分块离散余旋变换压缩图像一种统一后处理算法.pdf

分块离散余旋变换压缩图像一种统一后处理算法.pdf

  1. 1、本文档共11页,可阅读全部内容。
  2. 2、原创力文档(book118)网站文档一经付费(服务费),不意味着购买了该文档的版权,仅供个人/单位学习、研究之用,不得用于商业用途,未经授权,严禁复制、发行、汇编、翻译或者网络传播等,侵权必究。
  3. 3、本站所有内容均由合作方或网友上传,本站不对文档的完整性、权威性及其观点立场正确性做任何保证或承诺!文档内容仅供研究参考,付费前请自行鉴别。如您付费,意味着您自己接受本站规则且自行承担风险,本站不退款、不进行额外附加服务;查看《如何避免下载的几个坑》。如果您已付费下载过本站文档,您可以点击 这里二次下载
  4. 4、如文档侵犯商业秘密、侵犯著作权、侵犯人身权等,请点击“版权申诉”(推荐),也可以打举报电话:400-050-0827(电话支持时间:9:00-18:30)。
查看更多
Unified Artifacts Reduction for Block DCT Compressed 1 Images * Zhai Guangtao, Xu Yi , Yang Xiaokang, Zhang Wenjun, Yu Songyu Institute of Image Communication and Information Processing, Shanghai Jiao Tong University, Shanghai, China (200240) Abstract We propose a unified scheme to alleviate the blocking, ringing and blurring artifacts in block DCT compressed images. The image is decomposed into 2 levels using the over-complete wavelet transform, and the wavelet coefficients are filtered with consideration of local image properties. For the smooth region, the wavelet coefficients are averaged over the range of blocking artifacts’ extension in the over- complete wavelet transform domain on respective levels. For the texture region, only the coefficients on the first level are averaged to prevent over-smooth of the image. And for the edge region, the coefficients are first smoothed and then high-pass filtered to enhance the edges so as to restrain the ringing and blurring artifacts. The inverse transform of the filtered coefficients is performed towards the quality enhanced image. And a quantization constraint is imposed on the final result for better PSNR and visual performance. Keywords: Image postprocessing, Deblocking, Overcomplete Wavelet Expansion 1. INTRODUCTION Nowadays the Block-based DCT (BDCT) is the most widely used compression technique for image (JPEG) and video sequence (MPEG[1], h.26x[2]). The block-wise DCT

文档评论(0)

xuefei111 + 关注
实名认证
内容提供者

该用户很懒,什么也没介绍

1亿VIP精品文档

相关文档