Image brightness and color balancing method and system for video stitching

Abstract

The invention discloses an image brightness and color balancing method and system for video stitching. The method comprises the steps of S1, acquiring two adjacent images to be stitched, calculating an image transformation matrix according to the two adjacent images to be stitched and setting an iteration suspensive condition; S2, determining the overlapping part of the two adjacent images to be stitched according to the image transformation matrix, and determining the brightness and color compensation coefficients of the images according to the brightness and color difference of the images with similar textures in the overlapping part; S3, adjusting the brightness and color of the two adjacent images to be stitched to obtain an adjusted image according to the brightness and color compensation coefficients of the images, and recalculating the image transformation matrix according to the adjusted image; S4, repeating S2 and S3 in sequence till the iteration suspensive condition is met, obtaining a final image transformation matrix and the brightness and color compensation coefficients of the final image, and saving the current original images to be stitched as reference images. By the adoption of the method and system, brightness and color jump occurring during video stitching can be eliminated.

Description

technical field [0001] The invention belongs to the field of image processing, and in particular relates to an image brightness and color equalization method suitable for video splicing. Background technique [0002] Video stitching technology is to use multiple fixed cameras to shoot the same scene from different angles of view. According to the correlation between images, video image stitching technology can produce images with wider field of view and higher resolution than a single image, so as to realize the scene 360-degree monitoring without blind spots. It has broad application prospects in military and civilian fields such as vehicle (ship) monitoring, individual shooting, square monitoring, intelligent transportation, and panoramic map drawing. [0003] Due to differences in lighting conditions, exposure time, photosensitive devices, etc., there are certain brightness and color differences in the images collected by different cameras. In the final stage of video s...

AI Technical Summary

Problems solved by technology

[0003] Due to differences in lighting conditions, exposure time, photosensitive devices, etc., there are certain brightness and color differences in the images collected by different cameras.

In the final stage of video stitching, that is, the image fusion stage, whether the image fusion method based on the optimal seam line is used (dividing the overlapping area into two parts, in the stitched image, the pixels on both sides of the seam line are from the reference image and the input image respectively ), or simply weighted and averaged the pixels in the overlapping area, there will be jumps in brightness and color at the image stitching, making the image transition unnatural

Moreover, the traditional image brightness and color equalization methods are mostly for still image stitching applications, and the real-time performance is not high.

Images