Dense Attention Research Articles

Residual Dense Attention Networks for COVID-19 Computed Tomography Images Super Resolution

Residual Dense Attention Networks for COVID-19 Computed Tomography Images Super Resolution

Efficient dense attention fusion network with channel correlation loss for road damage detection

AbstractRoad damage detection (RDD) is critical to society's safety and the efficient allocation of resources. Most road damage detection methods which directly adopt various object detection models face some significant challenges due to the characteristics of the RDD task. First, the damaged objects in the road images are highly diverse in scales and difficult to differentiate, making it more challenging than other tasks. Second, existing methods neglect the relationship between the feature distribution and model structure, which makes it difficult for optimization. To address these challenges, this study proposes an efficient dense attention fusion network with channel correlation loss for road damage detection. First, the K‐Means++ algorithm is applied for data preprocessing to optimize the initial cluster centers and improve the model detection accuracy. Second, a dense attention fusion module is proposed to learn spatial‐spectral attention to enhance multi‐scale fusion features and improve the ability of the model to detect damage areas at different scales. Third, the channel correlation loss is adopted in the class prediction process to maintain the separability of intra and inter‐class. The experimental results on the collected RDDA dataset and RDD2022 dataset show that the proposed method achieves state‐of‐the‐art performance.

LDANet: Automatic lung parenchyma segmentation from CT images

Automatic segmentation of the lung parenchyma from computed tomography (CT) images is helpful for the subsequent diagnosis and treatment of patients. In this paper, based on a deep learning algorithm, a lung dense attention network (LDANet) is proposed with two mechanisms: residual spatial attention (RSA) and gated channel attention (GCA). RSA is utilized to weight the spatial information of the lung parenchyma and suppress feature activation in irrelevant regions, while the weights of each channel are adaptively calibrated using GCA to implicitly predict potential key features. Then, a dual attention guidance module (DAGM) is designed to maximize the integration of the advantages of both mechanisms. In addition, LDANet introduces a lightweight dense block (LDB) that reuses feature information and a positioned transpose block (PTB) that realizes accurate positioning and gradually restores the image resolution until the predicted segmentation map is generated. Experiments are conducted on two public datasets, LIDC-IDRI and COVID-19 CT Segmentation, on which LDANet achieves Dice similarity coefficient values of 0.98430 and 0.98319, respectively, outperforming a state-of-the-art lung segmentation model. Additionally, the effectiveness of the main components of LDANet is demonstrated through ablation experiments.

Gated Region-Refine pose transformer for human pose estimation

Implementing the transformer for global fusion is a novel and efficient method for pose estimation. Although the computational complexity of modeling dense attention can be significantly reduced by pruning possible human tokens, the accuracy of pose estimation still suffers from the problem of high overlap of candidate regions and severe background noise. Moreover, the undifferentiated fusion of features from different views also leads to a sizeable effective information loss. To address these challenges, we propose a Gated Region-Refine Pose Transformer (GRRPT) for human pose estimation. The proposed GRRPT can obtain the general area of the human body from the coarse-grained tokens and then embed it into the fine-grained ones to extract more details of the joints. Experimental results on COCO demonstrate that performing the Multi-Resolution Attention mechanism learns more refined candidate regions and improves accuracy. Furthermore, we design a Fusion Gate module consisting of two gates to pixel-wise select valid information from the auxiliary views, which significantly alleviates information redundancy. Finally, we evaluate the effectiveness of our method on Human3.6M and our dataset FDU-Motion and achieve state-of-the-art performance.

Dense Attention Fusion Network for Object Counting in IoT System

Dense Attention Fusion Network for Object Counting in IoT System

Multi-scale dense selective network based on border modeling for lung nodule segmentation.

