Recognize Human Head Shot Top Down Based on Hough Transform

Abstract

Hough transform has the advantage of incomplete edge recognition. A large number of practices show that: for the actual circular object detection, the detected points are usually concentrated in a relatively continuous arc. If the detected points are scattered in various positions of the circle, it is usually the interference of noise that causes the wrong recognition of the target. On this basis, this paper proposes an algorithm, that is, to calculate the standard deviation of the angle of the detected edge points, and compare the standard deviation value with the detected edge points, so as to judge whether the recognition of the circular object is correct. In this paper, the algorithm is applied to human head recognition and good results are obtained.