为了更好地对图像进行平滑同时保持边缘不模糊,设计一种新的滤波方法.对基于该方法的图像滤波、细节增强等算法进行研究.首先,根据图像的亮度和颜色对图像进行分割,将图像分成不同的区域.接着,在不同的区域进行导引滤波,得到互不交叠的多个滤波子图像.然后,将这些子图融合,得到基于分割的改进导引滤波结果.最后,利用提出的改进导引滤波方法提出一种多尺度的细节增强方法.实验结果表明,在图像光滑和细节增强方面,提出的方法都要好于传统的导引滤波:提出的方法不仅能较好的光滑图像,同时保持边缘清晰,减少了传统滤波方法在边缘处的光晕现象.
In order to smooth image detail and preserve edge information well, a new edge-preserving filter is proposed in this paper.The filtering and detail enhancement method based on the proposed filter is studied.First, based on image intensity and color, the mean shift method is used for image segmentation, and we obtain several areas.Second, the guided image filter is used for each area, and several filtered images are separated with each other.Third, we fuse these filtered images to obtain a final result.Finally, a multi-scale detail enhancement method based on improved guided filter is proposed.Experimental results indicate that the proposed filter approach can obtain edge-preserving smoothed image, and detail enhanced image with less hole than guided filter.The proposed filter can obtain better results than guided filter in smoothing and detail enhancement.
参考文献
- 下载量()
- 访问量()
- 您的评分：
-
10%
-
20%
-
30%
-
40%
-
50%