Submitted to: Biannual Workshop in Color Photography and Videography in Resource
Publication Type: Proceedings
Publication Acceptance Date: 8/10/2001
Publication Date: 12/6/2004
Citation: Thomson, S.J., Sudbrink, D.L. 2004. Digital imaging from agricultural aircraft: system configurations and constraints for integrated pest management, weed detection, and determination of crop status. Proceedings of the 18th Biannual Workshop in Color Photography and Videography in Resource. American Society for Photogrammetry & Remotte Sensing, Amherst, MA. CD-ROM, paper number 0025.pdf. 4 pgs. Interpretive Summary: Imaging systems for agricultural aircraft have been proposed for targeting specific field areas for chemical or water application. When weed patches, areas of crop stress, or insect pressure are detected in specific field locations, those field areas are isolated for subsequent spraying, fertilization, or water application. This can potentially reduce amounts of fapplied pesticide and water and still provide good crop yields. Three imaging systems are being evaluated for use on the agricultural aircraft, and they are in various stages of evaluation and development. A system using digital video has shown limited success in classifying weeds, and preliminary data suggests possibilities for detecting crop yield and vigor. A second system using a high-resolution still-digital camera has thus far been evaluated for the pilot's ability to control the camera and provide field images at correct field positions. Electronics and programming are being implemented to allow more consistent imaging of the proper field areas and remove the pilot from having to determine when to take the image. A third imaging system is the DuncanTech MS-3100 digital camera, which permits three bands of image information. This camera is presently being configured along with a computer-based imaging platform for the airplane.
Technical Abstract: Three imaging systems are being evaluated for potential use in agricultural aircraft as tools for remote sensing. The first imaging system uses a digital video camera with interchangeable filters. This system was evaluated for its preliminary capability in mapping and distinguishing weeds, developing vegetative index maps, and locating areas of insect infestation. Images for distinguishing weeds were successfully classified using ENVI 3.4 image analysis software. Normalized Difference Vegetation Index (NDVI) was estimated from two flight passes using digital video, but further study will be needed to determine the suitability of matching multiple field images for deriving vegetative indices. A second system using a high-resolution still-digital camera was evaluated for pilot controls, and modifications were proposed to facilitate positional accuracy for imaging. A third system for the spray plane is the DuncanTech MS-3100 three-band digital imaging camera, ordered in a color-infrared configuration. This system is being evaluated, and an on-board computer is being constructed for use in the spray plane.