TY - GEN
T1 - A novel metric for image denoising algorithms
AU - Zhang, Yingtao
AU - Cheng, H. D.
AU - Huang, Jianhua
AU - Tang, Xianglong
PY - 2013
Y1 - 2013
N2 - Denoising algorithms, especially, the ones with contrast enhancement capability have many important applications. However, there is no an effective and accurate measurement for evaluating their performance objectively. This introduces a new metric, HME (Homogeneity Mean Error), to assess the denoising algorithms, especially those with enhancement capability. HME is based on the homogeneity property of each pixel which is sensitive to the changes of the structural information and noise levels, but insensitive to the changes of the contrast. Therefore, it can be utilized for evaluating the denoising algorithms. Various experiments are performed on images corrupted with different type of noise, the results demonstrate that HME is an effective and accurate metric for assessing the denoising algorithms with/without contrast enhancement. .
AB - Denoising algorithms, especially, the ones with contrast enhancement capability have many important applications. However, there is no an effective and accurate measurement for evaluating their performance objectively. This introduces a new metric, HME (Homogeneity Mean Error), to assess the denoising algorithms, especially those with enhancement capability. HME is based on the homogeneity property of each pixel which is sensitive to the changes of the structural information and noise levels, but insensitive to the changes of the contrast. Therefore, it can be utilized for evaluating the denoising algorithms. Various experiments are performed on images corrupted with different type of noise, the results demonstrate that HME is an effective and accurate metric for assessing the denoising algorithms with/without contrast enhancement. .
KW - HMD (Homogeneity Mean Difference)
KW - contrast enhancement
KW - denoising performance
KW - image quality assessment (QA)
KW - objective criterion
UR - https://www.scopus.com/pages/publications/84892845868
U2 - 10.1007/978-3-642-42057-3_68
DO - 10.1007/978-3-642-42057-3_68
M3 - 会议稿件
AN - SCOPUS:84892845868
SN - 9783642420566
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 538
EP - 545
BT - Intelligence Science and Big Data Engineering - 4th International Conference, IScIDE 2013, Revised Selected Papers
PB - Springer Verlag
T2 - 4th International Conference on Intelligence Science and Big Data Engineering, IScIDE 2013
Y2 - 31 July 2013 through 2 August 2013
ER -