Adaptive deblocking technique based on separate modes for removing compression effects in JPEG coded images
Adaptive deblocking technique based on separate modes for removing compression effects in JPEG coded images
复制标题
DOI：
10.1080/1206212x.2019.1567045
复制
发表时间：
2019-01
影响因子：
--
通讯作者：
Amanpreet Kaur;J. S. Sidhu;J. S. Bhullar
中科院分区：
文献类型：
--
作者：
Amanpreet Kaur;J. S. Sidhu;J. S. Bhullar
摘要 自适应后处理滤波技术通过减轻压缩伪影显着提高了解码图像的感知质量。然而，基于块的编码图像中仍然存在一些明显的压缩伪影，尤其是在低比特率下。在本文中，我们提出了一种三步去块滤波技术，可以消除压缩伪影、不需要的噪声和角点异常值。所提出的去块技术已应用于各种 JPEG 压缩图像，并根据 PSNR、PSNR-B 等标准指标将结果与现有后处理技术进行比较。实验仿真结果表明，所提出的技术可以有效提高各种图像的感知质量，并且优于现有的去块技术。
ABSTRACT Adaptive post-processing filtering techniques significantly improve the perceptual quality of decoded images by alleviating compression artifacts. However, there are still some noticeable compression artifacts in the block-based coded images especially at a low bit rate. In this paper, we proposed a three-step deblocking filter technique that removes the compression artifacts, undesired noise, and corner outliers. The proposed deblocking technique has been applied to the variety of JPEG compressed images and results are compared with existing post-processing techniques on the basis of standard metrics such as PSNR, PSNR-B. Experimental simulation results illustrate that the proposed technique could effectively improve the perceptual quality of various images and outperform existing deblocking technique.