During the process of pear damage detection based on computer vision, there are many noise in image acquisition. For a long time, rapid detection equipment cannot be utilized to identify pear damage. To solve this problem, a method of identifying damage pear is studied. This paper proposes a method for pear damage identification based on geometric features. First, with morphological method to remove the noise of pearimage to obtain the best edge detection method and best threshold suitable to the pear damage, so as to detect the damaged edge of the image. The experimental results show that this method can effectively identify the damaged pear. Compared with conventional edge detection method, the recognition accuracy has been significantly improved, and solve the long-standing problem of automatic identification and improve the detection efficiency of pear damage.
展开▼