Source: UNIVERSITY OF MISSOURI submitted to NRP
PROCESS CONTROL AND COMPUTER VISION STRATEGIES FOR FOOD PROCESSES
Sponsoring Institution
State Agricultural Experiment Station
Project Status
COMPLETE
Funding Source
STATE
Reporting Frequency
Annual
Accession No.
0176157
Grant No.
(N/A)
Cumulative Award Amt.
(N/A)
Proposal No.
(N/A)
Multistate No.
(N/A)
Project Start Date
Oct 1, 1997
Project End Date
Sep 30, 2008
Grant Year
(N/A)
Program Code
[(N/A)]- (N/A)
Recipient Organization
UNIVERSITY OF MISSOURI
(N/A)
COLUMBIA,MO 65211
Performing Department
FOOD SCIENCE & ENGINEERING
Non Technical Summary
(N/A)
Animal Health Component
50%
Research Effort Categories
Basic
30%
Applied
50%
Developmental
20%
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
4043320202020%
4043320208020%
4045010202040%
4047410202020%
Knowledge Area
404 - Instrumentation and Control Systems;

Subject Of Investigation
3320 - Meat, beef cattle; 7410 - General technology; 5010 - Food;

Field Of Science
2020 - Engineering; 2080 - Mathematics and computer sciences;
Goals / Objectives
The overall goal of this project is to develop and apply advanced technologies for the measurement and control of food and biological processes. The research activities have the following two major objectives: 1. Develop on-line, long-range system identification and predictive control strategies for food processes; and 2. Develop computer vision techniques for food quality evaluation, pathological assessment, and appraisal of agricultural products.
Project Methods
The process control research will focus on new techniques of predictive control.The major activities will involve methodology development in the following areas: system identification for long-range prediction, generalized dual-target predictive control, and weighting normalization for dual-target control. Experimental validation will be based on an APV Baker 50-mm twin-screw food extruder. A dual-fluid mixing system has been constructed, which forms an MIMO process with variable time delays and dynamics. The computer vision work will include technique development and applications in meat quality evaluation, pathological assessment, agricultural commodity appraisal, and hardware development. Two Data Translation image processing systems, a new Matrox system and a new Windows-based image processing environment will be used. A near-infrared CCD camera with various bandpass filters will be added to expand the spectral ranges. A new SGI O2 workstation provides Khorus, 3-D reconstruction and other Unix-based image processing capabilities.

Progress 10/01/97 to 09/30/08

Outputs
OUTPUTS: The overall goal of this project is to develop and apply advanced technologies for the measurement and control of food and biological processes. Major progress was made in two primary areas: 1. Developing on-line, long-range system identification and predictive control strategies for food processes; and 2. Developing computer vision techniques for food quality evaluation, pathological assessment, and appraisal of agricultural products. Twenty-eight refereed journal papers and a similar number of conference papers resulted from the project (see Publications). In the first area, significant progress was made in developing methodologies for identifying and controlling food and biological processes. A new control strategy called Dual-Target Control was developed and applied in food extrusion control. Also with food extrusion as a model process, a method was developed to use sensory evaluations to determine process set points based on fuzzy set and neural network techniques. In the second area, much progress was made in developing and applying computer vision technologies in evaluating food (meat, grain, and processed food products) quality. New image processing and pattern recognition methods were developed and applied. Pathological specimen evaluation was also a significant part of the work. PARTICIPANTS: Partner organiation and collaborators: Dr. David Gerrard, Purdue University Dr. Gary Yao, Dr. Lei Bo, Dr. Randy Floyd, University of Missouri Other participants: Numerous graduate students, post doctoral fellows and visiting scientists evident from the publication list. TARGET AUDIENCES: Food industry PROJECT MODIFICATIONS: Not relevant to this project.

Impacts
The project resulted in some important technological advances in automating food and biological processes. These advances will not only reduce labor costs, which enhances the competitiveness or sustainability of agri-businesses, but also improves the objectivity and consistency of evaluations.

