Machine vision-based guidance system for agricultural grain harvesters using cut-edge detection

ABSTRACT. A machine vision guidance system was developed for an agricultural combine harvester. The guidance algorithm separated the uncut crop rows from the surrounding background material, parameterized the crop rows, and calculated a guidance signal. A single monochrome camera mounted on the cab of the combine supplied the crop images. The algorithm was developed for corn and was tested under both laboratory and field conditions. Test results showed that the algorithm was capable of accurately locating crop rows in the image and providing a satisfactory lateral position signal for automated combine guidance. At a speed of 1.3 m/s (3 mph), the system was capable of guiding the combine at the same accuracy level as the GPS recording system available