To solve the problem that overlapping grape is difficult to separate in natural environment,a method of overlapping grape segmentation based on impr oved FCN and concave point search was proposed.The main steps are as follows:F irst,a comparative analysis of different color spaces and different color chann els was conducted for the collected summer black overlapping grape images.It wa s found the contrast between the target and the background was large in the H co mponent under the HSV color space,so the H component was extracted from the ove rlapping grape images.The H component of the overlapping grape was input into t he convolutional multiscale fusion FCN to obtain the binary graph,and then the centroid of the overlapping grape was solved and the contour convex hull was det ected.Second,axis of a single bunch of grapes was detected according to the ma ximum angle of the vector.The concave region was obtained by subtracting the bi nary image from the convex hull of the contour before separation.Symmetrical co nvex hull was made according to the axis,and then the concave region containing segmented concave points was obtained.Third,the geometric calculation model o f overlapping grape boundary was established,the contour of the selected concav e region was traversed,and then the segmented concave point was searched accord ing to the shortest distance.In addition,the 400overlapping grape images were verified,the average standard segmentation error was 4.03,the detection success r ate was 92.5%,and the algorithm consumed time of 0.40~0.59s.Compared with the existing methods,the accuracy of overlapping grape segmentation in this paper w as not easily affected by the details of contour inflection points,and the accu racy of recognition was improved.It provided a new overlapping grape segmentati on method for picking robot.