Progress 10/01/23 to 09/30/24
Outputs
PROGRESS REPORT Objectives (from AD-416): Objective 1: Develop and validate an autonomous unmanned aerial vehicle with multimode imaging technologies for preharvest inspection of produce fields for animal intrusion and fecal contamination, and for irrigation water quality monitoring. Objective 2: Advance the development of customized compact spectral sensing technologies for food inspection and sanitation assessment in food processing, and for controlled-environment produce production, with embedded automated detection results for non-expert end users. Sub-objective 2.A: Develop a handheld line-scan hyperspectral imaging device with enhanced capabilities for contamination and sanitation inspection in food processing environments. Sub-objective 2.B: Develop a compact automated hyperspectral imaging platform for food safety and plant health monitoring for controlled- environment produce production in NASA space missions. Objective 3: Develop innovative spectroscopic and optical methods to characterize food composition and nondestructively detect adulterants and contaminants, for screening and inspecting agricultural commodities and commercially prepared food materials. Sub-objective 3.A: Develop a transportable multimodal optical sensing system for rapid, automated, and intelligent biological and chemical food safety inspection. Sub-objective 3.B: Develop a novel apparatus enabling dual-modality concomitant detection, along with associated methods and procedures, for assuring food integrity. Approach (from AD-416): The overall goal of this project is to develop and validate automated sensing tools and techniques to reduce food safety risks in food production and processing environments. Engineering-driven research will develop the next generation of rapid, intelligent, user-friendly sensing technologies for use in food production, processing, and other supply chain operations. Feedback from industrial and regulatory end users, and from stakeholders throughout the food supply chain, indicates that effective automated sensing and instrumentation systems require real-time data processing to provide non-expert users with a clear understanding and ability to make decisions based on the system output. Towards this end, we will develop unmanned aerial vehicles with multimodal remote sensing platforms and on-board data-processing capability to provide real- time detection and classification of animal intrusion and fecal contamination in farm fields and of irrigation water microbial quality. We will upgrade our existing handheld imaging device for contamination and sanitation inspection with multispectral imaging and embedded computing and artificial intelligence. We are also partnering with the NASA Kennedy Space Center to develop a novel, compact, automated hyperspectral platform for monitoring food safety and plant health of space crop production systems. Food safety and integrity requires identifying adulterants, foreign materials, and microbial contamination as well as authenticating ingredients. We will develop innovative multimodal optical sensing systems utilizing dual-band laser Raman, and Raman plus infrared, for simultaneous detection on a single sampling site. Spectroscopic and spectral imaging-based methodologies will be developed to enhance detection efficacy for liquid or powder samples. These systems will be supported with intuitive, intelligent sample-evaluation software and procedures for both biological and chemical contaminants. Significant progress has been made on all Objectives of the project, which fall under Food Safety National Program 108. For Objective 1, USDA-ARS scientists in Beltsville, Maryland, began design, testing, and development of a small unmanned autonomous vehicle (sUAV) mounted with a multimodal imaging system (including hyperspectral, thermal, and 3D/color cameras), and continued development and testing of the field transportable multimodal imaging system. Data were collected from both undisturbed fields and fields with evidence of animal intrusion using the sUAV and the ground truth imaging systems. During the testing, the performance of the ground-based platform was evaluated, and improvements were made in both ease of use and data collection quality. This testing also led to the start of a collaborative project with an industry partner to develop a multimodal imaging system for the sUAV that fulfills the needs of this research, as no commercially available systems meet the requirements of this project. The new sUAV system has now been flown for both functionality and early design testing as well as data collection. Work continues for gathering data and making design improvements, as well as evaluating the potential of emerging sensor technology for application in the new system. For Objective 2A, USDA-ARS scientists in Beltsville, Maryland, engaged in collaborative efforts to advance the CSI-D device. This device incorporates patented handheld fluorescence imaging technology and is specifically engineered for detecting contamination on food contact surfaces. It utilizes ultraviolet (UV)-A fluorescence imaging to identify contaminants, employs UV-C illumination for disinfection purposes, and facilitates documentation of cleanliness levels. Throughout rigorous experiments, the device demonstrated robust performance in automatically detecting contaminants, achieving notable accuracies aided by artificial intelligence models. Furthermore, the research team developed a sophisticated bench-top system capable of emitting UV-B and UV-C light. This system was meticulously designed and evaluated for effectiveness in eradicating harmful bacteria, particularly within food processing environments and potential applications in plant settings. The system will be transferred to USDA plant scientists for collaboration on exploring its use and effects in plant germicidal applications. For Objective 2B, USDA-ARS scientists in Beltsville, Maryland, collaborated with NASA Kennedy Space Center (KSC) to advance ARS hyperspectral imaging technology tailored for monitoring plant health and ensuring food safety in fresh produce production systems for future spaceflight. The ARS and KSC team developed a new multimodal imaging system, featuring an automated imaging gantry platform created by ARS scientists to acquire a range of multimodal spectral and phenotype data essential for monitoring plant health in controlled environments. This system was installed in a plant growth chamber at NASA KSC for full-scale plant imaging experiments and includes two LED line lights providing broad visible to near-infrared illumination for reflectance, along with 365 nm ultraviolet-A excitation for fluorescence imaging. Python 3.0- based software was developed to control movements of the 3-axis gantry system and acquisition of multimodal spectral image data. NASA KSC utilized this platform to gather imaging data on pick-and-eat salad crop seedlings cultivated under well-watered and stressed conditions. ARS scientists developed chemometric and machine learning models to analyze the imaging data collected at NASA KSC. A machine learning method employing an optimized discriminant classifier based on combination spectra with VNIR reflectance and fluorescence achieved classification accuracies exceeding 90% for drought stress treatments. This approach demonstrates significant potential for early detection of drought stress on lettuce leaves, preempting visible symptoms and size differences. In addition, the ARS team and other collaborators developed a customized light source aimed at improving data quality. In collaboration with the University of Florida (UF), USDA-ARS scientists in Beltsville, Maryland, continued work to develop citrus disease detection and classification methods. Using a portable hyperspectral imaging system recently developed by the ARS team, UF collaborators collected hyperspectral reflectance and fluorescence images from the front and back sides of both health and diseased citrus leaves at the Citrus Research and Education Center in Lake Alfred, Florida. Based on the hyperspectral reflectance image data, UF collaborators developed a leaf disease classification method based on hyperspectral band selection and YOLOv8 network architectures. In addition, USDA-ARS scientists in Beltsville, Maryland, developed a leaf disease classification method based on machine learning models using combined reflectance and fluorescence spectral data extracted from the hyperspectral images. For Objective 3A, in collaboration with the National Agricultural Products Quality Management Service (NAQS), South Korea, USDA-ARS scientists in Beltsville, Maryland, completed integration, calibration, and optimization for a multimodal optical sensing system for automated and intelligent biological and chemical assessment in food safety applications. New white and ultraviolet-A LED spot lights were added to the system to improve color and fluorescence imaging for bacterial colonies grown in agar Petri dishes, with an oblique angle of illumination to minimize reflected glare from various agars in the Petri dishes. The LabVIEW system software was also modified for lighting control, image saving, and synchronization functions. To prepare for full- scale bacterial experiments, we replaced the black foam boards previously used in the system⿿s aluminum-framed enclosure with black PVC panels for easier cleaning and sanitation. In addition to the bacteria study, the system has also been used by a visiting scientist from NAQS to conduct experiments for detection of mycotoxin contamination in grains and animal feeds. In collaboration with an industry partner, USDA-ARS scientists in Beltsville, Maryland, continued work to develop a portable multimode spectroscopy device for industrial applications, such as detecting fungi and mycotoxins. This device can measure four types of spectral data, including fluorescence from both 365 nm and 405 nm excitations, reflectance in the visible region, and reflectance in the near-infrared region. ARS scientists developed the system's interface software in Python 3.0 to calibrate and optimize system parameters, as well as to acquire and visualize multimodal spectral data. Machine learning and chemometric models were developed to classify contaminated corn using various spectra collected by this system, and were demonstrated to be capable of 91% accuracy in detecting mycotoxins in corn. In collaboration with the University of North Dakota (UND), USDA-ARS scientists in Beltsville, Maryland, continued work to develop fish authentication methods to address issues of species mislabeling and fraud as well as freshness of fish fillets. Based on a multimode hyperspectral image dataset (i.e., fluorescence with VNIR and SWIR reflectance) collected using ARS in-house developed imaging systems in Beltsville, Maryland, UND developed a method for detecting mislabeling of fish fillets that was based on multimode spectroscopy data fusion and machine learning techniques and which achieved 89% accuracy in classifying fish species. The industry partner will use the results to design and develop portable smart spectroscopy-based sensing devices for industrial applications for on-site fish species and freshness inspection. For Objective 3B, USDA-ARS scientists in Beltsville, Maryland, conducted dual-modality IR and Raman measurements of three commercial plasticizers to (1) identify spectral wavenumbers in which the IR and Raman signals for the same vibrational mode are the most different in normalized relative intensity; (2) use this dual modality data as a marker to determine the most sensitive signal ratio which is specific to BPA, BPS, and BPF; (3) assign these wavenumbers to vibrational modes characteristic to individual compounds. This expands continuing research demonstrating the application of dual-modality techniques in confirming identity of specific food safety related products. A point-scan IR and Raman dual spectral imaging system is being developed to overcome the lack of instrumentation