Publications

  • Sun, Y., X. Zhao and J. Tan, 1998. Intelligent Control Methods for Extruded Food Quality, Transactions of the CSAE, 14(1):183-187.
  • Tan, J., X. Gao and D.E. Gerrard, 1999. Application of Fuzzy Sets and Neural Networks in Sensory Analysis, J. Sensory Studies, 14(2):119-138.
  • Sun Y. and X. Zhao, J. Tan, 1999. Automatic Measurement of Apparent Quality of Extruded-Food, Transactions of the CSAM, 30(1):63-67.
  • Li, J., J. Tan, F. Martz and H. Heymann, 1999. Image Texture Features as Indicators of Beef Tenderness, J. Meat Science, 53:17-22.
  • Liu, X. and J. Tan, 1999. Acoustic Wave Analysis for Food Crispness Evaluation, J. Texture Studies, 30(4):397-408.
  • Gao, X, J. Tan, P. Shatadal and H. Heymann, 1999. Predicting Expanded-Food Sensory Properties by Image Analysis, J. Texture Studies, 30(3):291-304.
  • Hong, F., J. Tan and D.G. McCall, 2000. Application of Neural Network and Time Series Techniques in Wool Growth Modeling, Transactions of the ASAE, 43(1):139-144.
  • Wang, Y. and J. Tan, 2000. Dual-Target Predictive Control Development and Application in Food Extrusion Processes, Control Engineering Practice, 8:1055-1062.
  • Lu, J., J. Tan, P. Shatadal and D.E. Gerrard, 2000. Evaluation of Pork Color by Using Computer Vision, J. Meat Science, 56:563-566.
  • Li, J., J. Tan and P. Shatadal, 2001. Classification of Tough and Tender Beef by Image Texture Analysis, J. Meat Science, 57:341-346.
  • Shatadal, P and J. Tan, 2003. Identifying Damaged Soybeans by Color Image Analysis, Applied Eng. in Agriculture, 19(1):65-69.
  • Liu, X., J. Tan, I. Hatem and B.L. Smith, 2003. Image Processing To Facilitate Histological Evaluation of Tissue Specimens Stained with Perl's Prussian Blue, Toxicology Mechanisms & Methods, 13(3):213-220.
  • Hatem, I., J. Tan and D. Gerrard, 2003. Determination of Animal Skeletal Maturity by Image Processing, J. Meat Science, 65:999-1004.
  • Hatem, I. and J. Tan, 2003. Cartilage and Bone Segmentation in Vertebra Images, Transactions of the ASAE, 46(5):1429-1434.
  • Tan, J., 2004. Meat Quality Evaluation by Computer Vision, J. of Food Engineering, 61(1):27-35.
  • Lu, W. and J. Tan, 2004. Analysis of Image-Based Measurements and USDA Characteristics as Predictors of Beef Lean Yield, J. of Meat Science, 66:483-491.
  • Lu, W. and J. Tan, 2004. Grain Pattern Characterization and Classification in Walnut by Image Processing, Wood and Fiber Science, 36(1):311-318.
  • Kupongsak, S., J. Tan, I. Hatem, W. Lu, B. Guthrie and M. Tanoff, 2004. Set Point Determination from Sensory Evaluations for Food Process Control, J. of Food Process Engineering, 27(2):87-102.
  • Liu, X., J. Tan, I. Hatem and B.L. Smith, 2004. Image Processing of Hematoxylin and Eosin-Stained Tissues for Pathological Evaluation, Toxicology Mechanisms & Methods, 14:301-307.
  • Lu, W., J. Tan and R. Floyd, 2005. Automated Fetal Head Detection and Measurement in Ultrasound Images by an Iterative Randomized Hough Transform, Ultrasound in Medicine & Biology, 31(7):929-936.
  • Su, W.W., B. Liu, W.B. Lu, N.S. Xu, G.C. Du, and J. Tan, 2005. Observer-Based Online Compensation of Inner Filter Effect in Monitoring Fluorescence of GFP-Expressing Plant Cell Cultures. Biotechnology & Bioengineering, 91(2):213-226.
  • Kuponsak, S. and J. Tan, 2006. Application of Fuzzy Set And Neural Network Techniques in Determining Food Process Control Set Points, Fuzzy Set & Systems, 157:1169-1178.
  • Kuponsak, S. and J. Tan, 2006. Food Process Control Based on Sensory Evaluations, J. of Food Process Engineering. 29(4):675-688.
