Novel Neural Network Technology for Very Fast Analysis of Hyperspectral Imagery

Opto-Knowledge Systems, Inc.

Torrance, CA

An integrated benchtop hyperspectral sensor and analysis system was developed for use in laboratory and field applications

• Demonstrated the application of compact hyperspectral camera system utilizing a liquid crystal tunable filter for ground truth collection and analysis in support for remote sensing applications
• Demonstrated a novel algorithmic paradigm for neural network training specifically optimized for computationally intensive applications such as hyperspectral analysis
• Developed a Toolbox of additional algorithms for hyperspectral analysis

Integrated benchtop sensor and Toolbox to form a turnkey system

• The outcome of this STTR provided a basis for over five years of work in support of remote sensing and precision farming work in the San Joaquin Valley in California
• Precision Farming saves money by reducing the amount of chemicals used and reduces the environmental stress of such chemicals seeping into the ground water supply
• Precision farming can be a significant cost saver in California where the irrigation water is the main cost driver in the production of cotton

• Sponsored under various activities, including the NASA Hyperspectral EOCAP, OKSI implemented a remote sensing program to develop Precision Farming technology
• Under this technology, cotton growers would use the remote sensed data to monitor the health of their crops, and apply chemicals and irrigation only when and where needed, on a site specific basis

Return to NASA STTR Success Listings