Image Mosaic Algorithm and Its Application to the Microimage of Grass Seeds
Lina Ning, Xin Pan, Lin Zhai, Fan Han · 2013
This paper presents an image mosaic algorithm and its application to the microimage of grass seeds.The following main steps are involved, firstly in the registration stage, the scale invariant feature transform (SIFT) is employed to obtain initial matches.Then the Random Sample Consensus (RANSAC) is used to remove incorrect matches effectively.Finally in the fusion stage, fade-in and fade-out method is applied to smooth seams which exist in the stitched image.Since the effect is not obvious, the adaptive Gamma correction method is used to weaken illumination effect on image quality to obtain a stitched image, which contains more information than each of the original images.Experimental results demonstrate that the proposed approach achieves good performance for grass seeds microimage mosaic in both subjective and objective evaluations.