designed specifically for macro-scale sample measurement. Automated sample positioning, spectral acquisition, and synchronization functions are realized using in-house developed control software. System capabilities for food safety applications will be demonstrated by experiments and results for authenticating selected foods using the fusion of the IR and Raman data. A rapid spectral detection technique was also developed to analyze in-situ (as is) wheat-like products labeled as gluten-free. Three chemical standards, gliadin, gluten, and starch from wheat, and 62 different types of commercial flour products were scanned by FT-IR spectroscopy over the wavenumber range of 4000 and 400. The linear discriminant analysis models were successfully used to evaluate the data. Artificial Intelligence (AI)/Machine Learning (ML) Artificial intelligence (AI) is an essential part of this engineering project, which involves analyzing and modeling large collections of hyperspectral and multispectral images acquired from various sensing systems for different food safety and agricultural applications. Both machine learning (ML) and deep learning (DL) techniques were used for this project during FY2024. State-of-the-art deep learning algorithms, including convolutional neural networks (CNNs) and generative adversarial networks (GANs), were used to classify images to inspect E. coli colonies using a contamination, sanitization inspection and disinfection (CSI-D) handheld fluorescence imaging device (Objective 2A). Also, DL models, including MobileNetv3 and DeepLabv3+, and federated learning (FL) were used to identify invisible residues on food preparation equipment and surfaces using the CSI-D imaging device and to develop data management models/methods for assuring client data privacy in the food service industry when utilizing collective imaging inspection technologies (Objective 2A). Multiple ML classification models, including discriminant analysis (DA), support vector machine (SVM), and partial least squares regression (PLSR), were used for detecting aflatoxins in ground maize based on Raman spectral data collected from a compact and automated measurement system (Objective 3A). We used Box and Google Drive to share data and results with our collaborators. In FY2024, we conducted our AI work using local Windows-based computing hardware. We recently purchased a high-performance computing (HPC) server with Linux OS, and plan to use this new HPC server to conduct fast ML and DL classifications for the large data sets. The AI based methods have opened a new avenue for analyzing data and improving measurement accuracies in this project. We developed and used ML and DL models to classify and predict unknown samples, especially when the sensing signals were low intensity or difficult to explain using the existing knowledge. ACCOMPLISHMENTS 01 Detection of aflatoxins in ground maize using Raman spectroscopy and machine learning. Aflatoxin contamination of maize is becoming an important issue in human food and animal feed supply. There is a need for an effective and efficient detection method that can be used for rapid onsite inspection of aflatoxin contamination. USDA-ARS scientists in Beltsville, Maryland, developed an aflatoxin detection method based on a custom developed compact and automated laser Raman spectroscopy system. Using a customized sample holder, Raman spectral data were automatically collected from ground maize samples naturally contaminated with aflatoxin. The data were analyzed using a machine learning method. A classification accuracy of 95.7% was achieved using a machine learning model based on linear discriminant analysis to differentiate aflatoxin levels in ground maize samples. The system and the detection method have the potential to be used at onsite processing locations to rapidly screen food and feed for aflatoxin and other hazardous substances affecting human and animal health. The technique would benefit the food industry and regulatory agencies in enforcing standards of safety and quality for maize-related food products. 02 Inspection of E. coli colonies using handheld fluorescence imaging and deep learning. Fruits and vegetables (e.g., citrus) can host bacterial pathogens (e.g., E. coli) that can cause severe health issues for the consumers. Detection of bacterial colonies on fruits and vegetables is important to reduce food safety risks and foodborne diseases. ARS scientists in Beltsville, Maryland, developed a method for inspecting E. coli colonies using a contamination, sanitization inspection and disinfection (CSI-D) handheld fluorescence imaging device, which was commercialized based on an ARS patented technology. Fluorescence images were collected from different concentrations of E. coli populations inoculated on black rubber slides. State-of-the-art deep learning algorithms (i.e., convolutional neural networks and generative adversarial networks) were used for image classifications. The best accuracy was achieved at 97% to classify four concentration levels of the E. coli colonies. The combination of the CSI-D handheld imaging and deep learning techniques would benefit the food industry and regulatory agencies in ensuring and enforcing food safety standards for products related to fruits and vegetables. 03 Identification of fish species using multi-mode spectroscopy and machine learning. Seafood mislabeling poses risks for consumers⿿ health and gives rise to economic and environmental hazards. Mixing less expensive species with more expensive species is a frequently recurring fraudulent practice in the seafood industry. ARS scientists in Beltsville, Maryland, developed a method based on multi-mode spectroscopy and machine learning techniques for detecting mislabeling of fish fillets. Three modes of spectral data, including fluorescence and reflectance in visible and near-infrared and short-wave infrared regions, were extracted from hyperspectral images collected from 216 fish fillet samples of 43 species. Algorithms were developed to create a hierarchical decision process for higher classification performance. Based on a classifier incorporating global and dispute models, a classification accuracy of 89% was achieved using the fusion of the three spectroscopic modes. The method developed in this study can be used to develop a rapid and cost-effective spectral sensing device for on-site inspection of the fish fillet mislabeling, which can be used for authentication of the fish fillets and other related food products by the seafood industry and regulatory agencies. 04 Classification of citrus fruit and leaf diseases using hyperspectral imaging and machine learning. Citrus black spot and canker are two significant diseases that pose quarantine threat, restrict market access, and cause economic losses for citrus growers. Early detection and management of groves infected with black spot or canker through fruit and leaf inspection can greatly benefit the citrus industry. ARS scientists in Beltsville, Maryland, developed an AI-based hyperspectral imaging and classification method for detection of fruits infected with black spot and leaves infected with canker. Hyperspectral reflectance images were collected in visible and near-infrared wavelength range from Valencia orange fruits and leaves with black spot, canker, and other common citrus diseases. Using convolutional neural network generated features and machine learning classifiers, classification accuracies were achieved at over 92% for classifying fruits with black spot and four other conditions and at over 93% for classifying leaves with canker and four other conditions. The method would benefit citrus industry and regulatory agencies in ensuring and enforcing the quality and safety standards for the citrus-related food and beverage products. 05 Non-destructive evaluation of soybean protein and lipid content using Raman chemical imaging. Soybean is an important crop that serves as a rich source of proteins and lipids, both of which are essential for human nutrition and are considered key parameters in determining soybean quality and market value. Traditional chemical analysis approaches (e.g., Soxhlet and Kjeldahl methods) for determining protein and lipid content of soybeans are destructive, time-consuming, and labor-intensive. ARS scientists in Beltsville, Maryland, developed a rapid and non-destructive method to evaluate soybean protein and lipid content based on macro-scale Raman hyperspectral imaging technique. A line-scan Raman hyperspectral imaging system was used to collect images of whole, intact soybean seeds. Partial least squares regression method was used to develop prediction models to correlate the Raman spectral data with the protein and lipid content of the soybean seeds. Chemical images were created to show the distributions and amounts of protein and lipid on single soybean seeds. The method developed in this study can be used for accurate and efficient estimation of protein and lipid content, which would benefit the soybean industry for quality control and breeding programs. 06 A rapid, accurate, and sensitive dual-modality IR and Raman spectroscopic technique for confirming the identity of plasticizers used in food containers. Bisphenol A (BPA) has previously been used as a common plasticizer in polycarbonate plastics for containers that store food and beverages. Regulatory measures have restricted the use of BPA in plastic formulations, especially for those which come in contact with food products. Rapid, accurate spectroscopic measurements are required for uniquely distinguishing BPA from other bis-phenols. ARS scientists in Beltsville, Maryland, developed a rapid, accurate, and sensitive dual-modality IR and Raman spectroscopic technique to distinguish BPA from other commercially used bis-phenols (BPS and BPF). Dual modality analysis enables addressing exactly the wavenumbers in which the IR and Raman spectra are the most different. Analysis found that the major phenolic ring vibrational mode intensities in BPA, BPS and BPF are significantly different, sufficiently so to identify them from each other despite their structural similarity. This approach provides practical information for the use of dual-modality technology for food detection, which will benefit researchers who have interest in developing and using the dual-modality techniques for safety and quality inspection of food products. 07 Rapid spectral identification of wheat-like products labeled gluten- free. Food allergies are major food and health concern. Celiac disease is a food allergy condition related to wheat protein gluten. Currently, the most commonly used methods for gluten testing of foods require complicated chemical procedures. ARS scientists in Beltsville, Maryland, developed a Fourier-transform infrared (FT-IR) in-situ (as is) spectroscopic technique for authentication of gluten-free flour products. Three chemical standards⿿gliadin, gluten, and starch from wheat⿿and 62 different types of flour products were scanned by FT-IR spectroscopy over the wavenumber range of 4000 and 400. Notable signal differences were observed between the chemical standards and wheat samples over the wavenumber range of 1800 to 450. The linear discriminant analysis models were successfully used to evaluate the data. This approach provides a practical method for evaluating commercial products labeled gluten-free for the presence of wheat in- situ (as is).
Impacts
(N/A)
Publications
- Sanders, J., Alarcorn, V., Marquis, G., Tabb, A., Van Kessel, J.S., Sonnier, J.L., Haley, B.J., Baek, I., Qin, J., Kim, M.S., Vasefi, F., Sokolov, S., Hellberg, R. 2024. Disinfection of foodborne bacteria using the Contamination Sanitization Inspection and Disinfection (CSI-D) device bg. Food Microbiology. 10(9): Article e30490. https://doi.org/10.1016/j. heliyon.2024.e30490.
- Guo, Q., Peng, Y., Qin, J., Chao, K., Zhao, Z., Yin, T. 2023. Advance in detection technique of lean meat powder residues in meat using SERS: a review. Molecules. 28(22): Article e7504. https://doi.org/10.3390/ molecules28227504.
