| 基于小波系数分割的局部自适应阈值图像去噪 |
| |
| 引用本文： | 李亚丽,张松林,韩杰.基于小波系数分割的局部自适应阈值图像去噪[J].测绘通报,2020,0(5):43-46,122. |
| |
| 作者姓名： | 李亚丽 张松林 韩杰 |
| |
| 作者单位： | 同济大学测绘与地理信息学院, 上海 200092 |
| |
| 基金项目： | 上海市自然科学基金（19ZR1459700）；国家重点研发计划（2017YFB0502700） |
| |
| 摘 要： | 针对自适应小波阈值去噪中方形局部窗口不能很好地适应小波系数自身边缘特征的问题,本文提出了一种基于图像分割的局部自适应小波阈值去噪方法。该方法首先对图像进行多尺度小波分解,其次利用图像分割技术对图像的各尺度小波系数分别进行分割,分割技术选用SLIC超像素分割法得到具有相似统计特性的不规则局部块,然后对每个分割块内的小波系数进行BayesShrink阈值估计和软阈值收缩,最后通过小波逆变换得到去噪图像,并在3幅标准测试图像和一幅高光谱影像上进行试验。试验结果表明,本文提出的方法能更好地适应小波系数自身的边缘特征,不仅能够获得更好的视觉效果,而且能够达到较高水平的数值指标。
|
| 关 键 词： | 图像去噪 小波变换 图像分割 局部自适应 阈值去噪 |
| 收稿时间： | 2019-09-20 |
| 修稿时间： | 2019-11-18 |
Wavelet coefficients segmentation for locally adaptive threshold image denoising method |
| |
| LI Yali,ZHANG Songlin,HAN Jie.Wavelet coefficients segmentation for locally adaptive threshold image denoising method[J].Bulletin of Surveying and Mapping,2020,0(5):43-46,122. |
| |
| Authors: | LI Yali ZHANG Songlin HAN Jie |
| |
| Institution: | College of Surveying and Geo-Informatics, Tongji University, Shanghai 200092, China |
| |
| Abstract: | Aiming at the problem that the square local window of the adaptive wavelet threshold denoising can not well adapt to the edge features of the wavelet coefficients, a new image segmentation-based method is proposed. In this method, multi-scale wavelet decomposition is firstly performed on the noisy image. Then, the different scale wavelet coefficients are segmented by using the SLIC superpixel segmentation technology with which some local blocks with similar statistical characteristics will be obtained. BayesShrink thresholds of the wavelet coefficients in every irregular block are estimated and soft threshold shrinkage is utilized. Finally, a denoised image is acquired by the inverse wavelet transform. In this paper, three standard test images and a hyperspectral image are tested. The experimental results show that the proposed method can better adapt to the edge features of the wavelet coefficients, and can not only get better visual effects, but also achieve a higher level of numerical indicators. |
| |
| Keywords: | image denoising wavelet transform image segmentation local adaptation threshold denoising |
| 本文献已被 CNKI 维普 等数据库收录！ |
| 点击此处可从《测绘通报》浏览原始摘要信息 |
| 点击此处可从《测绘通报》下载免费的PDF全文 |
|
