Performance Of Super-resolution Reconstruction Research Articles

Image super resolution by double dictionary learning and its application to tool wear monitoring in micro milling

It is an effective means to improve the machining quality of product by monitoring the tool wear conditions in micro milling. However, due to the small size and high speed of the micro milling cutter, it is difficult to capture the clear image of tool wear by CCD camera. Therefore, a double dictionary sparse coding super-resolution (SR) method is proposed to improve the resolution of micro milling tool wear images in this work. Based on the traditional single dictionary, this work proposes the residual dictionary to reconstruct the residual image of the tool, achieving the double dictionary coefficient encoding reconstruction. In order to extract local features of low resolution (LR) tool wear images, the Gobor filter banks are designed, and then the PCA method is used to reduce dimension of extracted features. The dictionary learning based on a single high resolution (HR) image is realized, which reduces the dependence of super-resolution algorithm on large datasets. The effectiveness of Gabor filter and double dictionary for restoration tool wear image details is verified by experiments. Compared with proposed method, traditional interpolation method, sparse coding super resolution model [32] and super resolution convolutional neural network model [26], the PSNR mean value of those methods are 43.62, 42.19, 43.20, 43.21, respectively, which shows that the proposed method has excellent performance in super-resolution reconstruction of micro milling tool images for tool wear monitoring.

Lightweight image super-resolution reconstruction based on inverted residual attention network

In recent years, methods based on deep convolutional neural networks have made great progress in the field of image super-resolution reconstruction. However, mainstream approaches generally establish many network layers, leading to high computational costs and memory usage that are unsuitable for resource-limited edge devices. To alleviate this issue, a lightweight inverted residual attention network (IRAN) is proposed to obtain better super-resolution reconstruction performance with lower parameters and computation. The main structure of the IRAN consists of a series of inverted residual attention groups (IRAGs), which are mainly composed of several inverted residual attention blocks (IRABs). IRAB effectively reduces the network parameters and computation while extracting depth features by introducing the inverted residual structure and uses the simple channel attention mechanism to learn the important channel feature information. In addition, an enhanced spatial attention mechanism is introduced at the beginning and end of IRAG to further improve the reconstruction performance of the network. The experimental results show that compared with the mainstream lightweight networks, not only the peak signal-to-noise ratio and structural similarity of image quality metrics are better but also the parameters and the computational effort of the network are lower.

Remote sensing image super-resolution using multi-scale convolutional sparse coding network.

With the development of convolutional neural networks, impressive success has been achieved in remote sensing image super-resolution. However, the performance of super-resolution reconstruction is unsatisfactory due to the lack of details in remote sensing images when compared to natural images. Therefore, this paper presents a novel multiscale convolutional sparse coding network (MCSCN) to carry out the remote sensing images SR reconstruction with rich details. The MCSCN, which consists of a multiscale convolutional sparse coding module (MCSCM) with dictionary convolution units, can improve the extraction of high frequency features. We can obtain more plentiful feature information by combining multiple sizes of sparse features. Finally, a layer based on sub-pixel convolution that combines global and local features takes as the reconstruction block. The experimental results show that the MCSCN gains an advantage over several existing state-of-the-art methods in terms of peak signal-to-noise ratio and structural similarity.

Sparse Dictionary-Based Magnetic Resonance Superresolution Imaging with Joint Loss Function Learning.

Magnetic resonance image has important application value in disease diagnosis. Due to the particularity of its imaging mechanism, the resolution of hardware imaging needs to be improved by increasing radiation intensity and radiation time. Excess radiation can cause the body to overheat and, in severe cases, inactivate the protein. This problem is expected to be solved by the image superresolution method based on joint dictionary learning, which has good superresolution performance. In the process of dictionary learning, the loss function will directly affect the dictionary performance. The general method only uses the cascade error as the optimization function in dictionary training, and the method does not consider the individual reconstruction error of high- and low-resolution image dictionary. In order to solve the above problem, In this paper, the loss function of dictionary learning is optimized. While ensuring that the coefficients are sufficiently sparse, the high- and low-resolution dictionaries are trained separately to reduce the error generated by the joint high- and low-resolution dictionary block pair and increase the high-resolution reconstruction error. Experiments on neck and ankle MR images show that the proposed algorithm has better superresolution reconstruction performance on ×2 and ×4 compared with bicubic interpolation, nearest neighbor, and original dictionary learning algorithms.

Superresolution Reconstruction in Automatic Thai Sign Language Feature Extraction Using Adaptive Triangulation Interpolation

Superresolution is an image processing technique for improving image quality and enhancing low-resolution images. This paper presents a novel interpolation method for increasing superresolution reconstruction effectiveness using a triangulation interpolation algorithm for automatic Thai sign language feature extraction. This approach uses three neighboring pixels for estimation. The experiment compared the superresolution reconstruction performance using the triangulation interpolation algorithm and the nearest-neighbor, SRCNN, bilinear, bicubic, GPR, and NEDI methods. The super-resolution reconstruction using the triangulation-improved interpolation technique provided the best PSNR measurements of image quality between the original and superresolution-reconstructed images. The PSNR value of the sign language image was 40.608, improving performance by 13.15%. The Thai sign language gesture recognition using 2D convolutional neural networks showed that the designed model increased the gesture recognition effectiveness with an accuracy of 0.95 and loss of 0.14. Thus, this study provides state-of-the-art superresolution reconstruction for automatic Thai sign language gesture recognition.

