Multi-stage image denoising with the wavelet transform

•A dynamic convolution is used into a CNN to address limitations in depth and width of lightweight CNNs for pursuing good denoising performance.•The combination of a signal processing technique and discriminative learning technique is used for image denoising.•Enhanced residual dense architectures a...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Pattern recognition Ročník 134; s. 109050
Hlavní autoři: Tian, Chunwei, Zheng, Menghua, Zuo, Wangmeng, Zhang, Bob, Zhang, Yanning, Zhang, David
Médium: Journal Article
Jazyk:angličtina
Vydáno: Elsevier Ltd 01.02.2023
Témata:
CNN
Dynamic convolution
Image denoising
Signal processing
Wavelet transform
Signal processing
Dynamic convolution
CNN
Wavelet transform
Image denoising
ISSN:0031-3203, 1873-5142
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:•A dynamic convolution is used into a CNN to address limitations in depth and width of lightweight CNNs for pursuing good denoising performance.•The combination of a signal processing technique and discriminative learning technique is used for image denoising.•Enhanced residual dense architectures are used to remove redundant information for improving denoising effects. Deep convolutional neural networks (CNNs) are used for image denoising via automatically mining accurate structure information. However, most of existing CNNs depend on enlarging depth of designed networks to obtain better denoising performance, which may cause training difficulty. In this paper, we propose a multi-stage image denoising CNN with the wavelet transform (MWDCNN) via three stages, i.e., a dynamic convolutional block (DCB), two cascaded wavelet transform and enhancement blocks (WEBs) and a residual block (RB). DCB uses a dynamic convolution to dynamically adjust parameters of several convolutions for making a tradeoff between denoising performance and computational costs. WEB uses a combination of signal processing technique (i.e., wavelet transformation) and discriminative learning to suppress noise for recovering more detailed information in image denoising. To further remove redundant features, RB is used to refine obtained features for improving denoising effects and reconstruct clean images via improved residual dense architectures. Experimental results show that the proposed MWDCNN outperforms some popular denoising methods in terms of quantitative and qualitative analysis. Codes are available at https://github.com/hellloxiaotian/MWDCNN.
ISSN:0031-3203
1873-5142
DOI:10.1016/j.patcog.2022.109050