- Aline, U., Bhattacharya, T., Faqeerzada, M., Kim, M.S., Baek, I., Cho, B. 2023. Advancement of non-destructive spectral measurements for the quality of major tropical fruits and vegetables: a review. Frontiers in Plant Science. 14:1240361. https://doi.org/10.3389/fpls.2023.1240361.
- Hong, S., Morgan, B.J., Stocker, M.D., Smith, J.E., Kim, M.S., Cho, K., Pachepsky, Y.A. 2024. Estimating concentrations of Escherichia coli across a farm pond from the sUAS-based RGB imagery and water quality variables with machine learning techniques. Water Research. 260: Article e121861. https://doi.org/10.1016/j.watres.2024.121861.
- Guo, Q., Peng, Y., Chao, K., Qin, J., Chen, Y., Yin, T. 2023. A determination method for clenbuterol residue in pork based on optimal particle size gold colloid using SERS. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy. 302:1386-1425. https://doi.org/10. 1016/j.saa.2023.123097.
- Yadav, P., Burks, T.F., Dudhe, K., Frederick, Q., Qin, J., Kim, M.S., Ritenour, M.A. 2023. Classification of E. coli colony with generative adversarial networks, discrete wavelet transforms and VGG19. Veterinary Radiology and Ultrasound. 6(3):146-160.
- Aulia, R., Amanah, H., Lee, H., Kim, M.S., Baek, I., Qin, J., Cho, B. 2023. Proteins and lipids content estimation in soybeans using Raman hyperspectral imaging. Frontiers in Plant Science. 14. Article e1167139. https://doi.org/10.3389/fpls.2023.1167139.
- Gorji, H., Saeedi, M., Zadeh, H., Husairik, K., Mojtaba, S., Qin, J., Chan, D.E., Baaek, I., Kim, M.S., Akhbardeh, A., Mackinnon, N., Vasefi, F., Tavakolian, K. 2023. Federated learning for clients' data privacy assurance in food service industry. Applied Sciences. 13(16):9330. https:// doi.org/10.3390/app13169330.
- Sueker, M., Daghighi, A., Akhbardesh, A., Mackinnon, N., Bearman, G., Baek, I., Hwang, C., Qin, J., Tabb, A.M., Roungchun, J.B., Hellberg, R.S., Vasefi, F., Kim, M.S., Tavakolian, K., Kashani Zadeh, H. 2023. A novel machine learning framework based on a hierarchy of dispute models for the identification of fish species using multi-mode spectroscopy . Sensors. 23(22): Article e9062. https://doi.org/10.3390/s23229062.
- Kim, J., Kurniawan, H., Faqeerzada, M.A., Kim, G., Lee, H., Kim, M.S., Baek, I., Cho, B. 2023. Proximate content monitoring of black soldier fly larval (Hermetia illucens) dry matter for feed material using short-wave infrared hyperspectral imaging. Food Science of Animal Resources. 43(6) :1150-1169. https://doi.org/10.5851/kosfa.2023.e33.
- Chun, S., Song, D., Lee, K., Kim, M., Kim, M.S., Kyoung-Su, K., Mo, C. 2024. Deep learning algorithm development for early detection of Botrytis cinerea infected strawberry fruit using hyperspectral fluorescence imaging. Postharvest Biology and Technology. 214: Article e112918. https://doi.org/ 10.1016/j.postharvbio.2024.112918.
- Kim, M., Yu, W., Song, D., Chun, S., Kim, M.S., Lee, A., Kim, G., Mo, C. 2024. Prediction of soluble-solid content in citrus fruit using visible⿿near infrared hyperspectral imaging based on machine learning and effective-wavelength selection algorithm. Sensors. 24, 1512. https://doi. org/10.3390/s24051512.
- Kurniawana, H., Ariefa, M., Santosh, L., Kim, M.S., Baek, I., Cho, B. 2024. Dual imaging technique for a real-time inspection system of foreign object detection in fresh-cut vegetables. Current Research in Food Science. 9: article 100802. https://doi.org/10.1016/j.crfs.2024.100802.
- Patel, A., Park, E., Lee, H., Priya, G., Kim, H., Joshi, R., Arief, M.A., Kim, M.S., Baek, I., Cho, B. 2023. Deep learning-based plant organ segmentation and phenotyping of sorghum plants using LiDAR point cloud. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing. 16, 8492. https://doi.org/10.1109/JSTARS.2023.3312815.
- Liu, Z., Zhou, H., Huang, M., Zhu, Q., Qin, J., Kim, M.S. 2023. Packaged butter adulteration evaluation based on spatially offset Raman spectroscopy coupled with FastICA. Journal of Food Composition and Analysis. 117:105149. https://doi.org/10.1016/j.jfca.2023.105149.
- Prabhukhot, G., Yin, H., Eggelton, C., Kim, M.S., Patel, J.R. 2024. Impact of surface topography and shear stress on single and dual species biofilm formation by Escherichia coli O157:H7 and Listeria monocytogenes in presence of promotor bacteria. BIOFOULING. 201: e116240. https://doi.org/ 10.1016/j.lwt.2024.116240.
- Boyd, A., Luo, Y., Lunney, J.K., Kustas, B., Fukagawa, N.K., Mattoo, A.K., Crow, W.T., Pachepsky, Y.A., Kim, M.S., Lillehoj, H.S., Van Tassell, C.P., Zhang, H.Q., Blomberg, L., Dubey, J.P. 2023. Cross-cutting concepts to transform agricultural research. Frontiers in Sustainable Food Systems. 7. Article e1242665. https://doi.org/10.3389/fsufs.2023.1242665.
- Kim, Y., Baek, I., Lee, K., Kim, G., Kim, S., Kim, S., Chan, D.E., Herrman, T., Kim, N., Kim, M.S. 2023. Hyperspectral imaging techniques for rapid detection of single- and co-contaminant aflatoxins and fumonisins in ground maize. Toxins. 15(7):472. https://doi.org/10.3390/toxins15070472.
- Frederick, Q., Burks, T.F., Watson, A., Yadav, P., Qin, J., Kim, M.S., Ritenour, M.A. 2024. Selecting hyperspectral bands and extracting features with a custom shallow convolutional neural network to classify citrus peel defects. Smart Agricultural Technology. 6: Article e100365. https://doi. org/10.1016/j.atech.2023.100365.
- Kim, Y., Qin, J., Baek, I., Lee, K., Kim, S., Kim, S., Chan, D.E., Herrman, T.J., Kim, N., Kim, M.S. 2023. Detection of aflatoxins in ground maize using a compact and automated Raman spectroscopy with machine learning. Current Research in Food Science. 7: Article e100647. https://doi.org/10. 1016/j.crfs.2023.100647.
- Yadav, P., Burks, T.F., Qin, J., Kim, M.S., Frederick, Q., Dewdney, M.M., Ritenour, M.A. 2024. Automated classification of citrus disease on fruits and leaves using convolutional neural network (CNN) generated features from hyperspectral images and machine learning classifiers. Journal of Applied Remote Sensing (JARS). 18 (1): Article e014512. https://doi.org/10. 1117/1.JRS.18.014512.
