ABSTRACT:
In this paper an adaptive image enhancement technique has been discussed. image enhancement procedure is a necessary pre processing for facial image detection. First, the distribution of image information is being measured by using the entropy error rate (EER). Then, by integrating a histogram Ridge analysis technique and an optimal intensity transform method, a novel adaptive image enhancement algorithm has been implemented to minimise the EER of an enhanced image. Basically, the image enhancement problem is a parameter (EER) optimisation problem. The results demonstrate That this method of preprocessing can significantly improve the accuracy of facial image detection and performs much better than classical histogram-based enhancement techniques such as linear stretching and histogram equalization.