Accurate quantification of pulmonary nodules helps physicians to accurately diagnose and treat lung cancer. We try to improve the segmentation efficiency of irregular nodules while maintaining the segmentation accuracy of simple types of nodules. In this paper, we obtain the unique edge part of pulmonary nodules and process it as a single branch stream, i.e., border stream, to explicitly model the nodule edge information. We propose a multi-scale dense selective network based on border modeling (BorDenNet). Its overall framework consists of a dual-branch encoder-decoder, which achieves parallel processing of classical image stream and border stream. We design a dense attention module to facilitate a strongly coupled status of feature images to focus on key regions of pulmonary nodules. Then, during the process of model decoding, the multi-scale selective attention module is proposed to establish long-range correlation relationships between different scale features, which further achieves finer feature discrimination and spatial recovery. We introduce border context enhancement module to mutually fuse and enhance the edge-related voxel features contained in the image stream and border stream and finally achieve the accurate segmentation of pulmonary nodules. We evaluate the BorDenNet rigorously on the lung public dataset LIDC-IDRI. For the segmentation of the target nodules, the average Dice score is 92.78[Formula: see text], the average sensitivity is 91.37[Formula: see text], and the average Hausdorff distance is 3.06 mm. We further test on a private dataset from Shanxi Provincial People's Hospital, which verifies the excellent generalization of BorDenNet. Our BorDenNet relatively improves the segmentation efficiency for multi-type nodules such as adherent pulmonary nodules and ground-glass pulmonary nodules. Accurate segmentation of irregular pulmonary nodules can obtain important clinical parameters, which can be used as a guide for clinicians and improve clinical efficiency.

Arbitrary-scale super-resolution reconstruction of remote sensing images based on meta-learning and residual dense attention mechanism

Arbitrary-scale super-resolution reconstruction of remote sensing images based on meta-learning and residual dense attention mechanism

Multi-modal MR image super-resolution with residual dense attention network

Multi-modal MR image super-resolution with residual dense attention network

De-Aliasing and Accelerated Sparse Magnetic Resonance Image Reconstruction Using Fully Dense CNN with Attention Gates.

When sparsely sampled data are used to accelerate magnetic resonance imaging (MRI), conventional reconstruction approaches produce significant artifacts that obscure the content of the image. To remove aliasing artifacts, we propose an advanced convolutional neural network (CNN) called fully dense attention CNN (FDA-CNN). We updated the Unet model with the fully dense connectivity and attention mechanism for MRI reconstruction. The main benefit of FDA-CNN is that an attention gate in each decoder layer increases the learning process by focusing on the relevant image features and provides a better generalization of the network by reducing irrelevant activations. Moreover, densely interconnected convolutional layers reuse the feature maps and prevent the vanishing gradient problem. Additionally, we also implement a new, proficient under-sampling pattern in the phase direction that takes low and high frequencies from the k-space both randomly and non-randomly. The performance of FDA-CNN was evaluated quantitatively and qualitatively with three different sub-sampling masks and datasets. Compared with five current deep learning-based and two compressed sensing MRI reconstruction techniques, the proposed method performed better as it reconstructed smoother and brighter images. Furthermore, FDA-CNN improved the mean PSNR by 2 dB, SSIM by 0.35, and VIFP by 0.37 compared with Unet for the acceleration factor of 5.

Dense Attentive Feature Enhancement for Salient Object Detection

Attention mechanisms have been proven highly effective for salient object detection. Most previous works utilize attention as a self-gated module to reweigh the feature maps at different levels independently. However, they are limited to certain-level guidance and could not satisfy the need of both accurately detecting intact objects and maintaining their detailed boundaries. In this paper, we build dense attention upon features from multiple levels simultaneously and propose a novel Dense Attentive Feature Enhancement (DAFE) module for efficient feature enhancement in saliency detection. DAFE stacks several attentional units and densely connects attentive feature output from current unit to its all subsequent units. This allows feature maps at deep units to absorb attentive information from shallow units, thus more discriminative information can be efficiently selected at the final output. Note that DAFE is plug and play, which can be effortlessly inserted into any saliency or video saliency models for their performance improvements. We further instantiate a highly effective Dense Attentive Feature Enhancement Network (DAFE-Net) for accurate salient object detection. DAFE-Net constructs DAFE over the aggregation feature that contains both semantics and saliency details, the entire salient objects and their boundaries can be well retained through dense attentions. Extensive experiments demonstrate that the proposed DAFE module is highly effective, and the DAFE-Net performs favorably compared with state-of-the-art approaches.

