Vision localization algorithms for apple bagging robot

Vision localization apple bagging robot is researched in this paper for young apples. The key technologies of the young fruit stereoscopic images recognizing and positioning are studied in the visible light of the natural environment. Firstly, the Otsu segmentation algorithm is used to preprocess the collected young apple images. Secondly, the improved connected component labeling algorithm is used to mark and label the preprocessed young apple images. Small areas where do not meet the requirements are removed by setting the threshold. The shape feature value circularity is used to recognize young apples and mark automatically. Thirdly, the binocular stereo vision system is used for three-dimensional positioning of the apple fruit. The three-dimensional information is used to grasp the fruit. A large number of experimental results prove the effectiveness and feasibility.