  • Lu, W, J. Tan, K. Zhang and B. Lei, 2008 Computerized Mouse Pupil Size Measurement for Pupillary Light Reflex Analysis, Computer Methods & Programs in Biomedicine, 90(3):202-209 (http://dx.doi.org/10.1016/j.cmpb.2008.01.002).
  • Chung, S. O., K. A. Sudduth and J. Tan, 2008, Spectral Analysis of On-the-go Soil Strength Sensor Data, J. of Biosystems Engineering, 33(5):355-361.
  • Tan, J., 2007. Inaugural Issue Editorial, Sensing & Instrumentation for Food Quality & Safety, 1(1):1-2.
  • Lu W. and J. Tan, 2008. Detection of Incomplete Curves in Images with Strong Noise by Iterative Randomized Hough Transform (IRHT), Pattern Recognition, 41:1268-1279.
  • Guo, Y., G. Yao, B. Lei and J. Tan, 2008. A Monte-Carlo Model for Studying the Effects of Melanin Concentrations on Retina Light Absorption, J. of Optical Society of America, 35:304-311.
  • Tan, J., 1998. Predictive Control Development for Food Manufacturing Processes, Proc. 1998 NSF Design & Manufacturing Grantees Conference, NSF, Arlington, VA.
  • Li, J., J. Tan, X. Gao and G.C. Smith, 1998. Image Texture Features as Indicators of Beef Muscle Mechanical Properties, 1998 ASAE Mid-Central Conference, Paper No. MC98130, ASAE, St. Joseph, MI.
  • Lu, J., J. Tan, X. Gao and G.E. Gerrard, 1998. USDA Beef Classification Based on Image Processing, 1998 ASAE Mid-Central Conference, Paper No. MC98131, ASAE, St. Joseph, MI.
  • Liu, X., J. Tan, and B.L. Smith, 1998. Image Processing of Prussian-blue-stained Tissue Slides for Pathological Evaluation, 1998 ASAE Mid-Central Conference, Paper No. MC98138, ASAE, St. Joseph, MI.
  • Tan, J. and A. Sethi, 1998. System Identification for Long-range Prediction, Proc. 4th Int=l. Sym. on Automatic Control of Food and Biological Processes, Goteborg, Sweden.
  • Tan, J. and Y. Wang, 1998. Dual-target Predictive Control and Application in Food Processes, Proc. 4th Int=l. Sym. on Automatic Control of Food and Biological Processes, Goteborg, Sweden, 2:382 398.
  • Tan, J. and Y. Sun, 1998. Quality Data Pattern Recognition for On-line Statistical Process Control, Proc. 4th Int=l. Sym. on Automatic Control of Food and Biological Processes, Goteborg, Sweden, 1:89-101.
  • Tan, J., 1998. Meat Quality Evaluation by Computer Vision, Proc. 4th Int=l. Sym. on Automatic Control of Food and Biological Processes, Goteborg, Sweden, 2:258-270.
  • Lu, W. and J. Tan, 1998. Wood Grain Pattern Characterization and Classification, 1998 ASAE Annual International Meeting, Paper No. 983045, ASAE, St. Joseph, MI.
  • Shatadal, P. and J. Tan, 1998. Identifying Damaged Soybeans by Using Color Image Analysis, 1998 ASAE Annual International Meeting, Paper No. 986023, ASAE, St. Joseph, MI.
  • Lu, J. and J. Tan, 1998. Application of Image Segmentation to Meat Image Processing, 1998 ASAE Annual International Meeting, Paper No. 983016, ASAE, St. Joseph, MI.
  • Hatem, I. and J. Tan, 1998. Determination of Animal Skeletal Maturity by Image Processing, 1998 ASAE Annual International Meeting, Paper No. 983019, ASAE, St. Joseph, MI.
  • Liu, X. and J. Tan, 1998. Microscopic Tissue Image Segmentation for Pathological Analysis, 1998 ASAE Annual International Meeting, Paper No. 983051, ASAE, St. Joseph, MI.
  • Tan, J. and F. Hong, 1999. Development and Application of Generalized Dual-target Predictive Control, Proc. of 14th World Congress of IFAC, Vol. N, pp. 169-174.
  • Lu, W. and J. Tan, 2002. Object Identification for Real-Time Machine Vision Applications, 2002 ASAE Annual International Meeting, Paper No. 023101, ASAE, St. Joseph, MI.
