Skip to main content
ARS Home » Plains Area » College Station, Texas » Southern Plains Agricultural Research Center » Aerial Application Technology Research » Research » Publications at this Location » Publication #313876
Title: Early identification of cotton fields using mosaicked aerial imagery

Author
item Yang, Chenghai
item Suh, Charles
item Westbrook, John
item Eyster, Ritchie

Submitted to: National Cotton Council Beltwide Cotton Conference
Publication Type: Proceedings
Publication Acceptance Date: 2/6/2015
Publication Date: 5/2/2015
Citation: Yang, C., Suh, C.P., Westbrook, J.K., Eyster, R.S. 2015. Early identification of cotton fields using mosaicked aerial imagery. Proceedings of the Beltwide Cotton Conference. p. 901-906.

Interpretive Summary: Early identification of cotton fields and volunteer cotton via remote sensing is important for the boll weevil eradication program and reducing the risk of re-infestation. This study used aerial imagery to identify cotton fields over large areas before cotton plants start to bloom. Airborne color images taken from an 8 km by 14.5 km (5 mi by 9 mi) cropping area near College Station, Texas in the 2014 growing season were directly mosaicked as one single image and then classified into different crop and cover types using image classification techniques. Preliminary results showed that the mosaicked color image was able to accurately differentiate cotton fields from other crop and cover types. These results will be useful for boll weevil eradication program managers to quickly and efficiently identify cotton fields and potential areas for volunteer and regrowth cotton plants without having to physically go to each and every field.

Technical Abstract: Early identification of cotton fields is important for advancing boll weevil eradication progress and reducing the risk of re-infestation. Remote sensing has been used for crop identification for decades, but limited work has been reported on early identification of cotton fields. The objective of this study was to evaluate mosaicked aerial imagery for identifying cotton fields before cotton plants start to bloom. A two-camera imaging system was used to acquire RGB and NIR images with a pixel array of 5616 × 3744 over an 8 km by 14.5 km (5 mi by 9 mi) cropping area near College Station, TX in the 2014 growing season. The images were acquired at 3048 m (10000 ft) above ground level along two flight lines to achieve a pixel size of 1.0 m. The individual images were mosaicked as one single image and then classified into different crop and cover types using supervised classification techniques. Preliminary results showed that the mosaicked color image was able to accurately differentiate cotton fields from other crop and cover types. These results will be useful for boll weevil eradication program managers to quickly and efficiently identify cotton fields and potential areas for volunteer and regrowth cotton plants without having to physically travel to each and every field.