Image matching method and video processing method

Abstract

The invention relates to an image matching method and a video processing method. The image matching method comprises the steps: carrying out the regional division of an image scene, so as to obtain sub-regions; extracting feature points in a first image and a second image, and carrying out the matching of the feature points, so as to obtain the matched feature points between the first and second images; grouping the matched feature points according to the sub-regions; selecting a matching feature point subset from the matched feature points according to a matching model between the first image and the second image, wherein the matching feature point subset refers to groups as many as possible; and carrying out the fitting of the matching model through employing the matching feature point subset, so as to obtain an image matching result. The method can improve the matching robustness in a jitter removing process of a video.

Description

technical field [0001] The invention relates to the field of image processing, in particular to an image matching method and a video processing method. Background technique [0002] In an actual photography system, the video obtained by a mobile platform such as a vehicle, a handheld or an aircraft includes not only the active motion of the imaging system, but also the random motion of the mobile platform. The unstable video generated by the random motion will make people feel tired and make it difficult to extract useful information. Therefore, how to convert unstable video into stable video is of great significance. [0003] Video deshaking, also called video stabilization, is a very important video processing technology. This video processing technology is designed to eliminate video jitter, to ensure clear and stable video images, so that videos can be better compressed, thereby improving video quality and speed. [0004] Video jitter refers to the shaking and blurrin...

AI Technical Summary

Problems solved by technology

[0006] The pixel method uses the relationship between pixel gray values ​​for motion estimation, but it is sensitive to noise and requires rich image information

[0007] The block matching method performs motion estimation on the pixels in the block as a whole, so it is more robust than the pixel method, but the accuracy and computational complexity of the algorithm are greatly affected by the number, size, search range and search strategy of the block

[0008] The phase correlation method estimates the direction and speed of motion by calculating the cross-power spectrum of adjacent frames. It has strong noise resistance, but has a large computational complexity and is susceptible to local motion interference.

[0009] The feature matching method is based on human visual characteristics, and estimates the global motion parameters of the camera by extracting and matching the features of adjacent frames. Compared with other algorithms, it is closer to the process of processing motion information by the human visual system, but when the scene When there are other moving targets in the scene, different motion parameters will appear in the scene, and the feature point extraction may only be limited to the area of ​​a certain motion parameter. At this time, the result will be limited by the feature extraction, which will affect the robustness and accuracy of matching.

Images