SBIR-STTR Award

This project will focus on the strategic, routine incorporation of medium-resolution satellite imagery into operational agricultural assessments for the global crop market. Automated algorithms for rapid extraction of field-level crop area statistics from Landsat and Landsat-class imagery (including SLC-off L7 data, AWiFS, ASTER, and NPOESS/OLI) will be developed. For prototype development, the project will collaborate with the Production Estimates and Crop Assessment Division of the USDA Foreign Agricultural Service. The algorithms, based on Bayesian Probability Theory, will incorporate multiple lines of evidence in the form of prior and conditional probabilities and will implement an innovative approach to supervised image classification allowing for automated class delineation. The knowledge-based expert classifiers prototyped during Phase I will be tested and validated at selected pilot sites across the globe. The overall results of the project will enhance global agricultural production estimates by improving the timeliness and accuracy of field-level crop area estimates. It addresses the NASA SBIR subtopic by developing unique, rapid analyses for the extraction of crop area statistics from medium-resolution imagery. The developed technologies will support both the scientific and commercial applications of ES data and will be benchmarked for practical use against an international model for agricultural production estimates

This project is focusing on the strategic, routine incorporation of medium-resolution satellite imagery into operational agricultural assessments for the global crop market. Automated algorithms for rapid extraction of field-level crop area statistics from Landsat and Landsat-class imagery (including Landsat 5 TM, Landsat 7 ETM+, AWiFS, ASTER, SPOT, LDCM, etc.) are under development. For prototype development, the project is collaborating with the Production Estimates and Crop Assessment Division of the USDA Foreign Agricultural Service. The Phase I prototype algorithms, based on Bayesian Probability Theory, incorporate multiple lines of evidence in the form of prior and conditional probabilities and implement an innovative approach to supervised image classification allowing for automated class delineation. The knowledge-based expert classifiers prototyped during Phase I were tested and validated at selected pilot sites across the globe. The results of the Phase I work have clearly demonstrated the technical feasibility of the GDA approach to automated crop area assessment with medium resolution imagery. Development undertaken during Phase I resulted in a robust, fully functional set of modules that are capable of processing large volumes of data and allow for accurate crop detection, area estimation, and crop acreage change assessment with minimal user intervention. The prototype algorithms were tested on a range of test sites, sensors, crop types, and crop conditions. A non-rigorous validation study proved the reliability and accuracy of the prototype algorithms. The overall results of the project will enhance global agricultural production estimates by improving the timeliness and accuracy of field-level crop area estimates.