  • Lu, W-B., J. Tan and W. Su, 2002. Development of GFP-Based Sensing for Transgenic Plant Cell Cultures, 2002 ASAE Annual International Meeting, Paper No. 027024, ASAE, St. Joseph, MI.
  • Hatem, I., J. Tan and P. Shatadal, 1999. Beef Quality Prediction by Using Near-infrared Image Features, 1999 ASAE Annual International Meeting, Paper No. 993159, ASAE, St. Joseph, MI.
  • Cheng, Z., J. Tan and J. Kozak, 2003. Tissue Characterization by Ultrasound Speckle Features, 2003 ASAE Annual International Meeting, Paper No. 033052, ASAE, St. Joseph, MI.
  • Li, J., J. Tan and P. Shatadal, 1999. Discrimination of Beef Images by Texture Features, 1999 ASAE Annual International Meeting, Paper No. 993158, ASAE, St. Joseph, MI.
  • Lu, W. and J. Tan, 2000. Segmentation of Ultrasound Fetal Images, in Biological Quality and Precision Agriculture II, Proceedings of SPIE, 4203:81-90.
  • Kupongsak, S., J. Tan, B. Guthrie, H. Iyad and W. Lu, 2001. Set Point Determination From Sensory Data for Food Process Control, 2001 ASAE Annual International Meeting, Paper No. 016135, ASAE, St. Joseph, MI.
  • Hatem, I. and J. Tan, 2000. Cartilage Segmentation in Vertebra Images, 2000 ASAE Annual International Meeting, Paper No. 003125, ASAE, St. Joseph, MI.
  • Zhang, J., J. Tan, N. Xu and W. Su, 2001. Kalman Filter Implementation for Nonlinear Processes with General Time Delay Structure, 2001 ASAE Annual International Meeting, Paper No. 017011, ASAE, St. Joseph, MI.
  • Wirth, B. and J. Tan, 2002. Biophotonic Features as Indicators of Biological Material Properties, 2002 ASAE Annual International Meeting, Paper No. 023135, ASAE, St. Joseph, MI.
  • Hatem, I. and J. Tan, 2002. Beef Image Segmentation by Using Multivariate Classification, 2002 ASAE Annual International Meeting, Paper No. 023128, ASAE, St. Joseph, MI.
  • Kupongsak, S. and J. Tan, 2003. Food Process Control Based on Sensory Evaluations, 2003 ASAE Annual International Meeting, Paper No. 036166, ASAE, St. Joseph, MI.
  • Lu, W. and J. Tan, 2004. Automatic Fetal Head Measurements in Ultrasound Images Using an Iterative Randomized Hough Transform, Proc. of 2004 SPIE International Symposium on Medical Imaging, February 14-19, San Diego.
  • Liu, X. and J. Tan, 2004. Decoupling of Speckle and Signal in Ultrasound Images by Wavelet Transform, Proc. of 2004 CIGR International Conference, October 11-14, Beijing.
  • Lu, W. and J. Tan, 2005. Fetal Head Detection and Measurement in Ultrasound Images by a Direct Inverse Randomized Hough Transform, Proc. of 2005 SPIE International Symposium on Medical Imaging, February 12-17, San Diego.
  • Korampally, V., S. Bhattacharya, Y. Gao, SA. Grant, SB. Kleiboeker, K. Gangopadhyay, J. Tan, S. Gangopadhyay, 2006. Optimization of Fabrication Process for a PDMS-SOG-Silicon Based PCR Micro Chip through System Identification Techniques. Proc. of 19th IEEE Symposium on Computer-Based Medical Systems (CBMS'06), pp. 329-334, IEEE, New York.
  • Wirth, B. and J. Tan, 2005. Modeling of Delayed Fluorescence Kinetics for the Evaluation of Crop Stress, 2005 ASABE Annual International Meeting, Paper No. 05xxxx, ASABE, St. Joseph, MI.
  • Chung, S-O, K.A. Sudduth and J. Tan, 2005. Variability Structure of On-the-go Soil Strength Sensor Data, 2005 ASABE Annual International Meeting, Paper No. 051039, ASABE, St. Joseph, MI.
  • Guo, Y., B.J. Wirth and J. Tan, 2007. A Kinetic Model for Delayed Fluorescence from Plants, 2007 ASABE Annual International Meeting, Paper No. 077003, ASABE, St. Joseph, MI.
  • Guo, Y. and J. Tan, 2008. Enhancing Identifiability of Biological Systems by Recurrent-Pulse Excitation, 2008 ASABE Annual International Meeting, Paper No. 084481, ASABE, St. Joseph, MI.


