Superresolution imaging using superoscillatory diffractive neural networks

In contrast to sophisticated means of visual su-per resolution (SR), not much work has been done in the tactile SR field. Existing tactile SR algorithms for taxel-based sensors mainly focus on enhancing the localization accuracy, and generally associate with a specific type of hardware, sometimes not applicable to generic taxel-based tactile sensors. Inspired by image SR, we investigate the tactile pattern SR in this paper, and present how to transform successful image SR schemes, e.g. Convolutional Neural Network (CNN) and Generative Adversarial Network (GAN) to serve the tactile SR. We propose two tactile SR models, i.e. TactileSRCNN and TactileSRGAN, and establish a new tactile pattern SR dataset for model learning. The ground truth of high resolution (HR) tactile patterns in the dataset is obtained via multi-sampling (i.e. overlapping reception) and registration of low resolution (LR) sensor. One key contribution of this research lies in achieving ×100 (from 3×4×4 to 40×40) times tactile pattern SR with a one-time tapping of 3-axis taxel-based sensor. Different from existing tactile SR algorithms which improves the localization accuracy of a single contact point, the proposed scheme can provide multi-point contact detection to robotic applications.

Abstract. As deep learning models advance, their use in Super Resolution (SR) tasks has become pivotal for enhancing remote sensing low-resolution (LR) satellite images. However, the decision making processes within these models remain opaque, especially in remote sensing applications where transparency is critical. This paper focuses on applying Explainable Artificial Intelligence (XAI) techniques, particularly Local Attribution Maps (LAMs), to analyze and interpret the internal behavior of both general purpose and remote sensing specific SR neural networks. General purpose models like Generative Adversarial Network for Super Resolution (SRGAN), Enhanced Deep Super-Resolution Network (EDSR), Efficient Super-Resolution Transformer (ESRT), and Hybrid-Attention Transformer (HAT), although highly effective in SR tasks, were originally designed for broader image enhancement challenges; in contrast, Hybrid-Scale Self-Similarity Exploitation Network (HSENet) and Multi-scale Enhanced Network (MEN) are tailored for the unique complexities of remote sensing, such as varied textures and intricate scene features. By leveraging LAMs, we highlight how different networks prioritize and process features such as edges, textures, and high frequency details to generate super resolved outputs. The comparative study between general purpose and remote sensing specific networks outlines each model’s strengths and weaknesses in managing the data present in remote sensing imagery. This approach addresses the previously unexplored application of LAMs in remote sensing for SR, contributing to the research in this field and providing deeper insights into the interpretability and transparency of widely used SR models. Furthermore, by drawing parallels between feature attribution in SR and classification tasks, we suggest new pathways for integrating semantic information to refine model transparency and performance in remote sensing applications.

The resolution and noise levels of input images directly affect the three-dimensional (3D) structure-from-motion (SfM) reconstruction performance. Conventional super-resolution (SR) methods focus too little on denoising, and latent image noise becomes worse when resolution is improved. This study proposes two SR denoising training algorithms to simultaneously improve resolution and noise: add-noise-before-downsampling and downsample-before-adding-noise. These portable methods preprocess low-resolution training images using real-world noise samples instead of altering the basic neural network. Hence, they concurrently improve resolution while reducing noise for an overall cleaner SfM performance. We applied these methods to the existing SR network: super-resolution convolutional neural network, enhanced deep residual super-resolution, residual channel attention network, and efficient super-resolution transformer, comparing their performances with those of conventional methods. Impressive peak signal-to-noise and structural similarity improvements of 0.12 dB and 0.56 were achieved on the noisy images of Smartphone Image Denoising Dataset, respectively, without altering the network structure. The proposed methods caused a very small loss (<0.01 dB) on clean images. Moreover, using the proposed SR algorithm makes the 3D SfM reconstruction more complete. Upon applying the methods to non-preprocessed and conventionally preprocessed models, the mean projection error was reduced by a maximum of 27% and 4%, respectively, and the number of 3D densified points was improved by 310% and 7%, respectively.

