Source: UNIVERSITY OF FLORIDA submitted to NRP
USING MULTI-SENSOR SPECTRAL ANALYSIS FOR NATURAL RESOURCE MANAGEMENT AND PRECISION AGRICULTURE APPLICATIONS
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
COMPLETE
Funding Source
Reporting Frequency
Annual
Accession No.
1002792
Grant No.
(N/A)
Cumulative Award Amt.
(N/A)
Proposal No.
(N/A)
Multistate No.
(N/A)
Project Start Date
Mar 26, 2014
Project End Date
Feb 28, 2019
Grant Year
(N/A)
Program Code
[(N/A)]- (N/A)
Recipient Organization
UNIVERSITY OF FLORIDA
G022 MCCARTY HALL
GAINESVILLE,FL 32611
Performing Department
Gulf Coast Research and Education Center
Non Technical Summary
With the rapid technological advances in remote sensing imaging technologies, an increasing need for methods to integrate such technologies and maximize their use in natural resource management and precision agriculture applications is heightened. Multi-sensor technologies can provide different sources of data to account for the tradeoffs of each data type and fill in the needs of specific type of applications. Spectral analysis of different land cover types has been the subject of a considerable amount of research in the last few decades. For example, hyperspectral imagery has been used in applications that involve studying plant stress, estimating water quality parameters of both open oceans and turbid inland waters, agricultural crop classification and yield estimation, and characterization of ecosystems.Variations in the amount of emitted, absorbed, and reflected electromagnetic energy arise from spectral characteristics of each plant species, plant condition and spatial distribution, sensor specifications, and atmospheric condition. Each of these factors contributes to a plant's spectra to form a unique pattern, or signature. On the other hand, the typical low spatial resolution of hyperspectral imagery and the need for image calibration techniques affect data quality and suitability for targeting specific applications that require precise object characterization. Some of these issues can be resolved by integrating different data types such as high spatial resolution imagery captured by lower spectral resolution sensors or Lidar point clouds. This type of data can provide textural and structural information due to their high spatial resolution and their potential for three-dimensional analysis.The need to establish the relationship between plant status and spectral measurements is of interest to many state and national agencies as well as private industry. Being able to utilize spectral data for the early identification of invasive plant species or optimizing the use of fertilization in agricultural lands could save the economy billions of dollars, not to mention the positive impact on human health and the environment. Fortunately, recent advances in spatial data acquisition technologies including multispectral and hyperspectral imagery and their acquisition platforms (e.g. introducing Unmanned Air Systems (UAS)) made the acquisition and use of such technologies more feasible and promising. However, in order to facilitate the use of such technologies to answer specific research questions, the research community needs to develop the methodologies required for data acquisition system integration, experimental design, and analysis algorithms.Improving remote sensing data acquisition tools, methodologies, and analysis algorithms for land cover characterization continues to be the main focus of this research effort. In this context, the proposed develop and use integrated multi-sensor spatial data acquisition techniques that include hyperspectral, multispectral, positioning and attitude sensors, etc. to analyze vegetation response. This research will help close the gap in understanding and quantifying spatial changes of, for example, vegetation, soils, and water quality. Furthermore, successful development could lead to integrated multi-sensor platform that provide a suite of data and enables applied and basic research opportunities in the precision agriculture and natural resource management fields.
Animal Health Component
85%
Research Effort Categories
Basic
0%
Applied
85%
Developmental
15%
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
4027299202015%
1020199206035%
2050330206030%
4047210208020%
Goals / Objectives
This proposal aims at integrating and developing spatial data sources, methodologies and algorithms to study the relationship between remote sensing data and object signatures. In this context, the proposed develop and use integrated multi-sensor spatial data acquisition techniques that include hyperspectral, multispectral, positioning and attitude sensors, etc. to analyze landcover response and classification. This research will help close the gap in understanding and quantifying spatial changes of, for example, vegetation, soils, and water quality. Furthermore, successful development could lead to integrated multi-sensor platform that provide a suite of data and enables applied and basic research opportunities in the precision agriculture and natural resource management fields. In achieving these goals, the following objectives are identified:I. Develop, calibrate and test spectral and navigation measuring systemsII. Design experiments and develop methods to collect and analyze spectral and field measurementsIII. Use spatial data acquisition sensors and techniques (including multispectral and hyperspectral imagery) to study land cover characteristics
Project Methods
The following is project methods organized by project objectives: I. Develop, calibrate and test spectral and navigation measuring systemsSystems that combine hyperspectral, multispectral and navigation sensors are subject to continuous improvement. Tighter integration of the system components for accurate georeferencing and adding more sensors is proposed. This process involves acquiring the sensors, developing new integration techniques, implementing system calibration and testing steps.System integration involves integrating different remote sensing and navigation data acquisition sensors. For example, across platform integration involves utilizing ground-based system in combination with sensors onboard UAS. Such integration overcome many of the limitations associated with each system leading to wide scale implementation and potential commercialization.II. Design experiments and develop methods to collect and analyze spectral and field measurementsThis objective aims at designing field experiments to collect spatial data using onboard sensors. The data should be augmented by establishing ground control locations and spectral calibration techniques. Field spectrometer measurements and multi-sensor imaging systems will be collected and analyzed. The design includes identifying management procedures, frequency of observations and accompanied reference lab measurements. It is envisioned that the results of ground based remote sensing leads to efficient and wider scale implementation using aerial systems (e.g. using UAS).III. Use spatial data acquisition sensors and techniques (including multispectral and hyperspectral imagery) to study land cover characteristicsDifferent types of remote sensing imagery will be analyzed. Image correction algorithms will be implemented. Hyperspectral image analysis techniques such as dimensionality reduction using Minimum Noise Fraction Transform algorithm (Chen, et al., 2003; Pande-chettri & Abd-Elrahman, 2013), statistical modeling, object based analysis will be implemented. The results will be analyzed and compared to ground truth data for modeling and assessment. It is envisioned that this objective will lead not only to better methods for land cover (e.g. vegetation, soils, water quality, etc.) characterization, but also to the introduction of more efficient techniques to achieve these results.Techniques and results achieved by applying remote sensing techniques at certain scale (e.g. ground based) can be a great asset for a different scale implementation. For example, determining specific bands capable of revealing specific water quality parameters can be the basis of larger scale water sampling by mounting a light weight sensor capable of sensing only the few bands relevant to specific water quality parameters on an UAS. This process involves developing appropriate sensors, calibration and analysis techniques in addition to extensive testing and assessment.

