Target detection method based on millimeter wave radar and visual feature attention fusion

The invention provides a target detection method based on millimeter wave radar and visual feature attention fusion. The method comprises the steps that millimeter wave radar point cloud data and visual image information are acquired; preprocessing the point cloud data, and fusing the preprocessed point cloud data with the visual image in a data layer; performing preliminary detection on the image, and extracting image features; performing target association on the radar point cloud and the image information, and extracting radar features; inputting the image features and the radar features into a feature attention fusion network for fusion; and decoding a detection result of the target by using a 3Dbox decoder and outputting the detection result. According to the invention, the size of the spatial information of the point cloud projected onto the image is adaptively adjusted by using the millimeter wave radar cross section intensity, and the problem that the size of the existing radar point cloud projected o