Skip to main content
ARS Home » Research » Publications at this Location » Publication #158906
Title: EVALUATION OF AERIAL PHOTOGRAPHY FROM MODEL AIRCRAFT FOR REMOTE SENSING CROP BIOMASS AND NITROGEN STATUS

Author
item Hunt Jr, Earle
item Cavigelli, Michel
item Daughtry, Craig
item McMurtrey Iii, James
item Walthall, Charles

Submitted to: Precision Agriculture
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 4/30/2005
Publication Date: 8/1/2005
Citation: Hunt, E.R., Cavigelli, M.A., Daughtry, C.S., McMurtrey, J.E., Walthall, C.L. 2005. Evaluation of aerial photography from model aircraft for remote sensing crop biomass and nitrogen status. Precision Agriculture. 6(4):359-378.

Interpretive Summary: Remote sensing is a key technology for precision agriculture because the actual progress of a crop can be monitored. However, low cost satellite data, such as from Landsat, may not have the spatial resolution (pixel size) required for monitoring of a field, have a long time period between acquisition and delivery, and may have significant cloud cover during the growing season. High spatial resolution airborne and satellite data have high costs, which may outweigh the benefits of the information. Radio controlled model aircraft have been used to acquire aerial photography by hobbyists, and may provide a low cost alternative platform for high spatial resolution imagery. A digital color camera was attached to a fixed-wing aircraft and used to obtain imagery over various fields at the Beltsville Agricultural Research Station. Automated processing of the imagery requires an index to account for differences in camera exposure, topography, and position of the sun. The Normalized Green Red Difference Index (NGRDI) was developed from the green and red bands of the digital camera. This index was linearly related to dry biomass at low amounts, and reached a maximum value when biomass was greater than 120 grams per square meter for soybean and corn. This index was not directly related to chlorophyll content of corn. There are many advantages of model aircraft platforms for precision agriculture, and digital imagery acquired from these platforms show within-field variability in crop growth making these technologies potentially useful for precision agriculture.

Technical Abstract: Remote sensing is a key technology for precision agriculture to assess actual crop conditions. Commercial, high-spatial resolution imagery from aircraft and satellites are expensive so the costs may outweigh the benefits of the information. Hobbyists have been acquiring aerial photography from radio-controlled model aircraft; we evaluated these imagery for use in estimating nutrient status of corn and crop biomass of corn, alfalfa, and soybeans. Based on requirements determined from previous work, we optimized an aerobatic model aircraft for acquiring pictures from a digital color camera. Colored tarpaulins were used to calibrate the images, and due to differences in exposure, there were large differences in digital number (DN) for the same reflectance. To account for differences in exposure, a normalized green-red difference index of (DN green - DN red)/(DN green + DN red) was related to crop biomass. For soybeans, alfalfa and corn, dry biomass from zero to 120 g m-2 was linearly correlated to the green-red index, but for biomass greater than 120 g m-2 in corn and soybean, the index did not increase further. In a fertilization experiment with corn, the green-red index did not show differences in nitrogen status, even though areas of low nitrogen status were clearly visible on late-season digital photographs. The SAIL (Scattering of Arbitrarily Inclined Leaves) canopy radiative transfer model verified that the normalized green-red difference index would distinguish between vegetation and soil at low leaf area index and that nitrogen status would not be detectable. There are many advantages of model aircraft platforms for precision agriculture, and aerial photography acquired from these platforms show within-field variability in crop growth. Automated analysis of within-field variability requires more work on sensors that can used with these platforms.