Multiple Mixed Modes: Single-Channel Blind Image Separation
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
As one of the pivotal techniques of image restoration, single-channel blind source separation (SCBSS) iscapable of converting a visual-only image into multi-source images. However, image degradation often resultsfrom multiple mixing methods. Therefore, this paper introduces an innovative SCBSS algorithm to effectivelyseparate source images from a composite image in various mixed modes. The cornerstone of this approach is anovel triple generative adversarial network (TriGAN), designed based on dual learning principles. The TriGANredefines the discriminator's function to optimize the separation process. Extensive experiments havedemonstrated the algorithm's capability to distinctly separate source images from a composite image in diversemixed modes and to facilitate effective image restoration. The effectiveness of the proposed method isquantitatively supported by achieving an average peak signal-to-noise ratio exceeding 30 dB, and the averagestructural similarity index surpassing 0.95 across multiple datasets.
Abstract
As one of the pivotal techniques of image restoration, single-channel blind source separation (SCBSS) iscapable of converting a visual-only image into multi-source images. However, image degradation often resultsfrom multiple mixing methods. Therefore, this paper introduces an innovative SCBSS algorithm to effectivelyseparate source images from a composite image in various mixed modes. The cornerstone of this approach is anovel triple generative adversarial network (TriGAN), designed based on dual learning principles. The TriGANredefines the discriminator's function to optimize the separation process. Extensive experiments havedemonstrated the algorithm's capability to distinctly separate source images from a composite image in diversemixed modes and to facilitate effective image restoration. The effectiveness of the proposed method isquantitatively supported by achieving an average peak signal-to-noise ratio exceeding 30 dB, and the averagestructural similarity index surpassing 0.95 across multiple datasets.
참고문헌 (22)
- W. Fu, "The research of SCBSS technology : survey and prospect", Journal of Beijing University of Posts and Telecommunications, 40(5), pp.1~11 (2017)
- Hepzibah Christinal Anandharaj, "Review on Deep Learning Methodologies in Medical Image Restoration and Segmentation", Current Medical Imaging Formerly Current Medical Imaging Reviews, 19(8), pp.844~854 (2023), Bentham Science Publishers Ltd.[DOI]
- Ganghua Liu, "A Windowed-Total-Variation Regularization Constraint Model for Blind Image Restoration", JIPS(Journal of Information Processing Systems), 18(1), pp.48~58 (2022), 한국정보처리학회[DOI]
- Jiangyu Han, "Exploring single channel speech separation for short-time text-dependent speaker verification", International Journal of Speech Technology, 25(1), pp.261~268 (2022), Springer Science and Business Media LLC[DOI]
- Wen-Hsing Lai, "RPCA-DRNN technique for monaural singing voice separation", EURASIP Journal on Audio, Speech, and Music Processing, 2022(1) (2022), Springer Science and Business Media LLC[DOI]
- Bing Wu, "Fuzzy logic based dynamic decision-making system for intelligent navigation strategy within inland traffic separation schemes", Ocean Engineering, 197, pp.106909~ (2020), Elsevier BV[DOI]
- D. Kitamura, "Music signal separation by supervised nonnegative matrix factorization with basis deformation", Proceedings of 2013 18th International Conference on Digital Signal Processing (DSP), pp.1~6 (2013)
- P. Dey, "Single channel blind source separation based on variational mode decomposition and PCA", Proceedings of 2015 Annual IEEE India Conference (INDICON), pp.1~5 (2015)
- Z. Zhang, "Image restoration based on principal component analysis", Journal of Changchun Normal University (Natural Science Edition), 38(6), pp.21~23 (2019)[DOI]
- Z. C. Fan, "SVSGAN: singing voice separation via generative adversarial network", Proceedings of 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp.726~730 (2018)
- D. Stoller, "Adversarial semi-supervised audio source separation applied to singing voice extraction", Proceedings of 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp.2391~2395 (2018)
- X. Lin, "An improved resampling particle filter blind separation algorithm", Journal of Chongqing University of Posts and Telecommunications (Natural Science Edition), 31(4), pp.502~508 (2019)[DOI]
- Kai Zhang, "Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising", IEEE Transactions on Image Processing, 26(7), pp.3142~3155 (2017), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- Emad M. Grais, "Two-Stage Single-Channel Audio Source Separation Using Deep Neural Networks", IEEE, 2017, pp.25(9), 1773~1783[DOI]
- C. Zheng, "Blind enhancement algorithm for throat microphone speech based on LSTM recurrent neural networks", Journal of Data Acquisition and Processing, 34(4), pp.615~624 (2019)[DOI]
- T. Wang, "Natural image classification based on improved SAE network", Information Technology, 2016(8), pp.1~4 (2016)[DOI]
- "Conditional generative adversarial nets" (2014)[URL]
- Y. C. Subakan, "Generative adversarial source separation", Proceedings of 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp.26~30 (2018)
- "Single-channel signal separation and deconvolution with generative adversarial networks" (2019)[URL]
- T. Liu, "Research on single-channel blind deconvolution algorithm for multisource signals", Journal of Electronics & Information Technology, 44(1), pp.230~236 (2022)[DOI]
- T. Yin, "Research on single-channel blind deconvolution algorithm for ancient Chinese character restoration", Small Microcomputer System, 42(2), pp.414~417 (2021)[DOI]
- Z. Yi, "DualGAN: unsupervised dual learning for image-to-image translation", Proceedings of the IEEE International Conference on Computer Vision, pp.2849~2857 (2017)