Progress 03/26/14 to 02/28/19

Outputs
Target Audience:The audience of this phase of the project is agricultural scientificcommunity interested in using ground-basedand and unmanner aeircraft systems in crop biophysical paramater and yield modeling. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?OPS technical personnel were hired and trained to analyze image data. A PhD student is working on analyzing ground-based images How have the results been disseminated to communities of interest?The technology and results used in strawberry data collection and analysis werepresented in the annual North American Strawberry Symposium . What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? Ground-based imaging system were analyzed to model strawberry dry mass weight and yield estimates. The images were analyzed using photogrammetric techniques to create multispectral mosaics. Several metrics have been tested, including canopy area, volume, and average height. Weekly regression models as well as whole season model were experimented with.

Publications


    Progress 10/01/17 to 09/30/18

    Outputs
    Target Audience:The audience of this phase of the project is the natural resource management and agricultural scientific community interested in using ground-basedand and unmanner aeircraft systems in land cover classification and crop biophysical paramater modeling. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?OPS technical personnel were hired and trained to analyze image data. Four undergraduate students participated on strawberry image interpretation. A PhD student is working on analyzing ground-based images How have the results been disseminated to communities of interest?The technology used in strawberry data collection was presented in the annual Strawberry Agritech meeting. A referred journal manuscript is published to the scientific community. What do you plan to do during the next reporting period to accomplish the goals?Analyze and write manuscript for the dry mass weight models developed using image-based metrics. Build capacity of multi-spectral image acquisitio using drone-mounted cameras. Develop straberyy growth and radiative transfer models

    Impacts
    What was accomplished under these goals? Ground-based imaging system were analyzed to model strawberry dry mass weight. The images were analyzed using photogrammetric techniques to create multispectral mosaics. Several metrics have been tested, including canopy size. Weekly regression models as well as whole season model were experimented with. We developed a multi-view object based image classification method for classifying wetland land cover. tested clases included the cogan grass invasive plant. The method proved superior results compared to traditional orthoimage based image clasification.

    Publications

    • Type: Journal Articles Status: Published Year Published: 2018 Citation: Liu, T. & Abd-Elrahman, A. (2018). Multi-view object-based classification of wetland land covers using unmanned aircraft system images. Remote Sensing of Environment, 216, 122-138.


    Progress 10/01/16 to 09/30/17

    Outputs
    Target Audience:This audience of this phase of the project is the natural resource management and agricultural scientific communityinterestedin using ground-basedand unmanner aeircraft systemsin land cover classification andcrop biophysical paramaters.modeling Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?OPS technical personnel were hired and trained to analyze image data. A PhD student is working on analyzing radiative transfer models for vegetation. How have the results been disseminated to communities of interest?The technology used in strawberry data collection was demonstrated in the field to strawberry growers andextension agents. A referred journal manuscript is published to the scientific community. What do you plan to do during the next reporting period to accomplish the goals?Ground-based and unmanned aerial system imagery of strawberry vegetation will be acquired and analyzed to model biophysical parameters and disease stress. Work on developing radiative transfer models will continue.

    Impacts
    What was accomplished under these goals? Ground-based imaging system was utilized to acquire multispectral imagery of strawberry fields. The images were analyzed usingphotogrammetric techniques to create multispectral mosaics. Data extracted from these mosaics are being used to model canopy biophysical parameters. A manuscript is published on the use of a multi-resolution object-based analysis for wetland cover classification of unmanned aircraft system imagery.

