Underwater structure surface crack detection device and method based on compound-eye bionic vision

The invention discloses an underwater structure surface crack detection device and a method based on the compound-eye bionic vision. According to the technical scheme of the invention, the contradiction between the field-of-view and the resolution of a single aperture imaging system during the imaging process can be effectively overcome. During the super-resolution reconstruction process for a plurality of images of the same target, the complementary information in each image can be utilized to effectively handle with the underwater complex light environment and the noise interference, so that a high-resolution image with the details thereof rich and clear can be obtained. According to the self-adaptive detection method for detecting the surface crack of an underwater structure based on a gray-scale image, the foreground information in the image can be extracted more effectively and the crack detection range is minimized. The influence of the underwater noise on the crack detection is greatly reduced, and the crack detection accuracy is improved. Therefore, the method and the device have a good application prospect.