Full-Reference Metric Adaptive Image Denoising

Kenji Hara, Kohei Inoue, Kiichi Urahama

研究成果: 著書/レポートタイプへの貢献会議での発言

抜粋

The performance of image denoising algorithms generally depends largely on the selection of the parameters. We address the problem of optimizing the denoising parameters to achieve maximum denoising performance. Most existing methods for no-reference denoising parameter optimization either use the estimated image noise or individual no-reference image quality evaluation metrics. In the paper, we introduce a natural image statistics based on the generalized Gaussian distribution and an elastic net regularization regression model. Consequently, we propose their use for performing no-reference parameter optimization in the state-of-the-art BM3D denoising algorithm, adaptively depending on which of the following most widely used full-reference image quality evaluation metrics is optimized: FSIM, SSIM, MS-SSIM, without any knowledge of the noise distribution.

元の言語英語
ホスト出版物のタイトル2019 IEEE International Conference on Image Processing, ICIP 2019 - Proceedings
出版者IEEE Computer Society
ページ2419-2423
ページ数5
ISBN(電子版)9781538662496
DOI
出版物ステータス出版済み - 9 2019
イベント26th IEEE International Conference on Image Processing, ICIP 2019 - Taipei, 台湾省、中華民国
継続期間: 9 22 20199 25 2019

出版物シリーズ

名前Proceedings - International Conference on Image Processing, ICIP
2019-September
ISSN(印刷物)1522-4880

会議

会議26th IEEE International Conference on Image Processing, ICIP 2019
台湾省、中華民国
Taipei
期間9/22/199/25/19

    フィンガープリント

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Vision and Pattern Recognition
  • Signal Processing

これを引用

Hara, K., Inoue, K., & Urahama, K. (2019). Full-Reference Metric Adaptive Image Denoising. : 2019 IEEE International Conference on Image Processing, ICIP 2019 - Proceedings (pp. 2419-2423). [8803231] (Proceedings - International Conference on Image Processing, ICIP; 巻数 2019-September). IEEE Computer Society. https://doi.org/10.1109/ICIP.2019.8803231