|
Progress 10/01/22 to 09/30/23
Outputs
PROGRESS REPORT Objectives (from AD-416): Objective 1: Develop and validate an autonomous unmanned aerial vehicle with multimode imaging technologies for preharvest inspection of produce fields for animal intrusion and fecal contamination, and for irrigation water quality monitoring. Objective 2: Advance the development of customized compact spectral sensing technologies for food inspection and sanitation assessment in food processing, and for controlled-environment produce production, with embedded automated detection results for non-expert end users. Sub-objective 2.A: Develop a handheld line-scan hyperspectral imaging device with enhanced capabilities for contamination and sanitation inspection in food processing environments. Sub-objective 2.B: Develop a compact automated hyperspectral imaging platform for food safety and plant health monitoring for controlled- environment produce production in NASA space missions. Objective 3: Develop innovative spectroscopic and optical methods to characterize food composition and nondestructively detect adulterants and contaminants, for screening and inspecting agricultural commodities and commercially prepared food materials. Sub-objective 3.A: Develop a transportable multimodal optical sensing system for rapid, automated, and intelligent biological and chemical food safety inspection. Sub-objective 3.B: Develop a novel apparatus enabling dual-modality concomitant detection, along with associated methods and procedures, for assuring food integrity. Approach (from AD-416): The overall goal of this project is to develop and validate automated sensing tools and techniques to reduce food safety risks in food production and processing environments. Engineering-driven research will develop the next generation of rapid, intelligent, user-friendly sensing technologies for use in food production, processing, and other supply chain operations. Feedback from industrial and regulatory end users, and from stakeholders throughout the food supply chain, indicates that effective automated sensing and instrumentation systems require real-time data processing to provide non-expert users with a clear understanding and ability to make decisions based on the system output. Towards this end, we will develop unmanned aerial vehicles with multimodal remote sensing platforms and on-board data-processing capability to provide real- time detection and classification of animal intrusion and fecal contamination in farm fields and of irrigation water microbial quality. We will upgrade our existing handheld imaging device for contamination and sanitation inspection with multispectral imaging and embedded computing and artificial intelligence. We are also partnering with the NASA Kennedy Space Center to develop a novel, compact, automated hyperspectral platform for monitoring food safety and plant health of space crop production systems. Food safety and integrity requires identifying adulterants, foreign materials, and microbial contamination as well as authenticating ingredients. We will develop innovative multimodal optical sensing systems utilizing dual-band laser Raman, and Raman plus infrared, for simultaneous detection on a single sampling site. Spectroscopic and spectral imaging-based methodologies will be developed to enhance detection efficacy for liquid or powder samples. These systems will be supported with intuitive, intelligent sample-evaluation software and procedures for both biological and chemical contaminants. Significant progress has been made for all objectives of the project, which fall under National Program 108. For Objective 1, ARS scientists in Beltsville, Maryland, continued development, design, and construction of the field transportable multimodal imaging system. This system will be used to validate the results of the small Unmanned Autonomous Vehicle (sUAV) system and provide reference data during the processes of sensor calibration and post-collection data processing. The sensor suite (including hyperspectral, thermal, and 3D/color cameras) that will be present on both the field and UAV units was installed onto the cart, and testing and integration of control software was started. A full-scale first prototype was built and underwent initial testing and design revisions. A second iteration design has been produced and an evaluation of the new design is under way. Testing of settings for the sensors and the linear stage has been completed and the results will be integrated into the control software. We continue the design and completion of the power system that will provide the sensor suite, control laptop, and control mechanisms power during remote testing. For Objective 2A, ARS scientists in Beltsville, Maryland, continued working with multiple collaborators for testing and commercialization of ARS portable multispectral imaging technology for contamination and sanitization inspection. Based on an exclusive license for the ARS- patented (US patent no. US 8,310,544) handheld fluorescence imaging technology, a commercial prototype Contamination Sanitization Inspection and Disinfection (CSI-D) device was developed in 2021. The handheld CSI-D device provides visualization of contamination on food contact surfaces via ultraviolet-A (UVA) fluorescence imaging, disinfection via ultraviolet-C (UVC) illumination, and documentation of cleanliness. Experiments were conducted to determine the detection efficacy of the device for various vegetable and meat sample smears on food contact surfaces such as commercial-grade plastic cutting boards and stainless- steel. Furthermore, we completed the development of a fully automated bench-top UV illumination system to evaluate the effectiveness of ultraviolet-B (UVB) and UVC light for germicidal applications. The system comprises a 305-nm UVB LED, a fan for cooling, a sample holder that can be adjusted in height, a single-board computer equipped with a touchscreen monitor, and a safety trigger mechanism. This setup was used for assessing the intensity of UVB irradiance at different sample distances, with the aim of identifying optimal parameters for effectively eliminating foodborne bacteria through germicidal applications. Experiments were conducted to evaluate the effectiveness of UVC radiation in eradicating pathogenic bacteria that were cultivated in Petri dishes. For Objective 2B, in collaboration with the NASA Kennedy Space Center (KSC), ARS scientists in Beltsville, Maryland, continued to develop an advanced version of ARS hyperspectral imaging technology suitable for plant health and food safety monitoring in fresh produce production systems for future spaceflight. The ARS and KSC team designed the new multimodal imaging system and ARS scientists developed an automated imaging gantry platform to acquire a range of multimodal spectral and phenotype data to monitor the health of plants grown in a controlled environment. The system was installed in a plant growth chamber at NASA KSC for full-scale plant imaging experiments. The newly developed multimodal sensing platform consists of a 3-axis gantry system, two LED line lights that provide broadband visible to near-infrared illumination for reflectance, and 365 nm ultraviolet-A excitation for fluorescence imaging. The system interface software was developed on Python 3.0 to control the gantry motions and to acquire multimodal spectral image data. NASA KSC is using the multimodal sensing platform to acquire imaging data for pick-and-eat salad crop seedlings grown under control conditions (well-watered) and stress conditions (under- and over-watered). ARS scientists have been developing hyperspectral image processing and artificial intelligence / machine learning models for analyzing the imaging data collected at NASA KSC. ARS scientists in Beltsville, Maryland, developed a new portable hyperspectral imaging system for collaborators at the University of Florida (UF) conducting research on citrus diseases and bacteria. Because the transport of diseased citrus fruit and leaf samples from mid and south Florida is restricted, ARS scientists initiated a Material Transfer Agreement with UF and delivered a portable imaging system and control software so that UF collaborators can collect image data on-site from diseased citrus samples without transporting samples to UF. They have acquired hyperspectral reflectance and fluorescence images from diseased leaves and fruit peels with bacteria at two citrus research centers in Florida (Lake Alfred and Fort Pierce). They will develop image fusion algorithms and deep learning classification models using the collected data for disease and bacteria detection. For Objective 3A, in collaboration with the National Agricultural Products Quality Management Service (NAQS), South Korea, ARS scientists in Beltsville, Maryland, completed development of control software and algorithms for spectral and image preprocessing and machine learning for a newly developed multimodal optical sensing system for automated and intelligent biological and chemical assessment in food safety applications. We developed the system control software using LabVIEW. Visiting scientists from NAQS conducted experiments on identification of common foodborne bacteria and detection of aflatoxin contamination in ground maize. Using machine-learning AI models, classification accuracies were achieved at 98% to differentiate five common bacterial species grown on nonselective agar in Petri dishes and at 95% to classify aflatoxin contamination levels in naturally contaminated ground maize samples. ARS scientists filed a USPTO patent application for the methodology and the prototype system. In collaboration with a CRADA partner and the University of North Dakota (UND), we continued work to develop fish authentication methods based on multimode hyperspectral imaging techniques to address issues of species mislabeling and fraud as well as freshness of fish fillets. In this continuing study, all hyperspectral image data (including VNIR and SWIR reflectance, fluorescence, and Raman) were collected for raw samples from over 60 major fish species. UND developed a machine learning method using the fusion of three spectral data types (i.e., VNIR, SWIR reflectance, and fluorescence) and achieved 95% accuracy in classifying fish conditions spanning fresh to spoiled. The CRADA partner will use the results to design and develop portable smart spectroscopy-based sensing devices for industrial applications for on-site fish species and freshness inspection. For Objective 3B, ARS scientists in Beltsville, Maryland, conducted dual- modality IR and Raman measurements of the insecticide fipronil to determine which vibrational modes are more intense in IR relative to those more intense in the Raman. Although IR and Raman spectra are often deemed to be ⿿complementary in principle, experimental evidence on spectral data on individual specific compounds is sparse in practice. Experimental evidence comparing IR and Raman spectral information has also been studied for yellow turmeric powder mixed with white turmeric powder. Dual-modality IR and Raman measures of seven lipids (cis- polyunsaturated fatty acids) in situ at room temperature were performed. Since the food component is much more abundant than the adulterant, one spectrum confirming an adulterant in one modality can be diluted by a second spectrum at another site which detects the food only. Thus, single modality measurements can quickly default into measuring only variance in food spectral signature because most of the data collected is food. In contrast, dual-modality collection can often almost immediately confirm if one site contains an adulterant. A point-scan IR and Raman dual macro- scale imaging system is being developed to overcome the lack of instrumentation designed specifically for macro-scale sample measurement. The system uses a thermal infrared light source and a laser as separate sources for IR and Raman measurement. Artificial Intelligence (AI)/Machine Learning (ML) Artificial intelligence (AI) is an essential part of this engineering project, which involves analyzing and modeling large collections of hyperspectral and multispectral images acquired from various sensing systems for different food safety and agricultural applications. Both machine learning (ML) and deep learning (DL) techniques were used for this project during FY2023. ML classifiers from Classification Learner app in MATLAB, such as Naive Bayes, decision tree, ensemble, k-nearest neighbor, discriminant analysis, neural network, and support vector machine, were used for early detection of drought stress on Dragoon lettuce for space crop production (Objective 2B) and classification of common foodborne bacteria for potential regulatory applications (Objective 3A). Our collaborators from University of North Dakota (UND) developed a DL method based on convolutional neural network to evaluate cleaning and sanitation of various surfaces using multispectral fluorescence images acquired from a handheld imaging device (Objective 2A) . We and UND used Box and Google Drive to share data and results. In FY2023, we conducted our AI works using local Windows-based computing hardware. We submitted