Enhancement in Skin Cancer Detection using Image Super Resolution and Convolutional Neural Network

Super-resolution using deep neural networks typically relies on highly curated training sets that are often unavailable in clinical deployment scenarios. Using loss functions that assume Gaussian-distributed residuals makes the learning sensitive to corruptions in clinical training sets. We propose novel loss functions that are robust to corruptions in training sets by modeling heavy-tailed non-Gaussian distributions on the residuals. We propose a loss based on an autoencoder-based manifold-distance between the super-resolved and high-resolution images, to reproduce realistic textural content in super-resolved images. We propose to learn to super-resolve images to match human perceptions of structure, luminance, and contrast. Results on a large clinical dataset shows the advantages of each of our contributions, where our framework improves over the state of the art. © 2019 IEEE.

As an important display mode of underwater environments, the sonar image has limitations on the resolution, which often leads to problems with low resolution of underwater objects. Therefore, the image super-resolution algorithm is needed to transform the images from low-resolution to high-resolution. It can improve the visual effect and contribute to subsequent processing such as 3D reconstruction and object recognition. This paper proposes a method for sonar image super-resolution based on generative adversarial networks (GAN). By comparing the super-resolution effects of various interpolation and convolutional neural network algorithms on sonar images, a Residual-in-Residual Dense Block network is employed as the generator of GAN since it has the low distortion and high perceptual quality. Because the sonar image training set does not have enough data, the generator utilizes the transfer learning on the sonar images to produce an optimized network model which is more suitable for super-resolution of sonar image. The VGG19 network is employed as the discriminator. In addition, the perceptual loss is introduced into the loss function of $$\hbox {S}^2\hbox {RGAN}$$ to further improve the perceptual quality of super-resolution images. The experimental results indicate that the proposed $$\hbox {S}^2\hbox {RGAN}$$ shows excellent performance. The generated super-resolution images of $$\hbox {S}^2\hbox {RGAN}$$ have the remarkable advantages of both lower distortion and higher perceptual quality comparing with other methods. Because $$\hbox {S}^2\hbox {RGAN}$$ focuses more on the reality and overall visual effect of super-resolution sonar images, it is suitable for various underwater situations.

Image super-resolution (SR) techniques can improve the spatial resolution of remote sensing images to provide more feature details and information, which is important for a wide range of remote sensing applications, including land use/cover classification (LUCC). Convolutional neural networks (CNNs) have achieved impressive results in the field of image SR, but the inherent localization of convolution limits the performance of CNN-based SR models. Therefore, we propose a new method, namely, the dilated Transformer generative adversarial network (DTGAN) for the SR of multispectral remote sensing images. DTGAN combines the local focus of CNNs with the global perspective of Transformers to better capture both local and global features in remote sensing images. We introduce dilated convolutions into the self-attention computation of Transformers to control the network’s focus on different scales of image features. This enhancement improves the network’s ability to reconstruct details at various scales in the images. SR imagery provides richer surface information and reduces ambiguity for the LUCC task, thereby enhancing the accuracy of LUCC. Our work comprises two main stages: remote sensing image SR and LUCC. In the SR stage, we conducted comprehensive experiments on Landsat-8 (L8) and Sentinel-2 (S2) remote sensing datasets. The results indicate that DTGAN generates super-resolution (SR) images with minimal computation. Additionally, it outperforms other methods in terms of the spectral angle mapper (SAM) and learned perceptual image patch similarity (LPIPS) metrics, as well as visual quality. In the LUCC stage, DTGAN was used to generate SR images of areas outside the training samples, and then the SR imagery was used in the LUCC task. The results indicated a significant improvement in the accuracy of LUCC based on SR imagery compared to low-resolution (LR) LUCC maps. Specifically, there were enhancements of 0.130 in precision, 0.178 in recall, and 0.157 in the F1-score.

Super resolution is the process of reconstructing a high-resolution (HR) image from a low-resolution (LR) one. Super-resolution technology enables high-resolution video streaming, image zoom-in, and far object recognition. Fig. 4.7.1 shows such an application scenario. The details of the videos/images can be reconstructed and projected to a higher-resolution screen, thereby providing a better visual experience. A hardware accelerator is needed to speed up the super-resolution process to support real-time high-resolution video streaming. Conventionally, dictionary-based approaches, such as ANR/GR [1] and A+ [2], convert the LR image into the HR one from learned mapping functions. Neural network (NN)-based algorithms generate better-quality super-resolution images by extracting features from training [3]. However, the complexity of the dictionary-based and the NN-based algorithms is excessively high, making them unsuitable for high-speed applications [4]. A rapid and accurate image super resolution (RAISR) algorithm [4] is proposed to achieve comparable quality with a much faster processing speed when compared to the previous solutions. It employs pre-learned filters to enhance the image quality based on bicubic interpolation. A pre-learned filter (also known as kernel) is selected by a hash function to address the structure-related details.