    Publications

    • Type: Journal Articles Status: Published Year Published: 2017 Citation: Pande-Chhetri, R., Abd-Elrahman, A., Liu, T., Morton, J., & Wilhelm, V. L. (2017). Object-based classification of wetland vegetation using very high-resolution unmanned air system imagery. European Journal of Remote Sensing, 50(1), 564-576.


    Progress 10/01/15 to 09/30/16

    Outputs
    Target Audience:This audience of this phase of the project is the scientific community, system integrators personnel, and growers interested in unmanned aerial systems and ground-based remote sensing systems. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?A MS student student was trained on unmanned aerial system image acquisition and analysis. A PhD student is working on analyzing radiative tranfer models for vegetation areas. How have the results been disseminated to communities of interest?The results were disseminated as MS thesis. Previous results were published as a refereed journal publication. What do you plan to do during the next reporting period to accomplish the goals?Ground-based and unmanned aerial system imagery of strawberry vegetation will be acquired and analyzed to model biophysical parameters and disease stress. Work on developing radiative transfer models will continue.

    Impacts
    What was accomplished under these goals? Unmannedaerial system was used to acquire multispectral imagery of strawberry fields. The images were analyzed using photogrammetric techniques to create multispectral mosaics. Data extracted from these mosaics were used to model spider mite count and damage.A MS thesis was prepared and defended based on this work and a refereed journal manuscript is in preparation.

    Publications

    • Type: Journal Articles Status: Published Year Published: 2016 Citation: Abd-Elrahman, A., Sassi, N., Wilkinson, B., & Dewitt, B. (2016). Georeferencing of mobile ground-based hyperspectral digital single-lens reflex imagery. Journal of Applied Remote Sensing, 10(1), 014002-014002.
    • Type: Theses/Dissertations Status: Published Year Published: 2016 Citation: Cakir, M. 2015. Damage Detection of Twospotted Spider Mites (Tetranychus Urticae Koch) in Strawberry Using Unmanned Aerial Vehicle Multispectral Images. MS Thesis, University of Florida.


    Progress 10/01/14 to 09/30/15

    Outputs
    Target Audience:Data analysis efforts were made through this reporting period. Several faculty members were involved inreviewing the data and its analysis results. The data collection, processing, and analysisresults werecompiled in a defended MS thesis and refrereed journal that is under review. This outout shouldreach out to the scientific community, system integratorspersonnel and growers interested in mobile ground-based remote sensing systems. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?A MS student student was trained on ground based image acquisition andanalysis. A PhD student is working on analyzing radiative tranfer models for vegetation areas. How have the results been disseminated to communities of interest?The results are dissiminated as MS thesis. A refereed journal publication is in preparation. What do you plan to do during the next reporting period to accomplish the goals?We plan to experiment with radiative transfer modeling to improve our ability to analyze low altitude unmanned aerial system imagery in agricultural setup.

    Impacts
    What was accomplished under these goals? An integrated sensing system to capture ground based hyperspectral and multispectral imagery was calibrated and tested. Navigation data was successfully synchronized and processed with the captured imagery to achieve georeferenced orthoimages. The positional accuracy of the resulting imagery was estimated. A MS thesis was prepared and defended based on this work and a refereed journal manuscript is in preparation.

    Publications

    • Type: Theses/Dissertations Status: Published Year Published: 2015 Citation: Sassi, N., 2015. MOBILE GROUND-BASED HYPERSPECTRAL SYSTEM IMAGE GEOREFERENCING. MS Thesis, University of Florida.


    Progress 03/26/14 to 09/30/14

    Outputs
    Target Audience: Although only tecnical experiemntation and data collection efforts were made through this reporting period, the project overall goal reaches to the growers community. Several students, professionals, faculty and technical staff were involved in field preparation in addition to data collection and preprocessing. Some of these efforts should be (were) presented to the audiance of the Gulf Coast Research and Education Center Ag Expo event in November of 2014. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided? During this reporting period, a MS student was trained on how to use the equipement and pre-process the data. A professional surveyor was involved in establishing the ground control points and image acquisition. Field data acquisition and analysis to assess image georeferencing accuracy constitutes the main objective of the MS student research/thesis. How have the results been disseminated to communities of interest? Nothing Reported What do you plan to do during the next reporting period to accomplish the goals? The data collecting within this reporting period will be analyzed to assess the georeferencing accuracy of the system. A MS thesis and peer reviewed publication should be the outcomes of this phase of the research. Other experiments will be conducted to assess the spectral properties and analyze remote sensing imagery in agricultural and natural resource management applications.

    Impacts
    What was accomplished under these goals? Images and trajectory infromation were collected and pre-processed to achieve the first objective. The data will be used to assess the positional accuracy of the system. By the end of this reporting period, the images and navigation data were pre processed and ready for actual integration to produce georeferenced images. Ground control points will be used to assess the positional accuracy of the georeferenced imagery.

    Publications