a purchase request for a high-performance computing (HPC) server with Linux OS. In FY2024, we plan to use this new HPC server to conduct fast ML and DL classifications for the large data sets. The AI-based methods opened a new avenue for analyzing the data and improving the measurement accuracies in this project. We developed and used ML and DL models to classify and predict unknown samples, especially when the sensing signals were low or difficult to explain using the existing knowledge. ACCOMPLISHMENTS 01 A hyperspectral plant health monitoring system for space crop production. Plant monitoring in growth chambers onboard the International Space Station is currently conducted by estimating growth rates based on photographic analysis of daily increments in leaf areas. This limited approach cannot detect plant stresses or nutrient deficiencies that usually occur days before the leaves manifest any visible changes. In a collaborative project between scientists at USDA ARS, Beltsville, Maryland, and at NASA, Kennedy Space Center (KSC), Florida, a compact and automated hyperspectral imaging system was developed and installed at the KSC to monitor plant health for space crop production under controlled environments. The prototype system can collect both hyperspectral reflectance and fluorescence images in the visible and near-infrared region within a single imaging cycle, which can provide rich spectral and spatial information for possible early detection of abiotic stresses and diseases for pick-and-eat salad crops. In a preliminary study on Dragoon lettuce, the system showed potential for detecting drought stress before visible symptoms and leaf size differences were evident, using a machine learning method with the spectral reflectance data of the lettuce. The method would benefit NASA⿿s space crop production and other fresh produce production in controlled-environment agriculture in enhancing quality and safety of fruits and vegetables by reducing crop losses due to stress and disease and enabling earlier interventions to mitigate problems. 02 A multimodal optical sensing system for automated and intelligent food safety inspection. Commercial integrated spectroscopy systems are usually bulky and not flexible for testing various food and agricultural products. There exists a lack of compact sensing devices and methods for quick and routine analysis of the chemical and biological content of food samples. As an extension of previous ARS- developed macro-scale Raman technologies, ARS scientists at Beltsville, Maryland, developed a new transportable multimodal (transmission, color, fluorescence, and Raman) optical sensing system with embedded artificial intelligence capabilities for automated and intelligent food safety inspection. By using machine vision and motion control techniques, the system can conduct automated Raman spectral acquisition for a variety of sample types presented in customized well plates or in Petri dishes (from food materials to bacterial colonies grown on media). Interesting targets within the sample materials can be identified and labeled using real-time image and spectral processing and machine learning functions integrated into the in-house developed software. The system shows promise for use by food safety regulatory agencies as an initial screening tool for quick species identification of common foodborne bacteria. Compact and easily transportable, the prototype is suitable for field and on-site food safety inspection in potential regulatory and industrial applications. 03 Dual-modality IR and Raman in situ spectral measurements distinctly identify seven individual, biologically essential, lipids at room temperature. Commercially available fish oil products are mixtures of lipids, all of which contain the same redundant cis-polyunsaturated fatty acids structure. Spectral verification of specific lipid identities is possible but requires distinguishing spectral wavenumbers specific to individual lipid compounds⿿fingerprint-like markers⿿from redundant nonselective wavenumbers. ARS scientists in Beltsville, Maryland, determined that, surprisingly, infrared (IR) wavenumbers unique to seven lipids are different from the wavenumber markers in the Raman domain for those same lipids. Using dual-modality measurements⿿both IR and Raman markers together⿿can facilitate and enhance the identification of lipids for verification purposes, for example, by readily distinguishing between cis-polyunsaturated fatty acids with an even number of double bonds from those with an odd number of double bonds, such as the omega-3 fatty acid DHA (with six double bonds) and the omega-3 fatty acid EPA (with five double bonds) that each exhibit a unique spectral signature. This approach enables verifying the identity of lipids supplements, such as the omega-3 fatty acids that are often added to poultry feed to enhance egg quality. 04 A rapid spectroscopic technique for detecting veterinary drug residues on meat surfaces. Current methods to detect veterinary drug residues require a strict sample collection and processing protocol, the use of very expensive analytical instrumentation, and a high technical level of expertise to both operate the equipment and to interpret the results. ARS scientists at Beltsville, Maryland, have developed macro-scale Raman imaging and spectroscopy technologies and methodologies for research addressing food integrity concerns arising from adulteration or contamination. A 785-nm point-scan Raman system was used in the development of a detection method for drug residues including salbutamol, clenbuterol, and ractopamine. The method detects the spectral fingerprint of the residue compounds after they are first absorbed onto gold nanoparticles. These spectroscopic methods, once developed, are less complicated to run and more user-friendly than conventional methods, and can produce practical analytical results in real-time, which is useful to those raising healthy animals for safe human consumption. 05 Rapid assessment of fish freshness for multiple supply-chain nodes using multi-mode spectroscopy and fusion-based artificial intelligence. Fresh fish is a highly perishable product with more than 20% wasted at retail level every year. One reason for such waste is that early fish decay is not easily detectable by human senses. There is a need for sensing techniques that allow for onsite inspection of fish freshness in a rapid, cost-effective, and nondestructive manner. ARS scientists in Beltsville, Maryland, developed a multimode spectroscopy method for rapid assessment of fish freshness. Three types of spectral data (i.e., visible and near- infrared reflectance, short -wave infrared reflectance, and fluorescence) were collected from fish fillets of four species (i.e., farmed Atlantic salmon, wild coho salmon, Chinook salmon, and sablefish) over time as fillet conditions progressed from fresh to spoiled. A machine learning method using the fusion of the three spectral data types achieved 95% accuracy in classifying fresh and spoiled fish samples. The data analysis and classification methods developed in this research can be used to assist development of an easy- to-use handheld device to estimate remaining shelf life of the fish fillets, which could enable dynamic sales management and major reductions in waste for the seafood industry. 06 Citrus disease detection using convolutional neural network generated features and Softmax classifier on hyperspectral image data. Citrus diseases and peel blemishes can limit marketability of citrus crops and in some cases lead to shipping restrictions into certain regions. Proper and timely identification and control of citrus diseases can assure fruit quality and safety, improve production, and minimize economic losses. ARS scientists in Beltsville, Maryland, developed an AI-based hyperspectral imaging and classification method for identification of various diseased peel conditions on citrus fruit. Hyperspectral reflectance images were collected in the visible and near- infrared wavelength range from Ruby Red grapefruits with normal peels and with common peel diseases and defects, including canker, greasy spot, insect damage, melanose, scab, and wind scar. A classification accuracy of over 98% was achieved using a deep learning algorithm based on convolution neural network with the hyperspectral reflectance image data. Using this method could benefit the citrus industry and regulatory agencies (e.g., FDA and USDA APHIS) in helping to ensure and enforce quality and safety standards for citrus-related food and beverage products. 07 Nondestructive evaluation of packaged butter for adulteration based on spatially offset Raman spectroscopy coupled with FastICA. Butter is a dairy product that is prone to mixing with cheaper vegetable fat (e.g., margarine) in economically motivated adulteration. Traditional optical sensing techniques can be used for adulteration detection for unpackaged butter products. However, nondestructively authenticating packaged foods is challenging due to complicated interactions between light and packaging materials. ARS scientists in Beltsville, Maryland, employed a spatially offset Raman line-scan imaging technique using a point laser to nondestructively detect adulterated butter within its intact packaging. Animal butter was mixed with margarine in different ratios. Raman image and spectral data were acquired from the butter- margarine mixtures covered by original packaging sheets and plastic film. Analysis models were developed and successfully used to predict the adulteration content of the butter-margarine samples covered with different packaging materials. The detection method is useful for through-package safety and quality inspection of food materials. Use of the technique would benefit the food industry and regulatory agencies (e.g., FDA and USDA FSIS) in ensuring and enforcing safety and quality standards for packaged food products. 08 Evaluating performance of SORS-based subsurface signal separation methods using statistical replication Monte Carlo simulation. Nondestructive evaluation of safety and quality for packaged foods is a challenging task due to difficulties in acquiring optical signals from food samples through packaging materials. Spatially offset Raman spectroscopy (SORS) is a promising depth-profiling technique to tackle this problem. However, there is a lack of studies to evaluate the signal separation methods for the SORS technique. ARS scientists in Beltsville, Maryland, presented a method based on the line scan SORS and statistical replication Monte Carlo simulation to evaluate effectiveness of retrieving Raman signals from subsurface food samples. The simulation results were verified by three packaged foods (i.e., sugar in plastic jar, bagged rice, and boxed butter), in which fast independent component analysis method can effectively separate Raman signals from surface layer of the packaging materials and subsurface layer of the foods. The evaluation method can assist in developing and optimizing SORS-based methods for through-package safety and quality inspection of the foods and ingredients. The technique would benefit the regulatory agencies (e.g., FDA and USDA FSIS) and the food industry in enforcing standards of the safety and quality of the packaged food products.
Impacts
(N/A)
Publications
- Gorji, H., Van Kessel, J.S., Haley, B.J., Husarik, K., Sonnier, J.L., Shahabi, S., Zadeh, H., Chan, D.E., Qin, J., Baek, I., Kim, M.S., Akhbardeh, A., Sohrabi, M., Kerge, B., Mckinnon, N., Vasefi, F., Tavakolian, K. 2022. Deep learning and multiwavelength fluorescence imaging for cleanliness assessment and disinfection in food services. Frontiers in Remote Sensing. https://doi.org/10.3389/fsens.2022.977770.
- Guo, O., Peng, Y., Chao, K. 2022. Raman enhancement effect of different silver nanoparticles on salbutamol. Heliyon. 8(6):e09576. https://doi.org/ 10.1016/j.heliyon.2022.e09576.
- Guo, Q., Peng, Y., Chao, K., Zhuang, Q., Chen, Y. 2022. Raman enhancement effects of gold nanoparticles with different particle sizes on clenbuterol and ractopamine. Vibrational Spectroscopy. 123:103444. https://doi.org/10. 1016/j.vibspec.2022.103444.
- Zadeh, H., Hardy, M., Sueker, M., Li, Y., Tzouchas, A., Mackinnon, N., Bearman, G., Haughey, S., Akhbardeh, A., Baek, I., Hwang, C., Qin, J., Tabb, A.M., Hellberg, R., Ismail, S., Reza, H., Vasefi, F., Kim, M.S., Tavakolian, K., Elliott, C.T. 2023. Rapid assessment of fish freshness for multiple supply-chain nodes using multi-mode spectroscopy and fusion-based artificial intelligence. Sensors. 23:5149. https://doi.org/10.3390/ s23115149.
- Tunny, S., Kurniawan, H., Amanah, H., Baek, I., Kim, M.S., Chan, D.E., Farqeerzada, M., Wakholi, C., Cho, B. 2023. Hyperspectral imaging techniques for detection of foreign materials from fresh-Cut vegetables. Postharvest Biology and Technology. 201:112373. https://doi.org/10.1016/j. postharvbio.2023.112373.
- Qin, J., Monje, O., Nugent, M.R., Finn, J.R., O'Rourke, A.E., Wilson, K.D., Fritsche, R.F., Baek, I., Chan, D.E., Kim, M.S. 2023. A hyperspectral plant health monitoring system for space crop production. Frontiers in Plant Science. 14:1133505. https://doi.org/10.3389/fpls.2023.1133505.