This paper presents a new quality assessment parameter for the evaluation of deep learning based super resolution techniques applied on thermal images. Three widely used deep learning-based models namely Super-Resolution Convolutional Neural Network (SRCNN), Thermal Enhancement Network (TEN) and Very Deep Super Resolution (VDSR) have been implemented for achieving super resolution of thermal images. Peak Signal to Noise Ratio (PSNR) and Structural Similarity Index (SSIM) have been the most widely used conventional evaluation metrics for performance measurement of super resolution algorithms. Since these parameters require a reference image for the evaluation of the resultant images, we propose a new quality assessment parameter based on strength of the edges. Edge detection of the super resoluted image is performed utilizing Canny Edge Detection method and the entropy of the edge detection image is computed to provide a new parameter, Edge Detection Entropy Score (EDES). For the comparison and validation of the proposed image quality assessment techniques, Mean Opinion Score (MOS) of the target images have been obtained to be used as a benchmark. The obtained results indicate that the proposed EDES of the super resoluted images has high degree of correlation with the MOS as well as PSNR and SSIM.

Convolutional neural network (CNN)-based single image super-resolution (SR) methods have achieved superior performance on some discrete-scaling factors, including 2, 3, and 4. However, the scaling factors for SR should be continuous and not discrete in practical applications. Previous CNN-based SR models usually yield poor results for non-integer-scaling factors and are sometimes even worse than results derived from the conventional bicubic method. To extend CNN-based SR models to continuous scale, this paper proposes a multiple-scaling-based SR (MSSR) method that combines an integer-scaling-factor SR and once or twice non-integer-scaling-factor SR without retraining networks. For a non-integer-scaling factor, the MSSR method first computes an optimal integer-scaling factor according to the data similarity and choose the corresponding pre-trained model for the next stage. Then, an existing CNN-based model is used to perform the integer-scaling-factor SR. Finally, the output is scaled to the target size. The proposed MSSR method can extend a variety of existing CNN-based SR models from discrete to continuous-scaling factors. Experimental results with six CNN-based SR models demonstrated that the MSSR method could effectively improve the performance of existing CNN-based SR models for continuous-scaling-factor SR without retraining networks. Furthermore, the comparison with a magnification-arbitrary method, called Meta-SR, shows that the proposed MSSR method usually outperforms Meta-SR for scaling factors greater than or equal to 2.

Image super-resolution (SR) is widely used on mobile devices to enhance user experience. However, neural networks used for SR are computationally expensive, posing challenges for mobile devices with limited computing power. A viable solution is to use heterogeneous processors on mobile devices, especially the specialized hardware AI accelerators, for SR computations, but the reduced arithmetic precision on AI accelerators can lead to degraded perceptual quality in upscaled images. To address this limitation, in this paper we present SR For Your Eyes (FYE-SR), a novel image SR technique that enhances the perceptual quality of upscaled images when using heterogeneous processors for SR computations. FYE-SR strategically splits the SR model and dispatches different layers to heterogeneous processors, to meet the time constraint of SR computations while minimizing the impact of AI accelerators on image quality. Experiment results show that FYE-SR outperforms the best baselines, improving perceptual image quality by up to 2×, or reducing SR computing latency by up to 5.6× with on-par image quality.

