Image recognition method for multi-body separation trajectory prediction in delivery test

The invention relates to an image recognition method for predicting a multi-body separation track in a drop test, and aims to carry out multi-body drop separation test shooting in a wind tunnel by using a high-speed camera so as to capture a single-plane separation track. According to the method, a single-plane separation track is obtained, then axial displacement, longitudinal displacement and pitch angles of multiple bodies in the single-plane separation track are recognized in real time through the image recognition technology, the method belongs to the technical field of wind tunnel tests, and the method comprises the image calibration technology, the image recognition technology based on pixel edge detection, the multi-body recognition technology and the subject matter follow-up technology. The test is carried out by adopting a throwing separation method, the single-plane axial displacement, the longitudinal displacement and the pitch angle when one projectile body is separated into two projectile bodies