Distinctive Feature Detection Algorithm through Clustered Mask of Stereo Images

Abstract

Feature Detection and their similarities is a central problem in the field of vision based modeling. It is significant in the field of automation to locate objects and obstacles, or in the case of computer-guided surgery, or study of mathematical structures as well as in machine vision. It arises from the task of distinctive feature detection to classify and recognize structures of their observed silhouette. Defining natural distances between image discontinuities creates a metric space of shapes, whose mathematical structure is inherently relevant to the classification task. One intriguing metric space is identified from using conformal mappings through image matching with the help of mask from 2D stereo images. The poor image quality of many video surveillance cameras effectively renders them useless for the purposes of identifying features from grabbed image. Under certain conditions, however, it may be possible by comparing multiple stereo video frames for such identification tasks. For this a simple and computationally efficient technique for feature detection to detect the objects of an environment is discussed. The efficiency of this technique is shown on real video sequences.