An Automatic Target Image Mosaic Algorithm
Lingjia Gu · 2007
In order to resolve the problem of capturing double shadow images in the shooting range test, a novel automatic image mosaic algorithm is proposed in the paper. Each double shadow image will be split into the odd field image and the even field image, then two field images will be complemented to two frame images, finally mosaic the middle image of two frame images by the algorithm. The mosaic algorithm mainly adopts SUSAN algorithm to abstract the feature points in the two images, then a pseudo matching set is obtained by comparing local neighborhoods of features through intensity cross-correlation method, these pseudo matches are divided into inliers and outliers using robust RANSAC algorithm, then estimate the point transformation matrix between two images accurately, finally seamlessly stitch two overlapping images. The experimental results show that the image mosaic effect is wonderful; the algorithm expands the applicable scope of the traditional mosaic algorithm.