Image Mosaics Algorithm Based on SIFT Feature Point Matching and Transformation Parameters Automatically Recognizing
A mosaic algorithm of different scale image registration and adaptive is proposed in this thesis,against to the large amount of calculation and poor robustness,as well as cannot well solve the problem of image mosaic of images who are in different scales in the traditional image mosaic method.The match and mosaic of different scale and rotated images is achieved through feature point matching and automatically recognizing of transform geometric parameters between images.First,using SIFT extraction algorithm to extract the feature points of the image to be mosaic.Second,achieving feature points matching according to the principal of mutual information maximum.Third,automatically recognize the relationship of transformation parameters between two images according to the geometric information of the matching pairs.Finally,the projective transformation matrix that reflected the image translation,rotation and scaling information can be obtained and the image stable mosaic can be achieved.
SIFT Image mosaic Feature point matching Automatic recognize of transformation parameters
Pengrui Qiu Ying Liang Hui Rong
Kunming Metallurgy College,650033 Kunming,China
国际会议
杭州
英文
1561-1564
2013-03-22（万方平台首次上网日期，不代表论文的发表时间）