Adaptive High Boost Filtering for Increasing Grayscale and Color Image Details

摘要

Generally, the high boost filtering often produced a lot of unpleasant noise, especially the flat area. Moreover, it also gives color distortion in the edge regions when applied for the color image. In this paper, we propose the adaptive sharpening filter based on the high boost filtering for increasing the image details. This filter consists of two parts as the original and the High-frequency Component (HC) parts. The HC part was controlled by using the adaptive weight matrix, which was constructed from the non-linear combination. The Laplacian operator was employed as the high-pass filter for generating both the HC and the adaptive weight matrix. The simulated step-edge image, grayscale images, and color images were applied for observing our filter performance. The traditional bilateral filter and high boost filter are used as the comparison filters. The experimental results show that our proposed filter can increase the edge sharpness while it did not enhance the unpleasant noise for the flat area. Moreover, it also reduces the color distortion and edge ringing when applied to the color images.