- Qin, J., Hong, J., Cho, H., Van Kessel, J.S., Baek, I., Chao, K., Kim, M.S. 2023. A multimodal optical sensing system for automated and intelligent food safety inspection. Journal of the ASABE. 66(4):839-849. https://doi. org/10.13031/ja.15526.
- Faquurzada, M.A., Park, E., Kim, T., Kim, M.S., Baek, I., Joshi, R., Kim, J., Cho, B. 2023. Fluorescence hyperspectral imaging for early diagnosis of heat-stressed ginseng plants. Applied Sciences. 13:31. https://doi.org/ 10.3390/app13010031.
- Baek, I., Mo, C., Eggleton, C., Gadsden, S.A., Cho, B., Lee, H., Kim, M.S., Qin, J. 2022. Determination of spectral resolutions for multispectral detection of apple bruises using visible/near-infrared hyperspectral reflectance imaging. Frontiers in Plant Science. 13:963591. https://doi. org/10.3389/fpls.2022.963591.
- Nabwire, S., Suh, H., Kim, M.S., Baek, I., Cho, B. 2021. Review: Application of artificial intelligence in phenomics. Sensors. 21, 4363. https://doi.org/10.3390/s21134363.
- Joshi, R., Joshi, R., Kim, G., Kim, M.S., Baek, I., Lee, H., Mo, C., Cho, B., Kim, G., Park, E. 2022. Non-destructive identification of fake eggs using fluorescence spectral analysis and hyperspectral imaging. Korean Journal of Agricultural Science. 49:495-510. https://doi.org/10.7744/kjoas. 20220043.
- Omia, E., Bae, H., Park, E., Kim, M.S., Baek, I., Kabenge, I., Cho, B. 2023. Remote sensing in field crop monitoring: A comprehensive review of sensor systems, data analyses and recent advances. Remote Sensing. 15:354. https://doi.org/10.3390/rs15020354.
- Liu, Z., Huang, M., Zhu, Q., Qin, J., Kim, M.S. 2023. Evaluating performance of SORS-based subsurface signal separation methods using statistical replication Monte Carlo simulation. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy. 293:122520. https://doi.org/10. 1016/j.saa.2023.122520.
- Sun, D., Wang, X., Huang, M., Zhu, Q., Qin, J. 2023. Optical parameters inversion of tissue using spatially resolved diffuse reflection imaging combined with LSTM-attention network. Optics Express. 31(6):10260-10272. https://doi.org/10.1364/OE.485235.
- Yadav, P., Burks, T.F., Frederick, Q., Qin, J., Kim, M.S., Ritenour, M.A. 2022. Citrus disease detection using convolution neural network generated features and softmax classifier on hyperspectral image data. Frontiers in Plant Science. 13:1043712. https://doi.org/10.3389/fpls.2022.1043712.
|
Progress 10/01/21 to 09/30/22
Outputs
PROGRESS REPORT Objectives (from AD-416): Objective 1: Develop and validate an autonomous unmanned aerial vehicle with multimode imaging technologies for preharvest inspection of produce fields for animal intrusion and fecal contamination, and for irrigation water quality monitoring. Objective 2: Advance the development of customized compact spectral sensing technologies for food inspection and sanitation assessment in food processing, and for controlled-environment produce production, with embedded automated detection results for non-expert end users. Sub-objective 2.A: Develop a handheld line-scan hyperspectral imaging device with enhanced capabilities for contamination and sanitation inspection in food processing environments. Sub-objective 2.B: Develop a compact automated hyperspectral imaging platform for food safety and plant health monitoring for controlled- environment produce production in NASA space missions. Objective 3: Develop innovative spectroscopic and optical methods to characterize food composition and nondestructively detect adulterants and contaminants, for screening and inspecting agricultural commodities and commercially prepared food materials. Sub-objective 3.A: Develop a transportable multimodal optical sensing system for rapid, automated, and intelligent biological and chemical food safety inspection. Sub-objective 3.B: Develop a novel apparatus enabling dual-modality concomitant detection, along with associated methods and procedures, for assuring food integrity. Approach (from AD-416): The overall goal of this project is to develop and validate automated sensing tools and techniques to reduce food safety risks in food production and processing environments. Engineering-driven research will develop the next generation of rapid, intelligent, user-friendly sensing technologies for use in food production, processing, and other supply chain operations. Feedback from industrial and regulatory end users, and from stakeholders throughout the food supply chain, indicates that effective automated sensing and instrumentation systems require real-time data processing to provide non-expert users with a clear understanding and ability to make decisions based on the system output. Towards this end, we will develop unmanned aerial vehicles with multimodal remote sensing platforms and on-board data-processing capability to provide real- time detection and classification of animal intrusion and fecal contamination in farm fields and of irrigation water microbial quality. We will upgrade our existing handheld imaging device for contamination and sanitation inspection with multispectral imaging and embedded computing and artificial intelligence. We are also partnering with the NASA Kennedy Space Center to develop a novel, compact, automated hyperspectral platform for monitoring food safety and plant health of space crop production systems. Food safety and integrity requires identifying adulterants, foreign materials, and microbial contamination as well as authenticating ingredients. We will develop innovative multimodal optical sensing systems utilizing dual-band laser Raman, and Raman plus infrared, for simultaneous detection on a single sampling site. Spectroscopic and spectral imaging-based methodologies will be developed to enhance detection efficacy for liquid or powder samples. These systems will be supported with intuitive, intelligent sample-evaluation software and procedures for both biological and chemical contaminants. Significant progress has been made for all Objectives of the project, which fall under National Program 108. For objective 1, ARS scientists in Beltsville, Maryland, started development and design of the field transportable multimodal imaging system. This system will be used to validate the results of small Unmanned Autonomous Vehicle (sUAV) system and provide reference data during the processes of sensor calibration and post-collection data processing. The sensor suite that will be used in both field cart and (sUAV) units was finalized, and scientists began testing on these sensors. A full 3-D model of the field transportable system was completed to verify fit and dimensions, and the parts from this design will be purchased. Testing was done to confirm the viability of certain design parameters such as weight and linear actuator load capacity. Parts of this testing were corroborated using calculations within 3-D CAD with the specific goal of making the cart usable for a wide range of end users from both accessibility and crop variety standpoints. This design will be used as the first prototype platform to take sample images and work to optimize the data quality and simplify the collection process. For Objective 2A, ARS scientists in Beltsville, Maryland, continued to work with multiple collaborators for testing and commercialization of ARS portable multispectral imaging technology for contamination and sanitization inspection. Based on an exclusive license for the ARS- patented (US patent no. US 8,310,544) handheld fluorescence imaging, a commercial prototype Contamination Sanitization Inspection and Disinfection (CSI-D) device was developed in 2021. The handheld CSI-D device provides visualization of contamination on food contact surface via ultraviolet-A (UVA) fluorescence imaging, disinfection via ultraviolet-C (UVC) illumination, and documentation of cleanliness. Experiments were conducted to determine detection efficacy of the CSI-D devices for various vegetable and meat sample smears on food contact surfaces such commercial grade cutting boards and stainless-steel plates. Furthermore, ARS scientists completed the development of fully automated bench-top UV illumination systems to evaluate effectiveness of ultraviolet-B (UVB) and UVC for germicidal applications. The first system consists of a 305-nm UVB LED module with a cooling fan, a height- adjustable sample holder, a single-board computer with a touchscreen monitor, and a safety trigger. This system was used to measure UVB irradiance at varying distances to determine parameters suitable for germicidal applications on foodborne bacteria. Experiments were conducted to determine the efficacies of the UVC radiation for killing pathogenic bacteria grown in Petri dishes. ARS scientists in Beltsville, Maryland, developed a fully automated UV illumination and germicidal system which includes 305 nm UVB, 275 nm UVC, and mixed UVB-UVC LED modules, a programmable linear stage, a depth and RGB camera for sample imaging, a data logger, a single-board computer with a touchscreen monitor, and a safety trigger. This system will be used to establish optimal UVB and UVC parameters along with efficacy, safety, and functionality, for nonchemical control of pathogens For Objective 2B, in collaboration with NASA Kennedy Space Center (KSC), ARS scientists in Beltsville, Maryland, continued to develop next- generation hyperspectral imaging technology suitable for plant health and food safety monitoring in fresh produce production systems for future spaceflight. ARS scientists finished development and testing of a compact hyperspectral system equipped with broadband and UVA light for reflectance and fluorescence measurements. The prototype system and its control software were transferred and installed in a plant growth chamber at KSC for experiments on pick-and-eat salad crops. Hyperspectral reflectance and fluorescence images were acquired from Dragoon lettuce, pak choi, mizuna, and radish grown by KSC scientists under normal and abiotic stress conditions (e.g., drought and overwatering). ARS scientists developed hyperspectral image processing and machine learning classification programs for data analysis. Results from the lettuce experiment showed that machine learning classification models have the potential for early detection of drought stress on lettuce leaves prior to visible symptoms and leaf size differences. ARS scientists visited the KSC to improve the current system and plan to develop a new XYZ gantry imaging system for automated scanning of multiple plant growth chambers. For Objective 3A, in collaboration with National Agricultural Products Quality Management Service, South Korea, ARS scientists in Beltsville, Maryland, developed a multimodal optical sensing system for automated and intelligent biological and chemical assessment in food safety applications. The system uses two pairs of point lasers and spectrometers, at 785 and 1064 nm, to conduct dual-band Raman sensing, which can be used for samples generating low- and high-fluorescence interference signals, respectively. Automated data acquisition was realized using a fast XY-moving stage for solid, powder, and liquid samples placed in well plates or randomly scattered in standard Petri dishes (e.g., bacterial colonies). The system capability was demonstrated by an example application for rapid identification of five common foodborne bacteria. Using a machine-learning model based on a linear support vector machine, over 98% classification accuracy was achieved using spectra automatically collected from bacterial colonies for the five species grown on nonselective agar in Petri dishes. A patent application for the methodology and prototype system was approved by ARS National Mechanical & Measurement Patent Committee and will be filed to USPTO for a formal application. In collaboration with a CRADA partner, ARS scientists in