Evaluation of Stereo Images Matching

Image matching and finding correspondence between a stereo image pair is an essential task in digital photogrammetry computer vision. Stereo images represent the same scene from two different perspectives, therefore they typically contain high degree of redundancy. This paper includes evaluation implementing manual as well auto-match that acquired with overlapped area. Particular target points are selected to be matched manually (22 points). Auto-matching, based on feature-based (FBM) method, has been applied these by using BRISK, FAST, Harris, MinEigen algorithms. Auto conducted main phases: extraction (detection description) features. The techniques used prevalent algorithms depend local point (corner) Also, performance assessed according results obtained various criteria, such number auto-matched auto-matched. study aims determine evaluate total root mean square error (RMSE) comparing coordinates those auto-matching each According experimental results, BRISK algorithm gives higher points, which equals 2942, while Harris 378 representing lowest points. All FAST algorithms, 3 9 only respectively. Total RMSE its minimum value given match first image, it 0.002651206 mm, provide second 0.002399477 mm.