| 摘要 |
<p>Block noises at block boundaries are removed, but when a contour line in the original image matches a block boundary, the degree of block noise removal is adjusted and image quality is improved. An adaptive filter has a step of calculating the weight of pixels near the block boundary by a high-pass filter, and a step of determining an adjusted value for the pixels near the block boundary based on the weight. The pixel values near the block boundary are adjusted and block noises are removed. A low-pass filter is applied to pixels near the block boundary to determine a filtered value. The adjusted values of the pixels near the block boundary are determined by obtaining a weighted sum of the original pixel value and the filtered value. Because the high-pass filter recognizes the definition of pixels near the block boundary, and the low-pass filter reduces the difference between pixels near the block boundary, the adaptive filter outputs the adjusted value, and this value removes block noise while retaining image definition.</p> |
