Dual Attention Multiscale Network for Vessel Segmentation in Fundus Photography

The existing learning-based image feature point extraction methods ignore the contextual information in the image, and perform poorly in repetitive and weak texture scenes such as forest scenes, which brings great difficulties to the 3D reconstruction task of forest scenes. In order to solve the above problems, this paper proposes a new self-supervised image feature point extraction method with attention mechanism. First, the input image is processed by an encoder and encoded into a feature map containing multi-scale information. Then, the feature map is fed into the dual attention mechanism module, which includes a position attention module and a channel attention module, so as to obtain the image feature correlation in the spatial dimension and the channel dimension respectively, and obtain the enhanced feature map. Finally, the enhanced feature map is used to train the detector and descriptor respectively. Experiments show that our method has achieved some improvement on both the public dataset Hpatches and the virtual dataset Tartan Air, especially in the weakly textured forest scene.

Case instance segmentation of small farmland based on Mask R-CNN of feature pyramid network with double attention mechanism in high resolution satellite images

Dual attention per-pixel filter network for spatially varying image deblurring

Traditional semantic segmentation methods have problems such as poor multi-scale feature extraction ability, weak lightweight backbone network feature extraction ability, lack of effective fusion of context information, resulting in edge segmentation errors and feature discontinuity. In this paper, a novel semantic segmentation model based on multi-layer information fusion and dual convolutional attention mechanism is proposed. In this method, SegFormer network is used as the backbone network, and multi-scale features of encoder output are fused with overlapping features. The feature extraction subnetwork is optimized by constructing the object region enhancement module, and the intermediate feature map is refined adaptively in each convolutional block of the deep network, so as to strengthen the fine extraction of multi-dimensional feature information of complex images. Dual convolutional attention module is used to fusion high-level semantic information to avoid the loss of feature information caused by up-sampling operation and the influence of introducing noise, and refine the effect of target edge segmentation. At the same time, the feature pyramid grid is proposed to process the overlapping features, obtain the context information of different scales, and enhance the semantic expression of features. Finally, the features processed by the feature pyramid grid module are combined to improve the segmentation effect. The experimental results on the public data set show that the proposed method has better performance than the existing methods, and has better segmentation effect on the object edge in the scene.

The recognition of human activities using vision-based techniques has become a crucial research field in video analytics. Over the last decade, there have been numerous advancements in deep learning algorithms aimed at accurately detecting complex human actions in video streams. While these algorithms have demonstrated impressive performance in activity recognition, they often exhibit a bias towards either model performance or computational efficiency. This biased trade-off between robustness and efficiency poses challenges when addressing complex human activity recognition problems. To address this issue, this paper presents a computationally efficient yet robust approach, exploiting saliency-aware spatial and temporal features for human action recognition in videos. To achieve effective representation of human actions, we propose an efficient approach called the dual-attentional Residual 3D Convolutional Neural Network (DA-R3DCNN). Our proposed method utilizes a unified channel-spatial attention mechanism, allowing it to efficiently extract significant human-centric features from video frames. By combining dual channel-spatial attention layers with residual 3D convolution layers, the network becomes more discerning in capturing spatial receptive fields containing objects within the feature maps. To assess the effectiveness and robustness of our proposed method, we have conducted extensive experiments on four well-established benchmark datasets for human action recognition. The quantitative results obtained validate the efficiency of our method, showcasing significant improvements in accuracy of up to 11% as compared to state-of-the-art human action recognition methods. Additionally, our evaluation of inference time reveals that the proposed method achieves up to a 74× improvement in frames per second (FPS) compared to existing approaches, thus showing the suitability and effectiveness of the proposed DA-R3DCNN for real-time human activity recognition.

A largely ignored fact in spectral super-resolution (SSR) is that the subsistent mapping methods neglect the auxiliary prior of camera spectral sensitivity (CSS) and only pay attention to wider or deeper network framework design while ignoring to excavate the spatial and spectral dependencies among intermediate layers, hence constraining representational capability of convolutional neural networks (CNNs). To conquer these drawbacks, we propose a novel deep hybrid 2-D-3-D CNN based on dual second-order attention with CSS prior (HSACS), which can excavate sufficient spatial-spectral context information. Specifically, dual second-order attention embedded in the residual block for more powerful spatial-spectral feature representation and relation learning is composed of a brand new trainable 2-D second-order channel attention (SCA) or 3-D second-order band attention (SBA) and a structure tensor attention (STA). Concretely, the band and channel attention modules are developed to adaptively recalibrate the band-wise and interchannel features via employing second-order band or channel feature statistics for more discriminative representations. Besides, the STA is promoted to rebuild the significant high-frequency spatial details for enough spatial feature extraction. Moreover, the CSS is first employed as a superior prior to avoid its effect of SSR quality, on the strength of which the resolved RGB can be calculated naturally through the super-reconstructed hyperspectral image (HSI); then, the final loss consists of the discrepancies of RGB and the HSI as a finer constraint. Experimental results demonstrate the superiority and progressiveness of the presented approach in terms of quantitative metrics and visual effect over SOTA SSR methods.

