Abstract: In view of the external factors that affect the recognition and planning of tapping trajectory, such as complex and variable lighting conditions and similar target and background, a combinatorial optimization algorithm for the recognition and planning of tapping trajectory of natural rubber tree was proposed. Firstly, the algorithm preprocessed the original image to reduce the impact of image noise and to enhance the feature information between the profile and background. Then, OTSU algorithm was used for threshold segmentation to obtain binary image and morphology operation was used to eliminate residual noise points, after which the Canny operator was selected to extract the glue track by comparing and analyzing different boundary extraction operators. Finally, the trajectory fitting operation was used to plan the new tapping trajectory. In order to verify the stability and effectiveness of the proposed algorithm, the experiment was carried out on 180 images collected in Hainan National Natural Rubber Forest. The results showed that the proposed combinatorial optimization algorithm could recognize and plan the next tapping trajectory under natural light conditions and the success rate of recognition and planning reached 89.4%. It can provide technical reference for the research and development of intelligent tapping robot.