Beltsville, Maryland, continued work to develop fish authentication methods based on multimode hyperspectral imaging techniques to address issues of species mislabeling and fraud as well as freshness of fish fillets. In this continuing study, two ARS in-house developed line-scan hyperspectral systems were used to collect reflectance and fluorescence images from fish fillet samples of additional species and from selected fillets for freshness study. Imaging experiments and DNA tests have been completed for over 60 major fish species. A hyperspectral image database with DNA barcoding species labels was established and shared with the CRADA partner. Machine learning AI models and spectral and image fusion algorithms were developed to classify fish species and freshness. The results will be used to design and develop portable smart sensing devices for industrial applications for on-site fish species and freshness inspection. For Objective 3B, we report a simple procedure to obtain fipronil samples at concentrations from 0.5 ppm to 11 ppm and measure the infrared (IR) spectra of fipronil samples. A partial least squares regression model has been developed to estimate the concentration of fipronil. The 3D chemical structure of fipronil was described, vibrational modes were assigned to IR wavelengths and compared with Raman wavelengths in the same spectral range. Raman spectra and IR spectra differentially detect symmetrical and asymmetrical vibrational modes in the same molecule. The fipronil vibrational mode near 2249 cm-1 was found to be much more intense in Raman than in the IR spectrum. We also examined the remaining IR and spectral fingerprint of fipronil to determine which vibrational modes are bigger in IR relative to those more intense in the Raman. Although IR and Raman spectra are often deemed to be in principle �complementary,� experimental evidence on spectral data on any given specific particular compound in practice is sparse. We had previously published experimental evidence comparing IR and Raman spectral data for turmeric powder mixed with white turmeric powder. The word �complementary� means more information is present when two data sets are both present. Whether IR and Raman spectral information provide identical identification information is clearly undetermined because experimental data comparing the two data sets has so rarely been published. The question of whether the spectral difference between the two techniques can be an even more definitively precise fingerprint was addressed with fipronil. Instead of using a single mode fingerprint in either IR or Raman, we found through using a dual modality technique that only a small number of specific wavenumbers paired in IR and Raman together are fully sufficient to precisely identify fipronil structure. ACCOMPLISHMENTS 01 Handheld fluorescence imaging device for surface contamination detection and disinfection. Cleaning and sanitation are critical components of USDA and FDA Hazard Analysis Critical Control Point (HACCP) regulation and management systems for food safety. Currently contamination inspection is conducted by human inspectors via either visual examination or spot-check testing, which is a process that limits productivity and is prone to error. Based on an ARS patented technology, a commercial contamination, sanitization inspection and disinfection (CSI-D) handheld imaging device has recently been developed for preventing infection in food preparation and serving facilities. The CSI-D device provides an innovative solution encompassing visualization of contamination using UV fluorescence imaging, disinfection of contamination using UVC illumination, and documentation of cleanliness. In demonstration experiments, the device can achieve one hundred percent sterilization for three selected pathogens under ten seconds, including a fungus (Aspergillus fumigatus), a bacterium (Streptococcus pneumonia), and a virus (influenza A). The commercialized CSI-D device will help improve efficacies for USDA-FSIS and the food processing industry for HACCP contamination and sanitation inspections required in Food Safety and Modernization Act (FSMA). 02 Identification of fecal contamination on meat carcasses using handheld fluorescence imaging and deep learning techniques. Animal fecal matter and ingesta, which can host bacterial pathogens such as E. coli and Salmonella, are a potential contaminant sources for various meat products. Detection of fecal contamination on meat carcasses is important to reduce food safety risks from foodborne diseases for consumers. In this study, a contamination, sanitization inspection and disinfection (CSI-D) handheld imaging device, which was developed and commercialized based on an ARS patented technology, was used to collect fluorescence images from beef and sheep carcasses in three meat processing facilities in North Dakota. State-of-the-art deep learning algorithms were developed to segment and identify areas of the fecal residues in the fluorescence images. The results demonstrated that the clean and fecal contaminated carcasses can be differentiated with an approximate accuracy of ninety seven percent. The combination of the CSI-D handheld imaging and deep learning techniques would benefit the meat industry and regulatory agencies (e.g., USDA FSIS and FDA) in ensuring and enforcing food safety standards for meat and related products. 03 Species classification of fish fillets using simulated annealing-based hyperspectral data optimization. Many fish fillets are similar in appearance, which makes them a target for economically motivated fraud. Mixing less expensive species into more expensive species is a common fraudulent practice in the seafood industry. This study developed a data analysis methodology to support design of a future spectroscopy- based system for detecting mislabeling of fish fillets. Three types of spectra�fluorescence, visible and near-infrared reflectance, and short- wave infrared reflectance�were obtained from hyperspectral images of fish fillet samples for 25 common species. Algorithms were developed for wavelength selection, data fusion, and machine learning classification. Based on a multi-layer perceptron neural network classifier, a ninety-five percent classification accuracy was achieved using the fusion of the three spectral modes with seven wavelengths selected by a simulated annealing method. The data analysis methods developed in this study can facilitate development of a rapid and cost- effective spectral sensing device for on-site inspection of the fish fillet mislabeling, which can be used for authentication of the fish fillets and other related food products by the seafood industry and regulatory agencies. 04 Rapid detection of aflatoxins in ground maize using spectral imaging techniques. Food crops such as maize, peanuts, and tree nuts can be contaminated with aflatoxin, which is produced by certain fungi and is considered a carcinogen. Many countries have established maximum allowable limits for the presence of aflatoxin in foods and thus there is a great interest worldwide in developing rapid and nondestructive methods to screen high volumes of food products for aflaxtoxin contamination. This study investigated the development of classification models to use with four hyperspectral imaging methods�fluorescence by ultraviolet excitation, visible/near-infrared reflectance, short-wave infrared reflectance, and Raman�to detection naturally-occurring levels of aflatoxin contamination in samples of ground maize. The results demonstrated that effective classification models were possible with all four hyperspectral imaging methods, indicating great promise for the development of non-destructive imaging methods that can be used by regulatory agencies and processors for high volume sample screening of ground maize and other products that are vulnerable to contamination. 05 Identification of corn kernels infected with aflatoxin using line-scan hyperspectral Raman imaging. Corn is one of the most susceptible crops to fungal infection. Effective methods for detecting aflatoxigenic fungi on corn kernels are important to reduce the risk of aflatoxin contamination entering the food and feed chains. In this study, a new detection method based on high-throughput hyperspectral Raman imaging was developed to differentiate healthy and artificially inoculated corn kernels with aflatoxigenic and non-aflatoxigenic fungi. Raman spectral differences between the healthy and the contaminated corn samples on both endosperm and germ sides of the kernels were investigated. Three- class discriminant models were developed based on mean spectra extracted from the Raman images of each kernel, and the best classification accuracy was achieved at ninety percentusing the endosperm data. The proposed method provides new possibilities to inspect for aflatoxin contamination on the corn kernels. The technique would benefit the food industry in helping to ensure the safety and quality of the corn and related food and feed products and benefit regulatory agencies with an interest in enforcing standards of food safety and quality for the corn products. 06 Nondestructive detection of adulterated sugar through plastic packaging using spatially offset Raman imaging. Safety and quality inspection of packaged foods and ingredients is important and challenging for both the food industry and regulatory agencies. Nondestructive detection of food adulterants through packaging using optical sensing techniques is difficult due to complex interactions between light and the packaging materials. This study developed a novel method using a laser-based spatially offset Raman imaging technique for detection of adulterated sugar in plastic packaging. Raman image and spectral data were collected from adulterated sugar samples that were made by mixing soft sugar and cheap glucose as an adulterant in different ratios. A mathematic prediction model was developed and was successfully used to evaluate the adulteration ratios for the mixed sugar samples through packaging. The results proved that the proposed method can be used for through-packaging inspection of the foods and ingredients for safety and quality applications. The technique would benefit the food industry in ensuring the safety and quality of packaged food products and benefit regulatory agencies with an interest in enforcing standards of food safety and quality for packaged foods and ingredients. 07 A rapid dual modality method for detecting insecticide fipronil on solid surfaces. Fipronil is a broad-spectrum insecticide banned from use in the food supply in the U.S. and European Union. An incident of poultry eggs tainted with fipronil in 2017 caused a recall of millions of eggs affecting more than 40 countries. ARS scientists in Beltsville, Maryland, have developed an in situ spectroscopic process for assaying fipronil on surfaces, verifying its identity, and validated the methodology developed. For fipronil in a [500 cm-1] in-common wavenumber range, two maximum intensity peaks in IR spectra were easily differentiated along with two different maximum intensity peaks in Raman spectra. The differences in the wavenumber intensities between IR results and Raman results in this selected spectral range demonstrates the complementary nature of IR and Raman: each contains critical spectral information absent in the other mode. The practical analytical advantage of using dual modality detection was demonstrated using fipronil. The same methodology can be applied to other compounds either for product verification or for identifying specific contaminants/adulterants. The in situ dual modality measurements enable more reliable and more accurate food product testing and monitoring that can be useful in food processing operations.
