Algorithm Of Eye Location With Gray Projection And Hough Transform Based On ARM Platform
Caixia Liu · 2015
Image processing system based on ARM is a faster, smaller and more flexible application.Based on the analysis of human face and eye features, a new eye location algorithm based on ARM embedded system is presented.Firstly, the methods of histogram equalization and median filtering are used for the image enhancement and noise reduction.The adaptive Niblack method is used for segmentation and binarization.The approximate location of eyes is gotten by the horizontal and vertical gray-level histogram.Finally, on the basis of the analysis of the grayscale characteristics of human eye, Hough transform is used to get the right eye position.Experimental results show that this algorithm is easy to realize, reduces the blindness of the search process, and improves the speed.