The demand for ultra-high-resolution imaging in mobile phone photography is continuously increasing. However, the image resolution of mobile devices is typically constrained by the size of the CMOS sensor. Although deep learning-based super-resolution (SR) techniques have the potential to overcome this limitation, existing SR neural network models require large computational resources, making them unsuitable for real-time SR imaging on current mobile devices. Additionally, cloud-based SR systems pose privacy leakage risks. In this paper, we propose M3Cam, an innovative and lightweight SR imaging system for mobile phones. M3Cam can ensure high-quality 16× SR image (4× in both height and width) visualization with almost negligible latency. In detail, we utilize an optical image stabilization (OIS) module for lens control and introduce a new modality of data, namely gyroscope readings, to achieve high-precision and compact optical flow estimation modules. Building upon this concept, we design a multi-frame-based SR model utilizing the Swin Transformer. Our proposed system can generate a 16× SR image from four captured low-resolution images in real-time, with low computational load, low inference latency, and minimal reliance on runtime RAM. Through extensive experiments, we demonstrate that our proposed multi-modal optical flow model significantly enhances pixel alignment accuracy between multiple frames and delivers outstanding 16× SR imaging results under various shooting scenarios. Code and dataset are available at: https://github.com/liangjindeamo-yuer/M3CAM

Super-resolution (SR) can be defined as a method of reconstructing high resolution(HR) image by increasing spatial resolution of one or more low resolution(LR) image counterparts. SR is widely applied in areas like satellite imaging, drones, surveillance and medical imaging etc. SR techniques can be broadly classified as external database driven or internal database driven approaches. The first approach is time consuming during training phase, in learning LR-HR patch relationship but computes fast during testing phase. The second approach directly constructs the super resolved image from existing LR image without the need of any training procedures but found to be computationally intensive during test time. This paper focuses on reducing the computation time of transformed self exemplar based internal database driven technique [1] that uses 3D scene geometry and internal image characteristics for super resolution by exploiting its highly parallelizable nature. On parallelizing the original algorithm we achieved a speedup of 6 on multi-core, upto 10 on GPU, around 14 on hybrid platform of multi-core and GPU. Using our hybrid implementation we can achieve 32x super resolution factor. We compare our results with SRCNN[2] (Super Resolution Convolution Neural Network) an external database SR technique which has a very good performance among the current start-of-the-art SR methods.

Adversarially trained deep neural networks have significantly improved performance of single image super resolution, by hallucinating photorealistic local textures, thereby greatly reducing the perception difference between a real high resolution image and its super resolved (SR) counterpart. However, application to medical imaging requires preservation of diagnostically relevant features while refraining from introducing any diagnostically confusing artifacts. We propose using a deep convolutional super resolution network (SRNet) trained for (i) minimising reconstruction loss between the real and SR images, and (ii) maximally confusing learned relativistic visual Turing test (rVTT) networks to discriminate between (a) pair of real and SR images (T1) and (b) pair of patches in real and SR selected from region of interest (T2). The adversarial loss of T1 and T2 while backpropagated through SRNet helps it learn to reconstruct pathorealism in the regions of interest such as white blood cells (WBC) in peripheral blood smears or epithelial cells in histopathology of cancerous biopsy tissues, which are experimentally demonstrated here. Experiments performed for measuring signal distortion loss using peak signal to noise ratio (pSNR) and structural similarity (SSIM) with variation of SR scale factors, impact of rVTT adversarial losses, and impact on reporting using SR on a commercially available artificial intelligence (AI) digital pathology system substantiate our claims.

ABSTRACTDue to the instantaneous field-of-view (IFOV) of the sensor and diversity of land cover types, some pixels, usually named mixed pixels, contain more than one land cover type. Soft classification can predict the portion of each land cover type in mixed pixels in the absence of spatial distribution. The spatial distribution information in mixed pixels can be solved by super resolution mapping (SRM). Typically, SRM involves two steps: soft class value estimation, which is similar to the image super resolution of image restoration, and land cover allocation. A new SRM approach utilizes a deep image prior (DIP) strategy combined with a super resolution convolutional neural network (SRCNN) to estimate fine resolution fraction images for each land cover type; then, a simple and efficient classifier is used to allocate subpixel land cover types under the constraint of the generated fine fraction images. The proposed approach can use prior information of input images to update network parameters and no longer require training data. Experiments on three different cases demonstrate that the subpixel classification accuracy of the proposed DIP-based SRM approach is significantly better than the three conventional SRM approaches and a transfer learning-based neural network SRM approach. In addition, the DIP-SRM approach performs very robustly about small-area objects within multiple land cover types and significantly reduces soft classification uncertainty. The results of this paper provide an extension for utilizing SRCNN to address SRM issues in hyperspectral images.