An adaptive fusion panoramic image mosaic algorithm based on circular LBP feature and HSV color system
Hou Jia Kang, Xuefei Lv, Zhang Wenhui · 2020
In view of the sharp edge of stitching line, object splitting, uneven color fusion and obvious stitching traces in the process of panoramic image stitching, a new panoramic image stitching algorithm is proposed. Before stitching, the adaptive image color fusion based on HSV color system is carried out, the local texture features of the image are described by circular LBP Operator, and the energy accumulation distribution table is obtained by using the optimized pruning dynamic planning algorithm. Finally, the image fusion is carried out in the overlapping area by interlacing backtracking and optimizing Gaussian filtering wave. Experimental results show that compared with the traditional stitching algorithm, the image quality level is improved, the color is too uniform, and there is no obvious hum in stitching.