首页> 外文期刊>Journal of visual communication & image representation >Perceptually-lossless image coding based on foveated-JND and H.265/HEVC
【24h】

Perceptually-lossless image coding based on foveated-JND and H.265/HEVC

机译:偏心JND和H.265 / HEVC的视觉无损图像编码。

获取原文
获取原文并翻译 | 示例
获取外文期刊封面目录资料

摘要

Removing perceptual redundancy plays an important role in image compression. In this paper we develop a foveated just-noticeable-difference (FJND) model to quantify the perceptual redundancy in the image and integrate it in the H.265/HEVC intra encoding framework to provide a perceptually lossless image coding solution. Different to the conventional JND models, our proposed FJND model considers the relationship between contrast masking effect and the foveation properties of HVS. Furthermore, to achieving the perceptually lossless coding, the FJND model is integrated in the H.265/HEVC framework by determining the quantization parameter to ensure that the resulting distortion is no larger than the FJND threshold. The experiments demonstrate that the proposed method effectively improves the compression performance. (C) 2016 Elsevier Inc. All rights reserved.
机译:消除感知冗余在图像压缩中起着重要作用。在本文中,我们开发了一个偏心的正好注意差异(FJND)模型来量化图像中的感知冗余,并将其集成到H.265 / HEVC帧内编码框架中,以提供感知无损的图像编码解决方案。与传统的JND模型不同,我们提出的FJND模型考虑了对比度掩盖效果和HVS的偏光特性之间的关系。此外,为了实现感知上的无损编码,通过确定量化参数以确保所得失真不大于FJND阈值,将FJND模型集成到H.265 / HEVC框架中。实验表明,该方法有效地提高了压缩性能。 (C)2016 Elsevier Inc.保留所有权利。

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