Semi-automated analysis of citrus grove data from remotely-sensed devices can reduce ground surveying for appraisal, crop recording, yield analysis, and grove management. Although they generally start with the same aerial infrared imagery for their data collection, grove managers and independent appraisers each have their own independent requirements for data analysis. By making the analysis of this imagery more efficient and robust, image understanding technology can provide significant benefit to the entire citrus industry. The Phase I research effort submitted in this proposal will focus on the potential agricultural uses of knowledge-based computer image understanding technologies that were originally developed for military, manufacturing, and medical domains. These techniques will extend the standard suite of remote sensing tools to provide robust mechanisms for automatic identification and measurement of individual and/or hedge-row citrus trees from aerial color infrared image data. These new techniques will address difficult problems such as the isolation of individual tree crowns, automatic spatial registration and normalization, delineation of groves, tree counting and classification. A follow-on Phase II program will expand the recognition criteria, analysis goals, tree species and remote sensing media in an integrated prototype system.
Anticipated Results/Potential Commercial Applications of Research:: A semi-automated system that can assist in the analysis of aerial color infrared imagery of citrus groves will be commercially attractive to several markets. Such systems will isolate individual trees (even when canopies overlap), provide counts according to grove boundaries, classify the trees according to spectral characteristics, register collections of images according to GPS, and provide counts and summary information for GIS packages. The potential users of this system include: county appraisers, grove managers, and researchers. The system will be developed to run on personal computers to minimize cost and make the package cost effective for the largest possible market.