Progress 01/01/05 to 12/31/05

Outputs
Delayed fluorescence was modeled based on an analysis of the photo-biochemistry of the photosynthesis. The model was first derived by analyzing the biochemical kinetics and the model parameters were determined from experimental data. The model could closely describe behavior of delayed luminescence from a number of different plants tested. The model could also clearly indicate stresses caused by drought and herbicides.

Impacts
The results are useful for measuring plants stresses, which is important for irrigation scheduling and herbicide applications. The model also provides insights into the physiological changes induced by environmental factors.

Publications

  • Lu, W., J. Tan and R. Floyd, 2005. Automated Fetal Head Detection and Measurement in Ultrasound Images by an Iterative Randomized Hough Transform, Ultrasound in Medicine & Biology, 31(7):929-936.
  • Kuponsak, S. and J. Tan, 2005. Application of Fuzzy Set And Neural Network Techniques in Determining Food Process Control Set Points, Fuzzy Set & Systems. (accepted)
  • Lu, Wei and J. Tan, 2005. Fetal Head Detection and Measurement in Ultrasound Images by a Direct Inverse Randomized Hough Transform, Proc. of 2005 SPIE International Symposium on Medical Imaging, February 12-17, San Diego.
  • Wirth, B. and J. Tan, 2005. Modeling of Delayed Fluorescence Kinetics for the Evaluation of Crop Stress, 2005 ASABE Annual International Meeting, ASABE, St. Joseph, MI.


Progress 01/01/04 to 12/31/04

Outputs
Specific activities this year: * Developed a new method for fetal head measurement in ultrasound images, * Developed techniques for delayed bioluminescence detection, * Initiated a project on visual system modeling.

Impacts
The work on imaging and classifcation methods will lead to new imaging technologies for applications in medicine, agriculture, food and other areas. The GFP-based sensing work will result in new sensing techniques for medical imaging, bioprocessing control and food pathogen detection. The biosystem modeling work will provide a better understanding of the feedback phenomena in biological systems.

Publications

  • Lu, W., J. Tan & L. Lu, 2004. Hough Transforms for Shape Identification and Application in Image Processing, Provisional Patent No. 04UMC035.
  • Lu, W. and J. Tan, 2004. Analysis of Image-Based Measurements and USDA Characteristics as Predictors of Beef Lean Yield, J. of Meat Science, 66:483-491.
  • Lu, W. and J. Tan, 2004. Grain Pattern Characterization and Classification in Walnut by Image Processing, Wood and Fiber Science, 36(1):311-318.
  • Tan, J., 2004. Meat Quality Evaluation by Computer Vision, J. of Food Engineering, 61(1):27-35.
  • Lu, Wei, J. Tan and R.C. Floyd, 2004. Fetal Head Detection and Measurement in Ultrasound Images by an Iterative Randomized Hough Transform, in Medical Imaging 2004: Image Processing, J.M. Fitzpatrick and M Sonka, eds., SPIE, Bellingham, WA.
  • Liu, X., J. Tan, I. Hatem and B.L. Smith, 2004. Image Processing of Hematoxylin and Eosin-Stained Tissues for Pathological Evaluation, Toxicology Mechanisms & Methods, 14:301-307.
  • Kupongsak, S. J. Tan, I. Hatem, W. Lu, B. Guthrie and M. Tanoff, 2004. Set Point Determination from Sensory Evaluations for Food Process Control, J. of Food Process Engineering, 27(2):87-102.
  • Liu, X. and J. Tan, 2004. Decoupling of Speckle and Signal in Ultrasound Images by Wavelet Transform, Proc. of 2004 CIGR International Conference, October 11-14, Beijing.
  • Lu, Wei and J. Tan, 2004. Automatic Fetal Head Measurements in Ultrasound Images Using an Iterative Randomized Hough Transform, Proc. of 2004 SPIE International Symposium on Medical Imaging, February 14-19, San Diego.