Automatic segmentation of esophageal gross tumor volume in 18F-FDG PET/CT images via GloD-LoATUNet

Background and objectiveFor esophageal squamous cell carcinoma, radiotherapy is one of the primary treatments. During the planning before radiotherapy, the intractable task is to precisely delineate the esophageal gross tumor volume (GTV) on medical images. In current clinical practice, the manual delineation suffers from high intra- and inter-rater variability, while also exhausting the oncologists on a treadmill. There is an urgent demand for effective computer-aided automatic segmentation methods. To this end, we designed a novel deep network, dubbed as GloD-LoATUNet. MethodsGloD-LoATUNet follows the effective U-shape structure. On the contractile path, the global deformable dense attention transformer (GloDAT), local attention transformer (LoAT), and convolution blocks are integrated to model long-range dependencies and localized information. On the center bridge and the expanding path, convolution blocks are adopted to upsample the extracted representations for pixel-wise semantic prediction. Between the peer-to-peer counterparts, enhanced skip connections are built to compensate for the lost spatial information and dependencies. By exploiting complementary strengths of the GloDAT, LoAT, and convolution, GloD-LoATUNet has remarkable representation learning capabilities, performing well in the prediction of the small and variable esophageal GTV. ResultsThe proposed approach was validated in the clinical positron emission tomography/computed tomography (PET/CT) cohort. For 4 different data partitions, we report the Dice similarity coefficient (DSC), Hausdorff distance (HD), and Mean surface distance (MSD) as: 0.83±0.13, 4.88±9.16 mm, and 1.40±4.11 mm; 0.84±0.12, 6.89±12.04 mm, and 1.18±3.02 mm; 0.84±0.13, 3.89±7.64 mm, and 1.28±3.68 mm; 0.86±0.09, 3.71±4.79 mm, and 0.90±0.37 mm; respectively. The predicted contours present a desirable consistency with the ground truth. ConclusionsThe inspiring results confirm the accuracy and generalizability of the proposed model, demonstrating the potential for automatic segmentation of esophageal GTV in clinical practice.

Joint strong edge and multi-stream adaptive fusion network for non-uniform image deblurring

Non-uniform motion deblurring has been a challenging problem in the field of computer vision. Currently, deep learning-based deblurring methods have made promising achievements. In this paper, we propose a new joint strong edge and multi-stream adaptive fusion network to achieve non-uniform motion deblurring. The edge map and the blurred map are jointly used as network inputs and Edge Extraction Network (EEN) guides the Deblurring Network (DN) for image recovery and to complement the important edge information. The Multi-stream Adaptive Fusion Module (MAFM) adaptively fuses the edge information and features from the encoder and decoder to reduce feature redundancy to avoid image artifacts. Furthermore, the Dense Attention Feature Extraction Module (DAFEM) is designed to focus on the severely blurred regions of blurry images to obtain important recovery information. In addition, an edge loss function is added to measure the difference of edge features between the generated and clear images to further recover the edges of the deblurred images. Experiments show that our method outperforms currently public methods in terms of PSNR, SSIM and VIF, and generates images with less blur and sharper edges.

A Remote-Sensing Scene-Image Classification Method Based on Deep Multiple-Instance Learning with a Residual Dense Attention ConvNet

The spatial distribution of remote-sensing scene images is highly complex in character, so how to extract local key semantic information and discriminative features is the key to making it possible to classify accurately. However, most of the existing convolutional neural network (CNN) models tend to have global feature representations and lose the shallow features. In addition, when the network is too deep, gradient disappearance and overfitting tend to occur. To solve these problems, a lightweight, multi-instance CNN model for remote sensing scene classification is proposed in this paper: MILRDA. In the instance extraction and classifier part, more discriminative features are extracted by the constructed residual dense attention block (RDAB) while retaining shallow features. Then, the extracted features are transformed into instance-level vectors and the local information associated with bag-level labels is highlighted by the proposed channel-attention-based multi-instance pooling, while suppressing the weights of useless objects or backgrounds. Finally, the network is constrained by the cross-entropy loss function to output the final prediction results. The experimental results on four public datasets show that our proposed method can achieve comparable results to other state-of-the-art methods. Moreover, the visualization of feature maps shows that MILRDA can find more effective features.

A Novel RDA-Based Network to Conceal Image Data and Prevent Information Leakage

Image data play an important role in our daily lives, and scholars have recently leveraged deep learning to design steganography networks to conceal and protect image data. However, the complexity of computation and the running speed have been neglected in their model designs, and steganography security still has much room for improvement. For this purpose, this paper proposes an RDA-based network, which can achieve higher security with lower computation complexity and faster running speed. To improve the hidden image’s quality and ensure that the hidden image and cover image are as similar as possible, a residual dense attention (RDA) module was designed to extract significant information from the cover image, thus assisting in reconstructing the salient target of the hidden image. In addition, we propose an activation removal strategy (ARS) to avoid undermining the fidelity of low-level features and to preserve more of the raw information from the input cover image and the secret image, which significantly boosts the concealing and revealing performance. Furthermore, to enable comprehensive supervision for the concealing and revealing processes, a mixed loss function was designed, which effectively improved the hidden image’s visual quality and enhanced the imperceptibility of secret content. Extensive experiments were conducted to verify the effectiveness and superiority of the proposed approach.

Single image super-resolution based on multi-scale dense attention network

Single image super-resolution based on multi-scale dense attention network

ASSET

We present ASSET, a neural architecture for automatically modifying an input high-resolution image according to a user's edits on its semantic segmentation map. Our architecture is based on a transformer with a novel attention mechanism. Our key idea is to sparsify the transformer's attention matrix at high resolutions, guided by dense attention extracted at lower image resolutions. While previous attention mechanisms are computationally too expensive for handling high-resolution images or are overly constrained within specific image regions hampering long-range interactions, our novel attention mechanism is both computationally efficient and effective. Our sparsified attention mechanism is able to capture long-range interactions and context, leading to synthesizing interesting phenomena in scenes, such as reflections of landscapes onto water or fora consistent with the rest of the landscape, that were not possible to generate reliably with previous convnets and transformer approaches. We present qualitative and quantitative results, along with user studies, demonstrating the effectiveness of our method. Our code and dataset are available at our project page: https://github.com/DifanLiu/ASSET

Multi-Level Wavelet-Based Network Embedded with Edge Enhancement Information for Underwater Image Enhancement

As an image processing method, underwater image enhancement (UIE) plays an important role in the field of underwater resource detection and engineering research. Currently, the convolutional neural network (CNN)- and Transformer-based methods are the mainstream methods for UIE. However, CNNs usually use pooling to expand the receptive field, which may lead to information loss that is not conducive to feature extraction and analysis. At the same time, edge blurring can easily occur in enhanced images obtained by the existing methods. To address this issue, this paper proposes a framework that combines CNN and Transformer, employs the wavelet transform and inverse wavelet transform for encoding and decoding, and progressively embeds the edge information on the raw image in the encoding process. Specifically, first, features of the raw image and its edge detection image are extracted step by step using the convolution module and the residual dense attention module, respectively, to obtain mixed feature maps of different resolutions. Next, the residual structure Swin Transformer group is used to extract global features. Then, the resulting feature map and the encoder’s hybrid feature map are used for high-resolution feature map reconstruction by the decoder. The experimental results show that the proposed method can achieve an excellent effect in edge information protection and visual reconstruction of images. In addition, the effectiveness of each component of the proposed model is verified by ablation experiments.

Pyramidal dense attention networks for single image super‐resolution

Pyramidal dense attention networks for single image super‐resolution

Dense and shuffle attention U‐Net for automatic skin lesion segmentation

AbstractThe segmentation of skin lesions is a critical step in the detection of skin cancer. However, it is also difficult because of unclear boundaries, various size and shape of skin lesions, the appearance of interference objects and different backgrounds in dermoscopic images. This study presents proposed multiple attention‐based methods based on U‐net architecture. At first, dense attention gates (DAG) are proposed to obtain the focus area of the feature map by capturing information through extern feature maps. Then, to extend the capability of capturing feature relationships of the DAG module. We introduced The shuffle attention module which can fuse two self‐attention modules in a lightweight yet effective way. According to the experiments conducted on International Skin Imaging Collaboration (ISIC) 2017 and PH2 datasets, the proposed model can generate satisfactory segmentation results and great capability in practical computer‐aided diagnosis systems.

IMGAN: Infrared and visible image fusion using a novel intensity masking generative adversarial network

Fusion methods based on a deep convolutional neural network enable to catch visible details and infrared intensity. However, it is insufficient for most fusion models to extract the hierarchical features of visible images, leading to the loss of detailed information such as texture features in the final fused images. This paper proposes an unsupervised infrared and visible fusion model based on a novel intensity masking adversarial learning network (IMGAN). A designed deep generator with a residual dense attention module in the network can not only make full use of all characteristics of convolution layers but also achieve the more accurate extraction of visible detailed texture. Two discriminators are employed to balance the fused information, making it possible for the fused image to reserve much more information from source images. An infrared intensity masking loss is also proposed to reserve more detailed texture features in the weaker infrared area of the visible image. Furthermore, a large number of compared experiments with the infrared and visible benchmark datasets demonstrate the superiority of our IMGAN than the state-of-the-art methods in terms of both human eye visual perception and quantitative evaluation metrics. Our IMGAN can retain the infrared target information well and secure more detailed textures from the visible image.