CNN-Based Real-Time Parameter Tuning for Optimizing Denoising Filter Performance

被引:5
作者
Mukherjee, Subhayan [1 ]
Kottayil, Navaneeth Kamballur [1 ]
Sun, Xinyao [1 ]
Cheng, Irene [1 ]
机构
[1] Univ Alberta, Edmonton, AB T6G 2R3, Canada
来源
IMAGE ANALYSIS AND RECOGNITION, ICIAR 2019, PT I | 2019年 / 11662卷
关键词
Filter parameter tuning; CNN; Denoising; BM3D; GPU; IMAGE; SPARSE; REPRESENTATIONS; ALGORITHM;
D O I
10.1007/978-3-030-27202-9_10
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
We propose a novel direction to improve the denoising quality of filtering-based denoising algorithms in real time by predicting the best filter parameter value using a Convolutional Neural Network (CNN). We take the use case of BM3D, the state-of-the-art filtering-based denoising algorithm, to demonstrate and validate our approach. We propose and train a simple, shallow CNN to predict in real time, the optimum filter parameter value, given the input noisy image. Each training example consists of a noisy input image (training data) and the filter parameter value that produces the best output (training label). Both qualitative and quantitative results using the widely used PSNR and SSIM metrics on the popular BSD68 dataset show that the CNN-guided BM3D outperforms the original, unguided BM3D across different noise levels. Thus, our proposed method is a CNN-based improvement on the original BM3D which uses a fixed, default parameter value for all images.
引用
收藏
页码:112 / 125
页数:14
相关论文
共 50 条
  • [41] CNN-Based Deblurring of THz Time-Domain Images
    Ljubenovic, Marina
    Bazrafkan, Shabab
    Paramonov, Pavel
    De Beenhouwer, Jan
    Sijbers, Jan
    COMPUTER VISION, IMAGING AND COMPUTER GRAPHICS THEORY AND APPLICATIONS, VISIGRAPP 2020, 2022, 1474 : 477 - 494
  • [42] Real-time UFIR parameter identification
    Siegl, Steffen
    Svaricek, Ferdinand
    AT-AUTOMATISIERUNGSTECHNIK, 2020, 68 (03) : 176 - 195
  • [43] CNN-BASED PARAMETER SELECTION FOR FAST VVC INTRA-PICTURE ENCODING
    Tech, Gerhard
    Pfaff, Jonathan
    Schwarz, Heiko
    Helle, Philipp
    Wieckowski, Adam
    Marpe, Detlev
    Wiegand, Thomas
    2021 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP), 2021, : 2109 - 2113
  • [44] A New Real-Time Embedded Video Denoising Algorithm
    Petreto, Andrea
    Romera, Thomas
    Lemaitre, Florian
    Masliah, Ian
    Gaillard, Boris
    Bouyer, Manuel
    Meunier, Quentin L.
    Lacassagne, Lionel
    2019 CONFERENCE ON DESIGN AND ARCHITECTURES FOR SIGNAL AND IMAGE PROCESSING (DASIP), 2019, : 47 - 52
  • [45] Optimizing energy efficiency of CNN-based object detection with dynamic voltage and frequency scaling
    Weixiong Jiang
    Heng Yu
    Jiale Zhang
    Jiaxuan Wu
    Shaobo Luo
    Yajun Ha
    Journal of Semiconductors, 2020, (02) : 85 - 94
  • [46] SVD-based portable device for real-time hoarse voice denoising
    Manfredi, C
    Landini, L
    Faita, F
    Gemignani, V
    DSP 2002: 14TH INTERNATIONAL CONFERENCE ON DIGITAL SIGNAL PROCESSING PROCEEDINGS, VOLS 1 AND 2, 2002, : 857 - 860
  • [47] Real-time chinese traffic warning signs recognition based on cascade and CNN
    Yining Gao
    Guangyi Xiao
    Journal of Real-Time Image Processing, 2021, 18 : 669 - 680
  • [48] CNN-LSTM Based Smart Real-time Video Surveillance System
    Iqrar, Waqas
    Abidien, Malik ZainUl
    Hameed, Waqas
    Shahzad, Aamir
    2022 14TH INTERNATIONAL CONFERENCE ON MATHEMATICS, ACTUARIAL SCIENCE, COMPUTER SCIENCE AND STATISTICS (MACS), 2022,
  • [49] A Novel Denoising Approach Based on Improved Invertible Neural Networks for Real-Time Conveyor Belt Monitoring
    Guo, Xiaoqiang
    Liu, Xinhua
    Zhang, Xu
    Krolczyk, Grzegorz M.
    Gardoni, Paolo
    Li, Zhixiong
    IEEE SENSORS JOURNAL, 2023, 23 (03) : 3194 - 3203
  • [50] Real-time chinese traffic warning signs recognition based on cascade and CNN
    Gao, Yining
    Xiao, Guangyi
    JOURNAL OF REAL-TIME IMAGE PROCESSING, 2021, 18 (03) : 669 - 680