- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
[IEEE 2018 OCEANS - MTS/IEEE Kobe Techno-Ocean (OTO) - Kobe (2018.5.28-2018.5.31)] 2018 OCEANS - MTS/IEEE Kobe Techno-Oceans (OTO) - Underwater Image Denoising Based on Non-Local Methods
摘要： Due to the medium scattering and light absorption, the acquired underwater optical images suffer from poor visibility. Absorption and scattering effects are caused not only by the water itself but also by other floating particles. Moreover, backscattering is strengthened during the light is reflected and deflected multiple times by the suspended particles before reaching the camera. The backscattering gives rise to specific strong correlated noise together with a constant level degrading the image contrast. Although there are many effective dehazing methods to remove the veil background, they have ignored the influence of backscattering noise, which still exists in the dehazed images. In addition, because of the special characteristics of backscattering noise, the traditional image denoising methods are difficult to make sense, even cause distortion. Accordingly, this paper employed the non-local denoising methods which are referred to Non-Local Means algorithm and Block-Matching 3D collaborative filtering to suppress the backscattering noise. The experimental results show that they can effectively remove the backscattering noise of underwater image and achieve better image quality.
关键词： backscattering noise,BM3D,NLM,underwater image
更新于2025-09-23 15:22:29
-
[IEEE 2019 18th International Conference on Optical Communications and Networks (ICOCN) - Huangshan, China (2019.8.5-2019.8.8)] 2019 18th International Conference on Optical Communications and Networks (ICOCN) - Optimization BM3D Algorithm Based on Pseudo-3D Transform
摘要： A optimization BM3D algorithm based on pseudo-3D transformation is proposed to solve the problem of huge computational of BM3D algorithm. Firstly, one-dimensional wavelet transform was carried out on the third dimension of 3D similar blocks to get one-dimensional the lowest-frequency coefficient block gotten above was transformed by two-dimensional transformation. Following, two-dimensional inverse transformation was performed on the lowest-frequency coefficient blocks after hard threshold filtering or Wiener filtering. And finally, one-dimensional inverse transformation was carried out on the third dimension to get the noise reduced image. The performance of the proposed optimization method is very closed to the original BM3D algorithm, but computation of pseudo-3D transformation is only 1/3 of that of 3D transformation.
关键词： BM3D,Pseudo-3D Transform,Similar Image Blocks,Image De-noising
更新于2025-09-23 15:19:57