Progress 01/01/03 to 12/31/03

Outputs
Goals and Objectives: The current research program focuses on new measurement and control techniques for medical and biological applications. The specific objectives are to develop: * Imaging and classification techniques for biological tissue differentiation, * Green fluorescent protein-based biosensing techniques, and * Modeling and control methodology for biosystems. Specific activities this year: *Imaging and classification techniques for biological tissue differentiation: Effective techniques for biological tissue differentiation and property evaluation are important in medicine and bioresource utilization. Different imaging and sensing techniques are studied and multivariate classification methods are developed. The techniques being employed and developed include visible color imaging, near-infrared (NIR) imaging, ultrasound imaging, delayed luminescence, time-domain optical responses, and bioluminescence. *Green fluorescent protein-based biosensing: Expression of the green fluorescent protein (GFP) gene in a variety of cells and microbes makes GFP a useful mechanism for developing whole-cell biosensors for medical, bioprocessing, food safety and environmental applications. We are working on several key issues in GFP-based sensing for cell and protein density evaluation, which include excitation, fluorescence measurement, relationships between fluorescence and variables of interest, filtering, and reaction kinetics. The work involves both experiments and theoretical development. *Process modeling and control: Most biological and food processes are nonlinear systems with time delays. Effective process modeling and control strategies are needed. We are developing new methodologies for time delay system identification, predictive control, and statistical process control. This involves theoretical development, computer simulation, and experimental validation. Applications include food and cell culture processes.

Impacts
The work on imaging and classifcation methods will lead to new imaging technologies for applications in medicine, agriculture, food and other areas. The GFP-based sensing work will result in new sensing techniques for medical imaging, bioprocessing control and food pathogen detection. The biosystem modeling work will provide a better understanding of the feedback phenomena in biological systems.

Publications

  • Cheng, Z., J. Tan and J. Kozak, 2003. Tissue Characterization by Ultrasound Speckle Features, 2003 ASAE Annual International Meeting, Paper No. 033052, ASAE, St. Joseph, MI.
  • Kupongsak, S. and J. Tan, 2003. Food Process Control Based on Sensory Evaluations, 2003 ASAE Annual International Meeting, Paper No. 036166, ASAE, St. Joseph, MI.
  • Hatem, I and J. Tan, 2003. Image Analysis, in Encyclopedia of Agricultural, Food, and Biological Engineering, D.R. Heldman, ed., Marcel Dekker, New York.
  • Shatadal, P and J. Tan, 2003. Identifying Damaged Soybeans by Color Image Analysis, Applied Eng. in Agriculture, 19(1):65-69.
  • Liu, X., J. Tan, I. Hatem and B.L. Smith, 2003. Image Processing To Facilitate Histological Evaluation of Tissue Specimens Stained with Perl's Prussian Blue, Toxicology Mechanisms & Methods, 13(3):213-220.
  • Hatem, I., J. Tan and D. Gerrard, 2003. Determination of Animal Skeletal Maturity by Image Processing, J. Meat Science, 65:999-1004.
  • Hatem, I. and J. Tan, 2003. Cartilage and Bone Segmentation in Vertebra Images, Transactions of the ASAE, 46(5):1429-1434.
  • Kuponsak, S., J. Tan, F. Hsieh and H. Huff, 2003. Application of Fuzzy Set and Neural Network Techniques in Determining Food Process Set Points, 2003 IFT Annual Meeting and Food Expo, Chicago, IL, July 12-16.
  • Lu, W., J. Tan and B. Lei, 2003. Automated Size Measurement for Pupillary Light Reflex Analysis of Mice. BMES Annual Fall Meeting, Nashville, TN, October 1-4.


