• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>International conference on graphic and image processing >Shadowed Non-local Image Guided Filter
【24h】

Shadowed Non-local Image Guided Filter

机译：带阴影的非本地图像引导滤镜

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Guided image filter has been widely used in image processing. Considering the Non-local model is an excellent method for global information accumulation, the non-local image guided filter has been proposed and shown good performance in many image processing tasks by utilizing the non-local similarity of the guidance image. In this paper, we introduce a shadowed non-local image guided filter derived from the concept of shadowed sets. The shadowed non-local model applies more reliable non-local information by suppressing the low similarity values of the guidance image to zero and boosting high similarity values to the maximum of the non-local similarity set. The thresholds of suppression and boosting are determined automatically based on the concept of shadowed sets. Experimental results on several image processing tasks including image denoising, depth super-resolution, and image dehazing demonstrate the superiority of shadowed set based approach.
机译：引导图像过滤器已广泛用于图像处理中。考虑到非局部模型是一种用于全局信息累积的出色方法，因此提出了一种非局部图像导引滤波器，并利用指导图像的非局部相似性在许多图像处理任务中显示了良好的性能。在本文中，我们介绍了从阴影集的概念派生的阴影非局部图像导引滤波器。阴影非局部模型通过将引导图像的低相似度值抑制为零并将高相似度值提升为非局部相似度集的最大值来应用更可靠的非局部信息。抑制和增强的阈值是根据阴影集的概念自动确定的。在一些图像处理任务（包括图像去噪，深度超分辨率和图像去雾）上的实验结果证明了基于阴影集的方法的优越性。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号