Tuesday, 8 November 2005
11

Use of a Consumer Grade Digital Camera and Geographic Information System Image-Processing Algorithm to Correlate Potato Yield with the Red-Green-Blue Color Space.

Edgar Po, Crops and Soils Department, Michigan State University, 414 PSSB, Michigan State University, East Lansing, MI 48823 and Sieglinde Snapp, Crops and Soils Sciences Department, Michigan State University, A440A Plant & Soil Sci. Bldg., East Lansing, MI 48824-1325.

Plant growth dynamics leading ultimately to yield is highly influenced by its interaction with electromagnetic radiation, especially in the visible and the near infrared part of the spectrum. Spatial variability in plant utilization of the green, the red and the infrared part of the spectrum could provide an early indicator of plant vigor and may present an opportunity to do corrective measure within-season rather than after the crops have been harvested. Real time evaluation of plant spectral interaction calls for a system that can be implemented on a regular temporal interval, and with a minimal of pre-activity preparation and cost requirement. Currently, digital images are obtained using satellites and or aircraft involving use of sophisticated instrumentation, both at the acquisition and processing stages. Planning out image acquisition through those systems involves considerable preparations. A consumer grade digital camera previously calibrated with a research grade spectro-radiometer, was utilized to obtain field nadir images of various stages of potato plant growth. Images were inputted to a Geographic Information System which provided output image statistics for spatial analysis of potato yield variability. Preliminary results indicated significant correlation of the green, infrared and red bands of acquired images not only with potato yield recorded using a yield monitor, but with other factors affecting potato growth and development. Improvement in the correlation coefficient was accomplished by lowering the cut-off properties of the customized lens system used in the study. Mechanization of filter array, a customized water-down geographic information system and an improved platform to take a rapid whole field image are recommended to further enhance the utility of this monitoring system.

Back to Precision Farming and Remote Sensing in Agricultural Systems
Back to A08 Integrated Agricultural Systems

Back to The ASA-CSSA-SSSA International Annual Meetings (November 6-10, 2005)