An Improved Randomized Hough Transform Method for the Cotton Recognition

Abstract

The cotton recognition would become rather difficult during the application of the cotton harvesting robot (CHR) in the case where the cotton is sheltered or covered by other objects. In this paper, a novel approach of cotton detection based on the improved randomized Hough transform (IRHT) for this case is proposed. Based on the contour information from the boundary trace, the mathematic model-based IRHT is derived based on a modified circular detection technique. It yields a well agreement with the requirements of the precision and rating of CHR.