Image Enhancement based on Spatial Multi-scale Homomorphic Filtering and Local Entropy Guided Image Filtering

摘要

During the process of image acquisition, the quality of images will be affected by many factors, and the most prominent problem is the negative effect of illumination. In order to overcome above impacts, a new color image illumination enhancement technique is proposed for improving the quality of color images. First, the spatial multi-scale technique is applied to local homomorphic filtering, which can eliminate the unevenness of image illumination to the utmost extent while avoiding over-enhancement. Then, local entropy guided image filtering is proposed, which can remove the blocking artifacts while fully retaining the detailed information. In order to avoid the color distortion, the above two algorithms are performed on V channel in the HSV color space, which forms a new color image enhancement technique for removing negative effects of illumination. Experimental results show that our method outperforms other algorithms, which obtains the better and higher performance in subjective and objective aspects.