Progress 01/01/99 to 12/31/99

Outputs
In the area process control, a number of new control algorithms have been designed. They include dual-target predictive control, generalized dual-target predictive control,identification algorithms for long-term predictive control, timing scheme time-delay processes, and weighting normalization for optimal predictive control. These algorithms have been applied in food extrusion control. The computer vision work made progress on several fronts. A new multi-variate classification technique was derived. Image processing was found useful in predicting beef quality. Image texture features proved useful for beef mechanical property prediction. Image processing was successfully used to classify certain wood grain patterns. Several new primitive density-based algorithms were designed to segment microscopic images for pathological evaluation. Image features were computed from stained slides of infected livers, spleens and kidneys. The features predicted expert scores to a satisfactory degree of accuracy. Based on wavelet ransform, a useful filter was designed to improve the clarity of ultrasound images.

Impacts
The results of this research lead to new automation technology that improves the performance of several important food and other processes in terms of processing efficiency, consistency and product quality.

Publications

  • Tan, J., X. Gao and D.E. Gerrard, 1999. Application of Fuzzy Sets and Neural Networks in Sensory Analysis, J. Sensory Studies, 14(2):119-138.
  • Sun Y. and X. Zhao, J. Tan, 1999. Automatic Measurement of Apparent Quality of Extruded-Food, Transactions of the CSAM, 30(1):63-67.
  • Li, J., J. Tan, F. Martz and H. Heymann, 1999. Image Texture Features as Indicators of Beef Tenderness, J. Meat Science, 53:17-22.
  • Liu, X. and J. Tan, 1999. Acoustic Wave Analysis for Food Crispness Evaluation, J. Texture Studies, 30(4):397-408.
  • Hong, F., J. Tan and D.G. McCall, 1999. Application of Neural Network and Time Series Techniques in Wool Growth Modeling, Transactions of the ASAE. 42(6).
  • Tan, J., X. Gao, H. Heymann and P. Shatadal, 1999. Predicting Extrudate Textural Properties by Image Processing, J. Texture Studies, 30(3):291-304.
  • Tan, J. and F. Hong, 1999. Development and Application of Generalized Dual-target Predictive Control, Proc. of 14th World Congress of IFAC, Vol. N, pp. 169-174.
  • Hatem, I., J. Tan and P. Shatadal, 1999. Beef Quality Prediction by Using Near-infrared Image Features, 1999 ASAE Annual International Meeting, Paper No. 993159, ASAE, St. Joseph, MI.
  • Li, J., J. Tan and P. Shatadal, 1999. Discrimination of Beef Images by Texture Features, 1999 ASAE Annual International Meeting, Paper No. 993158, ASAE, St. Joseph, MI.


Progress 01/01/98 to 12/31/98

Outputs
In the area process control, progress was made in two major directions. First, a generalized dual-target predictive control algorithm was derived, which permits minimization of both input and output errors while limiting the size of input moves. The algorithm was tested through applications in food extrusion control. Second, a method for weighting normalization was derived for multi-variable optimal control and the method was tested on a lab-scale, multi-variable fluid system, which was constructed and instrumented during the year. The computer vision work made progress on several fronts. A new multi-variate classification technique was derived and implemented in Window-based image segmentation programs for meat image processing. Image processing was found useful in predicting beef maturity from cartilage features computed from vertebra images. Image texture features proved useful for beef mechanical property prediction. Image processing was successfully used to classify certain wood grain patterns. Several new primitive density-based algorithms were designed to segment microscopic images for pathological evaluation. Image features were computed from stained slides of infected livers, spleens and kidneys. The features predicted expert scores to a satisfactory degree of accuracy.

