Group-Based Sparse Representation for Compressed Sensing Image Reconstruction with Joint Regularization

Abstract

Achieving high-quality reconstructions of images is the focus of research in image compressed sensing. Group sparse representation improves the quality of reconstructed images by exploiting the non-local similarity of images; however, block-matching and dictionary learning in the image group construction process leads to a long reconstruction time and artifacts in the reconstructed images. To solve the above problems, a joint regularized image reconstruction model based on group sparse representation (GSR-JR) is proposed. A group sparse coefficients regularization term ensures the sparsity of the group coefficients and reduces the complexity of the model. The group sparse residual regularization term introduces the prior information of the image to improve the quality of the reconstructed image. The alternating direction multiplier method and iterative thresholding algorithm are applied to solve the optimization problem. Simulation experiments confirm that the optimized GSR-JR model is superior to other advanced image reconstruction models in reconstructed image quality and visual effects. When the sensing rate is 0.1, compared to the group sparse residual constraint with a nonlocal prior (GSRC-NLR) model, the gain of the peak signal-to-noise ratio (PSNR) and structural similarity (SSIM) is up to 4.86 dB and 0.1189, respectively.

Modelling of Block Feature-Content Based Efficient Image Compression System Using DWT and Machine Learning

“…The proposed method shows the values of SSIM are approximately identical to 1, and for the pirate image, the PSNR value is 46.37 dB. 38.0967 WDR coding [10] 39.1978 GSR-JR [11] 30.19 2D-SR [29] 39.65 BCS [30] 37.5160 MBCS [31] 39.9194 Proposed 45.3196 Peppers GSR-JR [11] 33.48 DWT-BP Compression [1] 35.5915 DWT [2] 32.4194 DWT-DFrFT [23] 28.9 2D-SR [29] 37.03 BCS [30] 32.0950 MBCS [31] 36.3212 Proposed 40.6447 Airplane GSR-JR [11] 32.70 DWT [2] 34.5345 DWT-DFrFT [23] 33.5 BCS [30] 31.8017 Proposed 38.2578 Lena DWT-BP Compression [1] 30.5113 DWT [2] 35.6323 BCS [30] 34.5962 CAIC [22] 31.5328 MBCS [31] 37.6716 Proposed 38.6308…”

“…It is proposed to employ a group sparse representation-based joint regularised image reconstruction model (GSR-JR) [11]. Group sparse coefficients are regularised to reduce model complexity and ensure sparsity.…”

“…1 𝑖𝑓 min ED 0 𝑖𝑓 max ED (10) Where euclidean distance is the distance between query instance vector 𝑎 and trained vector 𝑏. 𝐸𝐷 = √∑ (𝑎 𝑖 − 𝑏 𝑖 ) 2 𝑛 𝑖=1 (11) The K-NN algorithm makes use of the euclidean distance measure technique to find the minimum difference between training and testing features [19]. A query block is reconstructed based on the predicted label, which is the indexing code book.…”

“…This gives a comprehensive view concerning different image quality metrics. This is explained in Figure 8, which shows the trade-off between the PSNR and SSIM, which is in comparison to the current technique of group sparse representation (GSR-JR) [11] with the proposed method using different sample test images. The proposed method shows the values of SSIM are approximately identical to 1, and for the pirate image, the PSNR value is 46.37 dB.…”

ARET-IQA: An Aspect-Ratio-Embedded Transformer for Image Quality Assessment

“…However, various distortion types (e.g., blur and JPEG compression) or discordant elements (e.g., low light and monotonous color) may cause image quality degradation during the shooting and imaging process of camera devices. Consequently, image quality assessment (IQA) [1,2] that can automatically predict the technical and aesthetic quality of images is a fundamental task in the computational photography and computer vision communities, which is extremely valuable in optimizing many applications, such as image compression [3], image restoration [4], photo enhancement [5], image reconstruction [6], and image synthesis [7].…”