Research on Multi-Modal Image Target Recognition Based on Asynchronous Depth Reinforcement Learning

The detection ability of multimodal image is not good under low light intensity. In order to improve the target recognition rate of multi-modal images, a multi-modal image target recognition method based on asynchronous deep reinforcement learning is proposed. The edge contour detection model of multi-modal image is established, and the light intensity of multi-modal image is adaptive fusion in the atmosphere scattering environment, and the information enhancement of multi-modal image in low mode is carried out by template matching. In this technique, scene contour feature matching method is used to refine the multimodal image, image features are extracted by fuzzy information tracking method, and significant transmission analysis is performed by brightness component. A multi-modal image target recognition method based on asynchronous deep reinforcement learning is proposed. Experiments show that the image has higher resolution and shorter processing time, which effectively improves the ability of image target recognition.