Skip to main content
ARS Home » Research » Publications at this Location » Publication #296411

Title: A two-camera imaging system for pest detection and aerial application

item Yang, Chenghai
item Hoffmann, Wesley
item Martin, Daniel - Dan
item Lan, Yubin
item Fritz, Bradley - Brad
item Westbrook, John
item Suh, Charles

Submitted to: ASABE Annual International Meeting
Publication Type: Abstract Only
Publication Acceptance Date: 7/26/2013
Publication Date: 7/26/2013
Citation: Yang, C., Hoffmann, W.C., Martin, D.E., Lan, Y., Fritz, B.K., Westbrook, J.K., Suh, C.P. 2013. A two-camera imaging system for pest detection and aerial application. ASABE Annual International Meeting. Paper No. 131598742.

Interpretive Summary:

Technical Abstract: This presentation reports on the design and testing of an airborne two-camera imaging system for pest detection and aerial application assessment. The system consists of two digital cameras with 5616 x 3744 effective pixels. One camera captures normal color images with blue, green and red bands, while the other camera is equipped with a 720-nm long-pass filter to obtain near-infrared images. The color camera is also equipped with a GPS receiver to allow geotagged images. A remote control device is used to trigger both cameras simultaneously for image acquisition. Images from each camera are stored in 16-bit RAW and 8-bit JPEG files in a CompactFlash card. A post processing procedure is used to align the color and NIR images and the aligned images are then combined as four-band images. The imaging system was tested under various flight heights and speeds and the camera settings were optimized for image acquisition for diverse ground covers. Images were captured at altitudes from 305 m (1000 ft) to 3050 m (10000 ft) to achieve ground pixel sizes from 0.1 to 1.0 m. The system was used to detect invasive weeds, crop diseases, and volunteer and regrowth cotton plants, and to assess the efficacy of aerial herbicidal application. The system performed well in both stationary and airborne testing conditions. Preliminary analysis of the airborne images showed that this system will be useful for pest detection and aerial application efficacy assessment.