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

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.

Group feature selection for enhancing information gain in MRI reconstruction

Magnetic resonance imaging (MRI) has revolutionized the radiology. As a leading medical imaging modality, MRI not only visualizes the structures inside body, but also produces functional imaging. However, due to the slow imaging speed constrained by the MR physics, MRI cost is expensive, and patient may feel not comfortable in a scanner for a long time. Parallel MRI has accelerated the imaging speed through the sub-Nyquist sampling strategy and the missing data are interpolated by the multiple coil data acquired. Kernel learning has been used in the parallel MRI reconstruction to learn the interpolation weights and re-construct the undersampled data. However, noise and aliasing artifacts still exist in the reconstructed image and a large number of auto-calibration signal lines are needed. To further improve the kernel learning-based MRI reconstruction and accelerate the speed, this paper proposes a group feature selection strategy to improve the learning performance and enhance the reconstruction quality. An explicit kernel mapping is used for selecting a subset of features which contribute most to estimate the missing k-space data. The experimental results show that the learning behaviours can be better predicted and therefore the reconstructed image quality is improved.

3-D Image Analysis via Jacobi Moments with GPU-Accelerated Algorithms

This research has developed a parallel algorithm to compute 3-Dimensional Jacobi moments with high efficiency and accuracy. The algorithm was implemented in CUDA C. Our developing progress was in the order of Legendre moments, Gegenbauer moments, and Jacobi moments investigated on the 2-D image. Then, we extended research from 2-D to 3-D image. To verify the algorithm’s performance, we have implemented image reconstruction from higher orders up to 500 on testing image sized at 512×512×512. The experiment was deployed on Nvidia Tesla V100, which restrained computational time within 400 milliseconds, and the PSNR value of reconstructed image reached up to 53.6382.

CT Image Features under Reconstruction Algorithm in Analysis of the Effect of Probiotics Combined with Ursodeoxycholic Acid in Treatment of Intrahepatic Cholestasis of Pregnancy

This research was to explore the adoption value of computed tomography (CT) images based on adaptive statistical iterative reconstruction (ASIR) algorithm in the evaluation of probiotics combined with ursodeoxycholic acid in the treatment of intrahepatic cholestasis of pregnancy (ICP). A total of 82 patients with ICP were selected as the research subjects and they were randomly rolled into experimental group (380 mg probiotics enteric-soluble capsule twice a day, combined with 90 mg ursodeoxycholic acid soft capsule three times a day) and control group (90 mg ursodeoxycholic acid soft capsule three times a day), with 41 cases in each. The treatment course was four months. The ASIR algorithm was constructed and applied to the CT image analysis and diagnosis of ICP patients. The effects of filtering back projection (FBP) reconstruction and ASIR algorithm on CT image quality, denoising degree, and artifacts of ICP patients were compared. Moreover, blood indicator levels of ICP patients before and after treatment were assessed. The results showed that the SD values of liver and gallbladder (20.77 Hu and 27.58 Hu) in the reconstructed image of the ASIR algorithm were significantly lower than those of the FBP algorithm (40.58 Hu and 45.63 Hu) ( P < 0.05 ). The SNR values of the liver and gallbladder (3.68 and 2.05) of the reconstructed image were significantly higher than those of the FBP algorithm (1.91 and 1.19) ( P < 0.05 ). The overall image quality after ASIR reconstruction (3.92 points) was significantly better than that of the FBP algorithm (2.36 points), and the image noise score (3.21 points) reconstructed by the FBP algorithm was higher than that by the ASIR algorithm (1.83 points). The artifact score of FBP reconstructed image (4.47 points) was greatly higher than that of ASIR algorithm (2.26 points) ( P < 0.05 ). Before treatment, there was no remarkable difference in the indexes between the two groups of patients ( P > 0.05 ). After treatment, the γ-glutamyltransferase (γ-GT) and alkaline phosphatase (ALP) levels (327.55 U/L and 778.15 μmol/L) of the experimental group of ICP patients were higher than those of the control group (248.63 U/L and 668.43 μmol/L), with substantial difference between the two groups ( P < 0.05 ). The blood ammonia (BA) level (151.09 μmol/L) of the experimental group was lower than that of the control group (178.46 μmol/L), and the difference between the two groups was remarkable ( P < 0.05 ). To sum up, the CT image denoising degree based on ASIR algorithm was high, with few artifacts and good overall quality. Probiotics combined with ursodeoxycholic acid in the treatment of ICP can effectively improve the liver function and intestinal flora of patients, which was of great significance in the clinical diagnosis and treatment of the disease.

