|Pinter Jr, Paul|
Submitted to: Biannual Workshop in Color Photography and Videography in Resource
Publication Type: Proceedings
Publication Acceptance Date: 4/1/1997
Publication Date: N/A
Citation: N/A Interpretive Summary: Several new commercial companies are offering aircraft-based digital images to farmers to aid in managing crop and soil resources. Such precision information could result in both increases in yield and decreases in costly chemical and water applications. However, farmers need images that are independent of changing solar, sensor and atmospheric conditions, and few image providers are offering such a product. In this study, several approaches for reducing these effects were evaluated for both practicality and accuracy. We found that images could be corrected for unwanted effects by comparison with large, uniform objects within the image. That is, the farmer could deploy large tarps of known reflectance during each aircraft overpass and use these to correct the images for atmospheric variations and sensor geometry. When image providers incorporate this or other such approaches into their systems, their product will be infinitely improved and a multitude of potential agricultural applications will be realized.
Technical Abstract: Spectral imagery from airborne sensors is useful for monitoring seasonal trends in crop and soil conditions. However, for such applications it is necessary to convert uncorrected digital numbers (dn) from sensors to surface reflectance to minimize effects of sensor and solar variation. Such conversion has been attempted using ground-based reference targets, on-board up -and down looking sensors and in-flight atmospheric measurements. In this study, these approaches were evaluated for accuracy and practicality based on analysis with observed data and atmospheric models. The advantages of each approach were tabulated.