An Image Binarization Method Based on HVS

This paper proposes an image binarization method based on the human vision system. It firstly gets the image threshold value according to the Euclidean distance, then uses the standard deviation to determine the visual sensitivity coefficient of each pixel according to the human visual characteristic, finally performs image binarization according to the gray value, the image threshold and the visual sensitivity coefficient. This method can both effectively eliminate the ghosts, and keep the character stroke well. The experimental results show that this method is more effective than Ostu method and Bernsen method. It is superior to Ostu method and Bernsen method in the entropy. It also provides a new effective way for image binarization.