Abstract
Aiming at the problem that the traditional iris location method is not ideal for non-ideal iris location,a non-ideal iris location method based on BP neural network and Hough transform is proposed.Firstly,the iris image to be located is divided into blocks,and the feature vector of each image is extracted.Secondly,the method of BP neural network is used to detect the edge block and non-edge block,and the outer boundary is accurately located by mathematical operation of edge block.Then,the center of the outer boundary is the seed pixel,and the inner boundary is region grown.Finally,Hough transform is used to locate the inner boundary accurately,and it can also achieve good results when eyelashes and eyelids are blocked seriously and girls draw eyeliner.
|