In order to improve the recognition speed and accuracy, this study aimed at overlapping apple proposes a positioning method based on the maximum optimization, for firing of overlapping circles.Due to the diversity of apple growth posture, the overlap and shade between the apples will influence the recognition efficiency of apple harvesting robot vision system.Firstly,using the K-means method to segment the captured overlapping apple image under the L*a*b* color space, this color space has the uniformity, so it can obtained the overlapping apple target.The segmented image is still exists some burr, noise, holes, ere, so need carries on the some morphological processing, such as edge smoothing, noise removal and holes filling, etc, and then extracted the outline of the overlapping apple.And then to calculate the minimum distance between the pixels in the circle and the edge of outline, got the minimum distance function, and find out the local maximum in these functions, that is the centers of two circles.Finally, the radius is determined by the minimum distance between centers to the edge of the outline, to realize the positioning of overlapping apple.In order to verify the validity of maximum recognition method for the overlapping apples, and comparing with corrosion method and the Hough transform method,using two overlapping images captured by different apple varieties to repeat experiments.The experimental results show that, from the recognition speed and accuracy, the maximum optimization method is optimal.Through experimental analysis, under the L*a*b* color space, the K-means clustering algorithm can segment the overlapping apples target effectively.The opening function of morphology processing can effectively smooth edge of apple, binarization and hole filling can make apple outline is complete.Using the maximum method can accurately find apple's center, the radius be determined by the minimum distance of center to the edge, this method can achieve the recognition and positioning of overlap apple.Although maximum method has obvious advantages, however, there are certain defects, under the condition of the apple edge incomplete,may appear the bigger error and deviation, lead to inaccurate positioning, so in this situation, the image need to repair and preprocessing, and then to fitting.Under the condition of overlapping circles shade not severe and the edge relatively complete, the process effect of maximum method is good, from the precision and speed, it is better than Hough transform and corrosion method.It can satisfy the requirements of the harvesting robot for accuracy and real-time performance, has a certain practicality.