Fruits recognition in complex background using binocular stereovision
1.School of Electrical and Information Engineering， Jiangsu University, Zhenjiang, Jiangsu 212013, China; 2.Key Laboratory of Modern Agricultural Equipment and Technology, Ministry of Education and Jiangsu Province, Jiangsu University, Zhenjiang, Jiangsu 212013, China)
Abstract：According to standard binocular stereovision, the system was built to locate and distinguish fruits in complex background. Based on corner clustering, a novel algorithm was proposed to establish the invariant model of local feature. In order to obtain tridimensional information of position and shape, corner pixels were extracted to complete stereovision matching. Sufficient experrments of fruits recognition show that the algorithm which is based on single fruit characteristic has poor accuracy with parallax positioning accuracy of 78%, distance accuracy of 73% and colour accuracy of 68%. However, accuracy of the algorithm which bases on three fruit characteristics reaches 89%.The algorithm in the paper is proved to be effective to establish invariant mode and can enhance the efficiency of picking matching.
2011, Vol. 32 
