Target surface image processing method and device

Abstract

The invention discloses a target surface image processing method and device. The method comprises the following steps: acquiring a template frame image, and extracting an edge contour of a target surface; determining a number of first angular points on the edge contour of the target surface of the template frame image; acquiring a current frame image, and extracting the edge contour of the target surface; determining a number of second angular points on the edge contour of the target surface of the current frame image; determining to form feature point pairs with the second angular points corresponding to each first angular point at the periphery of each first angular point according to the feature matching degree; and correcting the current frame image according to each feature point pair to obtain the corrected current frame image. According to the target surface image processing method and device disclosed by the invention, the processing based on video frames can avoid the influence of the change of illumination conditions on the accuracy of target reporting; and meanwhile, distortion correction with higher accuracy can be implemented on shaking images, and the problem of target surface shaking can be solved.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a target surface image processing method and device. Background technique [0002] At present, there are mainly two methods of automatic target reporting algorithm based on image recognition, bullet hole recognition based on single frame image and bullet hole recognition method based on video sequence, how to solve the target surface caused by environmental factors or bullet impact in the recognition process The shaking problem, as well as the accurate extraction of bullet holes from video frames and the identification of duplicate holes (partially overlapping bullet holes) have been an important research topic. [0003] In the traditional technical solution based on video frames, after obtaining the area where the target surface is located, the difference between the front and back video frames is directly performed, and then the connected domain analysis is performed o...

AI Technical Summary

Problems solved by technology

[0006] For the bullet hole detection scheme based on video frames, although the detection effect is better when the target surface does not shake and there are no heavy holes, but because the shake is not corrected or the correction accuracy is not enough, and there is no correction for heavy holes. Special analysis, therefore, the detection of bullet holes is less accurate in these two special cases;

[0007] For the bullet hole detection scheme based on a single-frame image, since the image is directly analyzed, it can solve the impact of the target surface shaking, but this method is sensitive to illumination changes, and the outdoor illumination changes are obvious and there may be insufficient illumination. Not very adaptable under uniform conditions

Images