Project Summaries

10-776  Project Manager: P. F. O'Leary

IMPROVING DETECTION OF BOLL WEEVIL POPULATIONS AND VOLUNTEER COTTON PLANTS

Charles P. Suh, USDA-ARS

Timely detection and remediation of volunteer and regrowth cotton plants are deemed critical for completing boll weevil eradication in South Texas.  However, given the vast cotton acreage and diversity of habitats that can support volunteer plants, timely detection of such plants on an area-wide basis is a challenging process.  We explored the potential of using of airborne remote sensing imagery to detect volunteer and regrowth cotton plants over large geographical areas. One objective of this overarching project was to determine how spatial and radiometric resolutions of images affect the detection of cotton plants. Another objective was to apply airborne remote sensing techniques to compare spectral properties of regrowth cotton plants and to estimate the canopy coverage of regrowth cotton plants.  Results addressing the first objective demonstrated how image spatial and radiometric resolutions affect the detection and estimation of cotton canopy cover. Spatial resolution had a significant effect on plant identification and canopy cover estimation. If spatial resolution is less than half of the plant width, crop canopy cover can be accurately estimated using hard pixel classifiers; otherwise, mixed pixels will affect the estimation results and crop canopy cover cannot be accurately estimated using the four classifiers. Radiometric resolution reduced from 16 bits to 8 bits had little effect on single crop identification. Results from the second study indicate that airborne multispectral imagery can be used to estimate the development of regrowth cotton fields.  Although the LSU technique estimated canopy cover from mixed pixel reflectance, the detection threshold of about 0.5 m was higher than expected.  The non-uniformity (skippiness) of the regrowth plants was obvious at the onset of the regrowth stage and likely contributed to inflated measured ratios of plant width to row width.  Incorporation of additional reflectance signatures (e.g., cotton lint) in the LSU analysis may improve the accuracy of canopy cover estimates.  Nevertheless, the LSU technique of estimating regrowth development may provide boll weevil eradication program managers with a tool for monitoring previously-harvested fields for which the cotton stalks have not been destroyed due to limited access (e.g., flooding) or neglect. Collectively, our findings from both studies will be useful for determining the appropriate spatial and radiometric resolutions and classification methods for identifying volunteer and regrowth cotton plants.  More importantly, these findings put us one step closer to providing the Texas Boll Weevil Eradication Foundation and Texas Department of Agriculture with an effective and relatively economical approach for detecting volunteer and regrowth cotton plants over a large geographical area.

 

Project Year: 2012
 

Search 2012 Projects:


STATE PROGRAM
▸ Alabama
▸ Arizona
▸ Arkansas
▸ California
▸ Florida
▸ Georgia
▸ Louisiana
▸ Mississippi
▸ Missouri
▸ New Mexico
▸ North Carolina
▸ Oklahoma
▸ South Carolina
▸ Tennessee
▸ Texas
▸ Virginia

CORE PROGRAM
▸ Cotton Incorporated Fellow
▸ Cottonseed
▸ Crop Improvement
▸ Farm-to-Mill
▸ Production Efficiency
▸ Sustainable Cotton
▸ Variety Improvement

Project Manager
 

▸ All Project Nos.
 

Share This: