Joint low-light enhancement and deblurring is a challenging imaging inverse problem that estimates clean images from photography corrupted by both low-light and blurring artifacts. To address this task, we propose FEL...
ISBN:
(纸本)9798350344868;9798350344851
Joint low-light enhancement and deblurring is a challenging imaging inverse problem that estimates clean images from photography corrupted by both low-light and blurring artifacts. To address this task, we propose FELI, a Fast and physically Enriched deep neural network for joint Low-light enhancement and image deblurring. In a departure from recently proposed end-to-end networks, FELI employs a learnable Decomposer during training based on Retinex theory that helps with low-light scene recovery. FELI's encoded features are further enriched by an input reconstruction task cognizant of the blur model leading to effective deblurring. We introduce a new customized contrastive regularization (CCR) term that pulls the restored clean image closer to the ground truth while pushing it far away from both the input and reconstructed input. Experiments performed on challenging synthetic and real-world datasets demonstrate that FELI outperforms state-of-the-art methods at a lower computational cost.
High Dynamic Range (HDR) images can be reconstructed from multiple Low Dynamic Range (LDR) images using existing deep neural network (DNN) techniques. Despite notable advancements, DNN-based methods still exhibit ghos...
详细信息
ISBN:
(纸本)9798350344868;9798350344851
High Dynamic Range (HDR) images can be reconstructed from multiple Low Dynamic Range (LDR) images using existing deep neural network (DNN) techniques. Despite notable advancements, DNN-based methods still exhibit ghosting artifacts when handling LDR images with saturation and significant motion. Recent Diffusion models (DMs) have been introduced in HDR imaging, showcasing promising performance, especially in achieving visually perceptible results. However, DMs typically require numerous inference iterations to recover the clean image from Gaussian noise, demanding substantial computational resources. Additionally, DM only learns a probability distribution of the added noise in each step but neglects image space constraints on HDR images, limiting distortion-based metrics. To tackle these challenges, we propose an efficient network that integrates DM modules into existing regression-based models, providing reliable content reconstruction for HDR while avoiding limitations in distortion-based metrics.
With the rise of downstream image tasks, the requirements for the quality of images obtained upstream are becoming higher and higher. In view of the many structural features of remote sensing images, we propose a nove...
详细信息
With the rise of downstream image tasks, the requirements for the quality of images obtained upstream are becoming higher and higher. In view of the many structural features of remote sensing images, we propose a novel deep neural network architecture for hyperspectral image fusion that integrates attention mechanisms and multi-layer perceptron blocks. The proposed network can capture long-range spatial dependencies between image elements, which is critical for capturing multi-scale features in remote sensing applications. The attention mechanisms selectively focus on important image features while disregarding redundant information, and the multi-layer perceptron blocks can capture multi-scale features by processingimage features at different scales. The experimental results demonstrate that the proposed network outperforms other state-of-the-art methods in terms of both objective evaluation metrics and visual quality. The proposed method achieves higher Peak signal to Noise Ratio and Spatial Consistency and Contrast values compared to other methods while preserving fine details and textures in the fused images. Overall, the proposed network provides an effective and efficient solution for hyperspectral image fusion that can contribute to the development of more accurate and reliable remote sensing applications.
The effectiveness of positioning techniques that utilize the receiver signal strength (RSS) is highly dependent on the instability of the received signal strength indicator (RSSI). Up to now, there is no strategy that...
详细信息
With the rise of Augmented Reality (AR) technology, which enhances the real world by overlaying computer-generated content, immersive experiences are being offered in education, entertainment, healthcare, ... Assessin...
详细信息
ISBN:
(纸本)9798350387261;9798350387254
With the rise of Augmented Reality (AR) technology, which enhances the real world by overlaying computer-generated content, immersive experiences are being offered in education, entertainment, healthcare, ... Assessing the quality of AR scenarios is crucial for understanding and improving user satisfaction and engagement. However, developing objective AR quality assessment methods is challenging due to the lack of data and the inherent complexity of technology, particularly in the presence of visual confusion. Existing convolution neural network-based approaches suffer from limited receptive fields and are not effective at capturing global information in visually confused AR scenarios. Additionally, to the best of our knowledge, exploring transformer capabilities for AR quality assessment is missing. Therefore, this study introduces transformAR, a lightweight transformer-based model for objective quality assessment in AR applications. This approach leverages pre-trained vision transformer-based encoders to capture image content information, computes distance vectors to quantify distortions, and employs cross-attention-based decoders to model perceptual quality features. The model also integrates adapted regularization techniques and label smoothing to mitigate overfitting. Experimental results demonstrate the effectiveness of transformAR, outperforming the few existing state-of-the-art methods.
High-resolution (HR) medical images can provide rich details, which are important for discovering subtle lesions to make diagnoses. Convolutional neural networks (CNNs) are widely used in this field, but struggle to m...
详细信息
High-resolution (HR) medical images can provide rich details, which are important for discovering subtle lesions to make diagnoses. Convolutional neural networks (CNNs) are widely used in this field, but struggle to model long-range dependencies. Although transformer-based methods have improved in this respect, this method requires large quantities of data. Unfortunately, large quantities of low -resolution (LR) and HR medical image pairs may not always be available. In addition, most medical image superresolution (SR) methods are deterministic, while the degradation in real scenarios is stochastic. To address these problems, we introduce a probabilistic degradation model that combines natural and medical images for training. This design alleviates the problem of insufficient medical image pairs and learns the degradation process of the natural scene. In addition, we propose a new medical image SR model that consists of CNNs and the Swin Transformer structure to excavate both local and global semantic features. Moreover, to reduce computational stress, the spherical locality -sensitive hashing (SLSH) module is employed in the nonlocal attention (NLA) mechanism to form the ENLA module. This design enables the proposed Sparse Swin Transformer (SSFormer) model to generate HR medical images without extensive training images. Experiments on diverse datasets (natural images and medical images) demonstrate that the proposed method is robust and effective, qualitatively and quantitatively outperforming other medical image SR methods. Code is available at https://***/codehxj/SSFormer.& COPY;2023 Elsevier Ltd. All rights reserved.
Automatic diagnosis of the type of arrhythmia of a patient achieved by ECG plays an important role in the prevention and treatment of cardiovascular diseases. In recent years, convolutional neural network (CNN) and re...
详细信息
Automatic diagnosis of the type of arrhythmia of a patient achieved by ECG plays an important role in the prevention and treatment of cardiovascular diseases. In recent years, convolutional neural network (CNN) and recurrent neural network (RNN) have been widely used in ECG diagnosis, however, using a simple convolutional network to capture the complex local changes in the signal is difficult. RNN is not effective enough in modeling the context of long-distance signals with dense time steps, and most of the methods are mostly modeling the lead space or the time domain individually, failing to combine the two features effectively. Therefore, we propose a network (MLATANet) based on convolution-transformer architecture with multi-scale lead attention and time domain attention. In the shallow layers of the network, parallel multi-scale convolution is used to extract features at different temporal resolutions. Small convolution kernels are used to capture local subtle features, while larger convolution kernels are used to obtain local coarse contour features. After convolution, the lead attention is used to automatically assign more weights to important lead channels based on the importance of different channel information. In the deep layers of the network, Transformer's multi-head self-attention is used to model the global temporal dependencies, enriching the feature expression in both temporal and spatial dimensions. In summary, first, spatial local features were captured through shallow multi-scale convolution and lead attention, then temporal global features were captured through deep Transformer multi-head self-attention, enabling the model to not only deeply explore the subtle aspects of the signal, but also analyze the signal from the overall trend, achieving an organic combination of local and global features. Experiments were conducted on the 2018 China Physiological signal Challenge (CPSC2018) dataset, 2021 PhysioNet/Computing in Cardiology Challenge (CinC202
Algorithms for multisignals detection using imageprocessing are investigated. Approaches based on digital imageprocessing, as well as on the use of neural networks and deep learning are considered. A comparative ana...
详细信息
Deep learning has become the mainstream method in the field of single image super-resolution (SISR), and the neural architecture search has been gradually applied to build SISR networks in a non-hand-crafted way. Howe...
详细信息
Deep learning has become the mainstream method in the field of single image super-resolution (SISR), and the neural architecture search has been gradually applied to build SISR networks in a non-hand-crafted way. However, the existing methods can only search the structure of models and the searching speed is slow. To solve this problem, a neural component search (NCS) method is proposed. When searching for SISR networks, the color space and the composition of loss functions during training are also parts of the search space. Under a specific computational constraint, the peak signal noise ratio (PSNR) or structural similarity (SSIM) can be used as the reward to search out an optimal super-resolution network. In addition, a super graph is designed with the idea of parameter sharing to sample adaptive residual dense networks (ARDNs), thus the NCS can complete the search of SISR networks at faster speed compared to existing methods. Experimental results indicate that ARDNs searched by the NCS is competitive with the hand-crafted state-of-the-art networks, and ARDNs achieve favorable performance against state-of-the-art methods with similar computational consumption.
Noise is one of the rare aspects of experimental work that crosses all boundaries. It is present from scientific fields like ultrafast optical signal detection to applied fields such as imageprocessing, or even in ou...
详细信息
Noise is one of the rare aspects of experimental work that crosses all boundaries. It is present from scientific fields like ultrafast optical signal detection to applied fields such as imageprocessing, or even in our day-to-day lives when we are simply trying to have a conversation in a loud room. In all these cases, incoherent, stochastic noise tends to drown a signal we aim to detect, and various techniques may need to be employed to improve the clarity of the waveform, which is characterized by the signal-to-noise ratio (SNR). Yet, considering the ubiquity of noise in scientific and technology fields, it may be surprising how few methods there exists for denoising a signal. Active amplification techniques alone cannot be employed for weak, noisy signals, since the SNR is inevitably degraded due to fundamental laws of physics, while bandpass filtering schemes necessarily lead to an attenuation of the signal. In this article, we review recent advances on the concept of passive amplification techniques based on the Talbot effect to enhance the noise properties of signals through coherent energy redistribution. We demonstrate the basic framework starting from pulse repetition rate multiplication with the Talbot effect. We then extend this theory to show the principle behind passive amplification of periodic waveforms, and then how this idea can be extended to arbitrary (generally, aperiodic) signals. methods for passive amplification of both the time-domain and the frequency-domain representations of the signal of interest are reviewed. While here we focus on the application of the technique for optical signals in the standard telecommunication band (near wavelengths of 1550 nm), the proposed denoising scheme relies on widely available wave manipulations, such that it may offer exciting opportunities for any kind of physical wave support, such as acoustics, plasmonics and other regimes of the electromagnetic spectrum, like microwaves or X-rays.
暂无评论