Abstract
The developments of technology grow fast in recent years, and the digital equipment becomes essential in human live, such as mobiles and camera. People can use the digital equipment to take a picture fast and conveniently to record their live. However, poor illumination will impair the quality of images. The state-of-the-art methods can’t deal with the problems of different poorly illuminated environment at the same time. Therefore, this paper presents an adaptive method to improve the problems which is beneficial to related applications. We propose an adaptive method based on the human visual system. First, based on JND model, we use 127 of the intensity as the standard to separate images into two parts which are the dark region and the bright region respectively. Second, we calculate the weighted arithmetic mean of the dark region and the bright region to represent the influence respectively. Third, we use the relationship between the dark region and the bright region to determine the characteristic of images. We produce an enhancement curves with the characteristics to adjust images. In the other hand, we use not only local standard deviation but also the weight of JND to enhance the detail and the contrast, so the enhanced images are provided with more details and edge information. In the final experimental result, we compare the result with the proposed method and other image enhancement methods. We use the experiments to prove the proposed method not only have more details and better contrast but also keep the original color without distortion. Therefore, with the proposed method, we can get a better image for other applications, such as recognition and recording.