Super-Resolution Reconstruction of Remote Sensing Images Using Generative Adversarial Network With Shallow Information Enhancement

The super-resolution reconstruction method based on deep learning can significantly improve the spatial super-resolution of remote sensing images. However, the current methods make insufficient use of the remote context information and channel information in shallow feature extraction, resulting in the limited effect of super-resolution reconstruction. This paper proposed a new super-resolution reconstruction model, SIEGAN, which uses generative adversarial network with shallow information enhancement to improve the effect of super-resolution reconstruction of remote sensing images. Similar to other generative adversarial models, SIEGAN is composed of generator and discriminator. But SIEGAN enhances the generator's ability to extract shallow information by using three different scale convolution operations. Specifically, a depth-wise convolution is used to extract the local context information of each band of the image. A depth-wise dilation convolution is used to capture the remote context information in the image. Finally, a 1×1 convolution is used to extract the correlation features between different channels in remote sensing images. In addition, SIEGAN uses U-Net network as its discriminator to provide detailed feedback per pixel to the generator, to improve the model's ability to identify image details. And the spectral-spatial total variation loss function is introduced to ensure the spectral-spatial reliability of the reconstructed images. The experimental results on Gaofen-1 data proved that compared with the state-of-the-art models, SIEGAN has achieved better super-resolution reconstruction performance. Furthermore, the reconstructed images by SIEGAN demonstrate better performance in land cover classification.

Bionic vision system and its application in license plate recognition

Conventional computer vision systems detect object after super-resolution (SR) or image reconstruction of the whole image, which is not an economical manner. By imitating the visual system of human beings, we proposed the bionic vision system (BVS), which is mainly composed by three parts: object detection by visual attention model, object-oriented SR reconstruction and object recognition by convolutional neural networks. The visual attention model contains both bottom-up and top-down cues. The bottom-up cues integrate low-level features by the feature integration theory. An Adaboost detector imitates the top-down cues. Sparse coding and compressed sensing reconstruction realize the object-oriented SR reconstruction. The BVS was validated on license plate recognition task. Both detection performance and SR reconstruction performance are tested. Besides of these, we also test the final recognition rate, all the experimental results are quite encouraging.

Super resolution reconstruction based on block matching and three‐dimensional filtering with sharpening

Super resolution (SR) reconstruction is often considered to be an inverse problem in the sense that unknown high resolution images are sought for giving low resolution images. Recent studies have shown that the sparsity regularisation used in compressed sensing (CS) reconstruction improves the performance of SR reconstruction. Furthermore, under the assumption that mutually similar regions exist within a natural image, non-local (NL) estimation produces accurate estimates for given degraded images. The incorporation of this NL estimation in SR reconstruction has been shown to yield better reconstructions. In this study, the authors propose the use of block matching and three-dimensional filtering with sharpening estimation as the regularisation constraint under the CS-based SR framework. This estimation collects similar blocks and adaptively filters them by the shrinkage of the transform coefficients. It recovers detailed structures while attenuating ringing artefacts. In addition, a sharpening technique used in the estimation also emphasises edges. As a result, the proposed SR algorithm searches for the solution that is similar to this enhanced estimate from among all feasible solutions. The experimental results demonstrate that the proposed method provides high-quality SR images, both numerically and subjectively.

A Collaborative Nonlocal-Means Super-resolution Algorithm Using Zernike Monments

Super-resolution (SR) with probabilistic motion estimation is a successful algorithm to circumvent the limitation of motion estimation upon conventional super-resolution methods. However, the algorithm can’t match similar patches with rotation or scale. This paper presents an efficient improved algorithm by introducing Zernike moments as representation of image invariant features into similarity measure. A collaborative strategy is proposed combining the moment based proximity and the bilateral proximity of nonlocal means (NL-means) algorithm for joint determination of weights. For the invariant property of Zernike moments, structure-similar pixels with rotation or scale can also be matched for computation of weights. Furthermore, the collaborative mechanism ensures higher accuracy of weights for a better estimation for each pixel of SR images. Experimental results indicate the proposed method is able to handle general video sequences with superior performance in SR reconstruction.

Performance of reconstruction-based super-resolution with regularization

From the perspective of linear algebra, the performance of super-resolution reconstruction (SR) depends on the conditioning of the linear system characterizing the degradation model. This is analyzed in the Fourier domain using the perturbation theory. By proposing a new SR error bound in terms of the point spread function (PSF), we reveal that the blur function dominates the condition number (CN) of degradation matrix, and the advantage of non-integer magnification factors (MFs) over the integer ones comes from sampling zero crossings of the DFT of the PSF. We also explore the effect of regularization by integrating it into the SR model, and investigate the influence of the optimal regularization parameter. A tighter error bound is derived given the optimal regularization parameter. Two curves of error bounds vs. MFs are presented, and verified by processing real images. It explains that with proper regularization, SR at the integer MFs is still valid.