Detection Method for the Buds on Winter Vines Based on Computer Vision

According to the requirement of grape pruning in winter, an algorithm of detecting the buds of grape vines based on machine vision was proposed, laying the foundation of grape vines automotive pruning. The grapevines´ color image was captured indoor. The blue component of the color image was selected for the image preprocessing such as filter, threshold segmentation and noise removal. After that a binary image was gained. With the binary image, Rosenfeld algorithm was used in thinning to extract the skeleton of the grape branches. Because the morphological characteristic of the buds was similar to the corners, Harris algorithm was chosen to detect the point of buds from the skeleton image. The experiment result showed that it´s effective to detect the buds with the strategy of this paper. The recognition rate reached 70.2%.