A New Hybrid Method for Caption and Scene Text Classification in Action Video Images

Achieving a better recognition rate for text in action video images is challenging due to multiple types of text with unpredictable actions in the background. In this paper, we propose a new method for the classification of caption (which is edited text) and scene text (text that is a part of the video) in video images. This work considers five action classes, namely, Yoga, Concert, Teleshopping, Craft, and Recipes, where it is expected that both types of text play a vital role in understanding the video content. The proposed method introduces a new fusion criterion based on Discrete Cosine Transform (DCT) and Fourier coefficients to obtain the reconstructed images for caption and scene text. The fusion criterion involves computing the variances for coefficients of corresponding pixels of DCT and Fourier images, and the same variances are considered as the respective weights. This step results in Reconstructed image-1. Inspired by the special property of Chebyshev-Harmonic-Fourier-Moments (CHFM) that has the ability to reconstruct a redundancy-free image, we explore CHFM for obtaining the Reconstructed image-2. The reconstructed images along with the input image are passed to a Deep Convolutional Neural Network (DCNN) for classification of caption/scene text. Experimental results on five action classes and a comparative study with the existing methods demonstrate that the proposed method is effective. In addition, the recognition results of the before and after the classification obtained from different methods show that the recognition performance improves significantly after classification, compared to before classification.

Anthropocene nostalgia for the 1980s in Netflix’s Stranger Things

Expressions of nostalgia for the 1980s in contemporary American culture are diverse. The most interesting of them go beyond a wistful longing for the past. A complex ‘nostalgia trip’ offered by Netflix’s Stranger Things serves as a notable case study of a distinctive type of this sentiment. Instead of yearning for the restoration of previous times, it plays with past aesthetics in a critically articulate manner, effectively demythologizing the depicted decade. I argue that this significant alteration of the traditional sentiment stems largely from the recent acknowledgment of the Anthropocene and its irreversibility. This article aims to examine the peculiar, self-aware, paradoxical nostalgia, which is coloured by the current, Anthropocene-induced fears for the environment and, thus, our future. The analysis of Stranger Things – its thematics, genre, visuals and the meticulously reconstructed image of the presented era – draws parallels to the techniques employed by the ‘novel nostalgia’: bitter, ironic depiction of the past and references to natural phenomena. The study thus investigates the show at the intersection of contemporary nostalgia for the 1980s and the cultural repercussions of the Anthropocene. In so doing, it will unravel the innovation in the programme’s discourse on the 1980s decade in American culture.

Holographic Near-Eye 3D Display Method Based on Large-Size Hologram

In this paper, we propose a holographic near-eye 3D display method based on large-size computer-generated hologram (CGH). The reconstructed image with a large viewing angle is obtained by using a time multiplexing and spatial tiling system. The large-size CGHs are generated and they record the information of the 3D object from different angles. The CGHs are reproduced at different moments. For a certain reconstructed moment, three spatial light modulators (SLMs) spatially spliced into a linear structure are used to load a single CGH. The diffraction boundary angle of the reconstructed light forming each image point is equal to the maximum diffraction angle of the SLM, so the viewing angle of the image generated by the CGH is enlarged. For different CGHs, the incident angle of reconstructed light is changed. Through time multiplexing, the reconstructed images of the CGHs are combined into a reconstructed image whose viewing angle is further enlarged. Due to the large viewing angle of the reconstructed image, the proposed method has unique advantages in near-eye display. The feasibility of the proposed method is proved by experimental results.

Performance Estimation of Intensity Accumulation Display by Computer-Generated Holograms

The image generated by binary computer-generated holograms (CGHs) always suffers from serious speckle noise. Thanks to the fast frame rate of the binary spatial light modulator, the speckle can be significantly suppressed by intensity accumulation, i.e., the sequential display of multiple CGHs of the same scene. If enough randomness is added to the CGHs, the speckle noise can be mostly averaged out. Intuitively, the quality of the reconstructed image should be proportional to the number of intensity accumulation. However, there is no simple method to predict the dependence of the average noise and accumulation number, and we can only know the results after finishing the full computation. In this paper, we propose an empirical formula of the average noise based on the speckle phenomenon in a laser projector. Using this model, we have confirmed that the randomness induced by random phase is equivalent to that induced by random down-sampling for the generation of binary CGHs. In addition, if the computational efficiency is a concern, the CGH calculated with iterations is not recommended for intensity accumulation display. Finally, there is an upper-quality limit of the reconstructed image by intensity accumulation. Thus, a strategy for efficient intensity accumulation is suggested.

Encryption Based Image Watermarking Algorithm in 2DWT-DCT Domains

This paper proposes an encryption-based image watermarking scheme using a combination of second-level discrete wavelet transform (2DWT) and discrete cosine transform (DCT) with an auto extraction feature. The 2DWT has been selected based on the analysis of the trade-off between imperceptibility of the watermark and embedding capacity at various levels of decomposition. DCT operation is applied to the selected area to gather the image coefficients into a single vector using a zig-zig operation. We have utilized the same random bit sequence as the watermark and seed for the embedding zone coefficient. The quality of the reconstructed image was measured according to bit correction rate, peak signal-to-noise ratio (PSNR), and similarity index. Experimental results demonstrated that the proposed scheme is highly robust under different types of image-processing attacks. Several image attacks, e.g., JPEG compression, filtering, noise addition, cropping, sharpening, and bit-plane removal, were examined on watermarked images, and the results of our proposed method outstripped existing methods, especially in terms of the bit correction ratio (100%), which is a measure of bit restoration. The results were also highly satisfactory in terms of the quality of the reconstructed image, which demonstrated high imperceptibility in terms of peak signal-to-noise ratio (PSNR ≥ 40 dB) and structural similarity (SSIM ≥ 0.9) under different image attacks.