Evaluation method of laser jamming effect based on deep learning

Aiming at the problem that the evaluation of laser jamming effect is influenced by subjective experience and difficult to evaluate quantitatively, a laser jamming effect evaluation method based on deep learning was proposed. Firstly, the overall introduction of the YOLOV5 algorithm was given. Secondly, 3 020 laser jamming image from different angles and distances were produced. Then, the labeled data sets were trained to obtain the laser jamming effect evaluation model. Finally, the model was trained 300 times under the network models of YOLOV5x, YOLOV5l, YOLOV5m and YOLOV5s respectively. The experimental results show that the trained model can be used to evaluate the effect of laser interference image. The model could not only automatically label the laser interference area and evaluate the interference effect level, but also integrate the traditional strategy. It could calculate the area of the labeled area in the whole image as an auxiliary decision. The percentage of laser jamming area was automatically marked. The recognition accuracy was more than 80%, which is of great significance to the evaluation of laser jamming effect.