Full-Text Search:
Home|About CNKI|User Service|中文
Add to Favorite Get Latest Update

A fast correlation method for image matching using pixels abstract

CAO Ju~(1,2),MA Jie~1,TAN Yi-hua~1,TIAN Jin-wen~1 gent Control, Inst.For Pattern Recognition and Artificial Intelligence,HUST,Wuhan 430074,China; 2.Beijing Aerospace Autocontrol Institute,Beijing 100854,China)  
The common method for image matching is correlation matching.The calculating number was very large when correlation function was used between two matching images.Two steps approach was proposed in this paper for image matching that was coarse and exact way.Firstly every pixel's gray level was summed up along every column to calculate the mean of column pixels.Then one dimension signal was coarsely matched using one order correlation function.Secondly the points with larger correlation matching values were chosen as the better matching points.These less points were used in the original image as the coordinate origins to calculate the two orders correlation functions.The point with maximum correlation value would be the optimal matching position.Experiences with three kinds of images prove that this approach is faster than conventional one and the matching result is accurate.
Download(CAJ format) Download(PDF format)
CAJViewer7.0 supports all the CNKI file formats; AdobeReader only supports the PDF format.
©CNKI All Rights Reserved