Virtual exhibition image stitching fusion method based on light compensation and improved optimal stitching lines

To address the problems of uneven quality of the panoramic image in the case of constructing a virtual exhibition hall and the existence of uneven lighting and color differences at the stitching seams, an optimal stitching line multi-scale image stitching fusion method using lighting compensation and new energy function is proposed. First, the method uses gamma correction for global luminance compensation of the image to be stitched, where the coefficient of gamma correction is the ratio of the sum of its luminance components in HSV (Hue, Saturation, Value) color space for each of the overlapping regions, secondly, the saturation component of HSV color space is used instead of the original color energy function to obtain the best stitching mask map according to the best stitch search criterion, and finally, the Laplace image pyramid method is used to Finally, the images to be stitched are stitched and fused according to the mask map using the Laplace image pyramid method in different scale-spaces. The results show that the method mitigates the visual fragmentation near the stitching seams due to uneven illumination and color differences in the case of constructing a virtual exhibition hall, and has a better stitching fusion effect.