Underwater Image Denoising via AMP with Renyi Entropy

被引：0

作者：

Lu, Jihua ^{[1
]}

Yan, Lei ^{[2
]}

Wang, Huan ^{[1
]}

Feng, Lihui ^{[1
]}

Li, Xinbin ^{[2
]}

机构：

[1] Beijing Inst Technol, Beijing, Peoples R China

[2] Yanshan Univ, Qinhuangdao, Peoples R China

来源：

WUWNET'19: PROCEEDINGS OF THE INTERNATIONAL CONFERENCE ON UNDERWATER NETWORKS & SYSTEMS | 2019年

关键词：

Underwater image denoising; shot noise; approximate message passing; Renyi Entropy; wavelet transform;

D O I：

10.1145/3366486.3366531

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

Approximate message passing (AMP) in wavelet domain has been widely used in image denoising. However, the AMP-based denoising method is not robust to the 'shot noise', which is dynamic and short-period in the wavelet domain. First, a preprocessing filter is designed with Entropy to combat the restlessness of noise component in spatial domain. Then, a shot noise filter is designed via AMP with Renyi Entropy in wavelet domain. By using the two-step denoising method, we can 'individualized' denoise both in spatial and wavelet domains. The experimental results show that, with pre-processing filtering, our denoising method outperform the design via AMP with wavelet thresholds.

引用

页数：2

共 3 条

[1] A method to estimate the abundance of fish based on dual-frequency identification sonar (DIDSON) imaging [J].