Detecting Bilateral Symmetry in Single Object Image Through Slope Matching

The computational treatment of symmetry is a complex problem in computer vision and computational geometry as well. In this paper detection of global bilateral symmetry in gray level images has been considered without prior segmentation. The proposed methodology detects symmetry based on slopes of the symmetric points which are filtered from the feature points extracted by Scale Invariant Feature Transform (SIFT) algorithm. This approach depends on all the points inherent to line of bilateral symmetry exhibits same inclination in 2D plane. This concept helps us further to find out bilateral symmetry in the image. For demonstration, the algorithm is tested on random images The main contribution of this paper is that the approach is computationally very simple and can be extended very easily to be applied to images having multiple symmetries.