Impacts
(N/A)

Publications

  • Sun, Y., X. Zhao and J. Tan, 1998. Intelligent Control Methods for Extruded Food Quality, Transactions of the CSAE, 1491):183-187.
  • Tan, J., X. Gao and D.E. Gerrard, 1998. Application of Fuzzy Sets and Neural Networks in Sensory Analysis, J. Sensory Studies, 30(1)
  • Tan, J., 1998. Meat Quality Evaluation by Computer Vision, Proc. 4th Int'l. Sym. on Automatic Control of Food and Biological Processes, Goteborg, Sweden, 2:258-270.
  • Lu, W. and J. Tan, 1998. Wood Grain Pattern Characterization and Classification, 1998 ASAE Annual International Meeting, Paper No. 983045, ASAE, St. Joseph, MI.
  • Tan, J. and L. Wang, 1998. Vacuum Regulator, U.S. Patent No. 5,813,426.
  • Tan, J., 1998. Predictive Control Development for Food Manufacturing Processes, Proc. 1998 NSF Design & Manufacturing Grantees Conference, NSF, Arlington, VA.
  • Hatem, I., 1998. Determination of Beef Maturity by Using Image Processing, M.S. Thesis, University of Missouri-Columbia.
  • Li, J., J. Tan, X. Gao and G.C. Smith, 1998. Image Texture Features as Indicators of Beef Muscle Mechanical Properties, 1998 ASAE Mid-Central Conference, Paper No. MC98130, ASAE, St. Joseph, MI.
  • Lu, J., J. Tan, X. Gao and G.E. Gerrard, 1998. USDA Beef Classification Based on Image Processing, 1998 ASAE Mid-Central Conference, Paper No. MC98131, ASAE, St. Joseph, MI.
  • Liu, X., J. Tan, and B.L. Smith, 1998. Image Processing of Prussian-blue-stained Tissue Slides for Pathological Evaluation, 1998 ASAE Mid-Central Conference, Paper No. MC98138, ASAE, St. Joseph, MI.
  • Tan, J. and A. Sethi, 1998. System Identification for Long-range Prediction, Proc. 4th Int'l. Sym. on Automatic Control of Food and Biological Processes, Goteborg, Sweden.
  • Tan, J. and Y. Wang, 1998. Dual-target Predictive Control and Application in Food Processes, Proc. 4th Int'l. Sym. on Automatic Control of Food and Biological Processes, Goteborg, Sweden, 2:382-398.
  • Tan, J. and Y. Sun, 1998. Quality Data Patten Recognition for On-line Statistical Process Control, Proc. 4th Int'l. Sym. on Automatic Control of Food and Biological Processes, Goteborg, Sweden, 1:89-101.
  • Shatadal, P. and J. Tan, 1998. Identifying Damaged Soybeans by Using Color Image Analysis, 1998 ASAE Annual International Meeting, Paper No. 986023, ASAE, St. Joseph, MI.
  • Lu, J. and J. Tan, 1998. Application of Image Segmentation to Meat Image Processing, 1998 ASAE Annual International Meeting, Paper No. 983016, ASAE, St. Joseph, MI.
  • Hatem, I. and J. Tan, 1998. Determination of Animal Skeletal Maturity by Image Processing, 1998 ASAE Annual International Meeting, Paper No. 983019, ASAE, St. Joseph, MI.
  • Liu, X. and J. Tan, 1998. Microscopic Tissue Image Segmentation for Pathological Analysis, 1998 ASAE Annual International Meeting, Paper No. 983051, ASAE, St. Joseph, MI.


Progress 01/01/97 to 12/31/97

Outputs
This is a new project initiated in August, 1997. The overall goal of this project is to develop and apply advanced technologies for the measurement and control of food and biological processes. The research activities have the following two major objectives: (1) Develop on-line, long-range system identification and predictive control strategies for food processes; and (2) Develop computer vision techniques for food quality evaluation, pathological assessment, and appraisal of agricultural products. A timing scheme has been developed for predictive control of processes with arbitrary time delay structures. A nonlinear transform devised appeared useful for multi-variate classification and image segmentation. A DSP-based vision system was designed.

Impacts
(N/A)

Publications

  • Tan, J., 1998. Predictive Control Development for Food Manufacturing Processes, Proc. 1998 NSF Design & Manufacturing Grantees Conference,