Impacts
(N/A)
Publications
- Delwiche, S.R., Baek, I., Kim, M.S. 2021. Does spatial region of interest (ROI) matter in multispectral and hyperspectral imaging of segmented wheat kernels. Biosystems Engineering. 212:106-114.
- Stocker, M., Pachepsky, Y.A., Hill, R.L., Kim, M.S. 2022. Elucidating spatial patterns of E. coli in two irrigation ponds with empirical orthogonal functions. Journal of Hydrology. 609:127770. https://doi.org/10. 1016/j.jhydrol.2022.127770.
- Ahmed, M.R., Yasmin, J., Park, E., Kim, G., Kim, M.S., Wakholi, C., Mo, C., Cho, B. 2020. Classification of watermelon seeds using morphological patterns of X-ray imaging: A comparison of conventional machine learning and deep learning. Sensors. 23(20), 6753. https://doi.org/doi:10.3390/ s20236753.
- Broadhurst, C., Schmidt, W.F., Qin, J., Chao, K., Kim, M.S. 2021. Continuous gradient temperature Raman spectroscopy of 1-stearoyl-2- docosahexonyl, 1-stearoyl- 2-arachidonoyl, and 1,2-steroyl phosphocholines. Chemistry and Physics of Lipids. https://doi.org/10.1016/j.chemphyslip. 2021.105116.
- Liu, Z., Huang, M., Zhu, Q., Qin, J., Kim, M.S. 2021. Detection of adulterated sugar with plastic packaging based on spatially offset Raman imaging. Journal of the Science of Food and Agriculture. 101:6281-6288. https://doi.org/10.1002/jsfa.11297.
- Kim, G., Lee, H., Baek, I., Cho, B., Kim, M.S. 2021. Quantitative detection of benzoyl peroxide in wheat flour using line-scan short-wave infrared hyperspectral imaging. Sensors and Actuators B: Chemical. 352:130997. https://doi.org/10.1016/j.snb.2021.130997.
- Kim, G., Lee, H., Cho, B., Baek, I., Kim, M.S. 2021. Quantitative evaluation of food-waste components in organic fertilizer using visible�near-infrared hyperspectral imaging. Applied Sciences. 11(17):8201. https://doi.org/10.3390/app11178201.
- Joshi, R., Sathasivam, R., Park, S., Lee, H., Kim, M.S., Baek, I., Cho, B. 2021. Application of Fourier transform infrared spectroscopy and multivariate analysis methods for the non-destructive evaluation of phenolics compounds in moringa powder. Agriculture. https://doi.org/10. 3390/agriculture12010010.
- Liu, Z., Huang, M., Zhu, Q., Qin, J., Kim, M.S. 2022. A packaged food internal Raman signal separation method based on spatially offset Raman spectroscopy combined with FastICA. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy. 275:121154. https://doi.org/10.1016/j.saa. 2022.121154.
- Nabwire, S., Wakholi, C., Faquurzada, M., Arief, M., Kim, M.S., Baek, I., Cho, B. 2022. Estimation of cold stress, plant age, and number of leaves in watermelon plants using image analysis. Frontiers in Plant Science. 13:847225. https://doi.org/10.3389/fpls.2022.847225.
- Tao, F., Yao, H., Hruska, Z., Rajasekaran, K., Qin, J., Kim, M.S. 2021. Use of line-scan Raman hyperspectral imaging to identify corn kernels infected with Aspergillus flavus. Journal of Cereal Science. 102:103364. https://doi.org/10.1016/j.jcs.2021.103364.
- Stocker, M., Pachepsky, Y.A., Smith, J., Morgan, B.J., Hill, R., Kim, M.S. 2021. Persistent patterns of E. coli concentrations in two irrigation ponds from three years of monitoring. Water, Air, and Soil Pollution. https://doi.org/10.1007/s11270-021-05438-z.
- Park, E., Kim, Y., Omari, M., Suh, H., Faqeezada, M., Kim, M.S., Baek, I., Cho, B. 2021. High-throughput phenotyping approach for the evaluation of heat stress in Korean ginseng (Panax ginseng Meyer) using hyperspectral reference image. Sensors. 21:5634. https://doi.org/10.3390/s21165634.
- Tunny, S., Amanah, H., Faqeerzada, M., Wakholi, C., Baek, I., Kim, M.S., Cho, B. 2022. Multispectral wavebands selection for the detection of potential foreign materials in fresh-cut vegetables. Sensors. 22:1775. https://doi.org/10.3390/s22051775.
- Schmidt, W.F., Chen, F., Broadhurst, C.L., Qin, J., Crawford, M.A., Kim, M. S. 2022. Unique and redundant spectral fingerprints of docosahexaenoic, alpha-linolenic and gamma-linolenic acids in binary mixtures. Journal of Molecular Liquids. 358:119222. https://doi.org/10.1016/j.molliq.2022. 119222.
- Wakholi, C., Nabwire, S., Kim, J., Bae, J.H., Baek, I., Kim, M.S., Cho, B. 2021. Economic analysis of an image-based beef carcass yield estimation system in Korea. Animals. 12:7. https://doi.org/10.3390/ani12010007.
- Amanah, H., Tunny, S.S., Masithoh, R., Choung, M., Kim, K., Kim, M.S., Baek, I., Lee, W., Cho, B. 2022. Nondestructive prediction of isoflavones and oligosaccharides in intact soybean seed using Fourier transform near- infrared (FT-NIR) and Fourier transform infrared (FT-IR) spectroscopic techniques. Foods. https://doi.org/10.3390/foods11020232.
- Nam, S., Baek, I., Hillyer, M.B., He, Z., Barnaby, J.Y., Condon, B.D., Kim, M.S. 2022. Thermosensitive textiles by silver nanoparticle-filled brown cotton fibers. Nanoscale Advances. 4:3725-3736. https://doi.org/10.1039/ D2NA00279E.
- Joshi, R., Baek, I., Joshi, R., Kim, M.S., Cho, B. 2022. Detection of fabricated eggs using Fourier Transform Infrared (FT- IR) spectroscopy coupled with multivariate classification techniques. Food Analytical Methods. https://doi.org/10.1016/j.infrared.2022.104163.
- Chao, K., Schmidt, W.F., Qin, J., Kim, M.S. 2022. A rapid and precise spectroscopic method for detecting fipronil insecticide on solid surfaces. Journal of Food Measurement and Characterization. https://doi.org/10.1007/ s11694-022-01384-4.
- Gorji, H.T., Shahabi, S.M., Sharma, A., Tamde, L.Q., Husarik, K., Qin, J., Chan, D.E., Baek, I., Kim, M.S., Mackinnon, N., Morro, J., Sokolov, S., Akhbardeh, A., Vasefi, F., Tavakolian, K. 2022. Combining deep learning and fluorescence imaging to automatically identify fecal contamination on meat carcasses. Nature Scientific Reports. 12:2392. https://doi.org/10. 1038/s41598-022-06379-1.
- Joshi, R., Sathasivam, R., Kumar, P., Patel, A.K., Nguyen, B.V., Faqeerzaada, M.A., Park, S., Lee, S., Kim, M.S., Baek, I., Cho, B. 2022. Comparative determination of phenolic compounds in Ara-bidopsis Thaliana leaf powder under distinct stress conditions using Fourier-Transform Infrared (FT-IR) and Near-Infrared (FT-NIR) Spectroscopy. Plants. 11(7) :836. https://doi.org/10.3390/plants11070836.
- Kim, Y., Baek, I., Lee, K., Qin, J., Kim, G., Shin, B.K., Chan, D.E., Herman, T.J., Cho, S., Kim, M.S. 2021. Investigation of reflectance, fluorescence, and Raman hyperspectral imaging techniques for rapid detection of aflatoxins in ground maize. Food Control. 132:108479. https:// doi.org/10.1016/j.foodcont.2021.108479.
- Chauvin, J., Duran, R., Tavakolian, K., Akhbardeh, A., Mackinnon, N., Qin, J., Chan, D.E., Hwang, C., Baek, I., Kim, M.S., Isaacs, R., Yilmaz, A., Roungchun, J., Hellberg, R., Vasefi, F. 2021. Simulated annealing-based hyperspectral data optimization for fish species classification: Can the number of measured wavelengths be reduced?. Food Control. 11:10628. https:/ /doi.org/10.3390/app112210628.
- Kumar, P., Faqeerzada, M., Park, E., Kim, Y., Joshi, R., Amanah, H., Sultana, T., Kim, H., Nabwire, S., Baek, I., Kim, M.S., Cho, B. 2022. Analysis of RGB plant images to identify root rot disease in Korean ginseng plants using deep learning. Applied Sciences. 12:2489. https://doi. org/10.3390/app12052489.
- Sueker, M., Stromsodt, K., Gorji, H.T., Vesafi, F., Khan, N., Schmidt, T., Varma, R., Mackinnon, N., Sokolov, S., Akhbardeh, A., Qin, J., Chan, D.E., Baek, I., Kim, M.S., Tavakolian, K. 2021. Handheld multispectral fluorescence imaging system to detect and disinfect surface contamination. Sensors. https://doi.org/10.3390/s21217222.
- Li, L., Peng, Y., Li, Y., Yang, C., Chao, K. 2021. Rapid and low-cost detection of moldy apple core based on an optical sensor system. Postharvest Biology and Technology. https://doi.org/10.1016/j.postharvbio. 2020.111276.
|
|