This paper describes a complementary mechanism that attempts to represent the Interest points (key points)[7][9][10] by a few of the intrinsic parameters in a rotation, scale and translation invariant manner. The parameter for this mechanism of finding interest point is that, the feature points or interest points[7][9][10] correspondences when the shapes of interest are each defined by a single, closed contour and the binary shape we obtained through segmentation represents some real-world object, which was sampled and binarized, and it is that object's identification that we want to estimate. That means by joining those key points, an image can be extracted. Corner is so special since it is the intersection of two edges; it represents a point in which the directions of these two edges change. Hence, the gradient of the image (in both directions) have a high variation, which can be used to detect it.

Computer Vision Object recognition Image matching Contour Tracing Interest Point detection.