In recent years, vein biometrics has gained significant attention due to its high security and privacy features. While deep neural networks have become the predominant classification approaches for their ability to automatically extract discriminative vein features, they still face certain drawbacks: 1) Existing transformer-based vein classifiers struggle to capture interactive information among different attention modules, limiting their feature representation capacity; 2) Current label enhancement methods, although effective in learning label distributions for classifier training, fail to model long-range relations between classes. To address these issues, we present ALE-IVT, an Attention Label Enhancement-based Interactive Vein Transformer for palm-vein recognition. First, to extract vein features, we propose an interactive vein transformer (IVT) consisting of three branches, namely spatial attention, channel attention, and convolutional module. In order to enhance performance, we integrate an interactive module that facilitates the sharing of discriminative features among the three branches. Second, we explore an attention-based label enhancement (ALE) approach to learn label distribution. ALE employs a self-attention mechanism to capture correlation between classes, enabling the inference of label distribution for classifier training. As self-attention can model long-range dependencies between classes, the resulting label distribution provides enhanced supervised information for training the vein classifier. Finally, we combine ALE with IVT to create ALE-IVT, trained in an end-to-end manner to boost the recognition accuracy of the IVT classifier. Our experiments on three public datasets demonstrate that our IVT model surpasses existing state-of-the-art vein classifiers. In addition, ALE outperforms current label enhancement approaches in term of recognition accuracy.

Road segmentation from high-resolution remote sensing images is a hot research topic in recent years. The roads in high-resolution remote sensing images are irregular with large scale variations. In addition, the scenes of such images are complicated with interference factors that have negative impacts on the segmentation results. On this account, we propose a dual attention network (DA-LinkNet), which combines the advantages of D-LinkNet and the dual attention mechanism. The network is built with D-LinkNet architecture and the Coordinate Attention block is added in its center part, which aims to enhance feature representation. To better combine features from different branches, we adopt the Attentional Feature Fusion block to replace the linear feature fusion operation of skip connection. The comparative experiments on DeepGlobe Road Extraction dataset demonstrate the effectiveness and superiority of the proposed method, which exceed state-of-the-art methods by at least 2 % for mean Intersection over Union (mIoU).

Monitoring the health of rolling mills has always been a concern for the steel industry. However, existing methods make it difficult to obtain fault features under the conditions of unbalanced data and serious disturbances during rolling. Therefore, we propose a fault diagnosis of the rolling mill based on the gramian angular difference field and the dual-attention residual network. Firstly, original one-dimensional vibration signals are converted into two-dimensional GADF images as the input of the network. Secondly, the dual attention mechanism of channel and space is introduced into the residual network to focus on the channel and spatial information of the input image for the re-integration of the extracted features. Finally, the model is trained to complete the recognition of faults. To validate the superiority of the proposed method, datasets with different data distributions were designed on the rolling mill test bench. Experimental comparisons have proven the superiority of GADF and the dual-attention mechanism. Moreover, when compared with other methods on several imbalanced datasets, the proposed method achieves an average accuracy that is at least 1.8% higher than that of other methods.

WiFi-based human activity recognition (HAR) leveraging channel state information (CSI) faces challenges due to multipath effects, leading to ambiguous feature representation. To address this, we propose AMCS-TCN, a model integrating dual attention mechanisms and a multi-channel, multi-scale temporal convolutional network (TCN). This model utilizes spatial diversity from MIMO systems to enhance feature representation. The dual attention module, comprising feature map fusion attention (FMFA) and Feature Channel Attention (FCA), adaptively focuses on discriminative information. Experimental results on a public dataset and real-world environments show an average accuracy of 98.85% and robust performance in complex settings. Our code and models are released at https://github.com/boyuanzhang759-rgb/AMCS-TCN-by-BoYuan for reproduction.

In computer-aided diagnosis systems, precise feature extraction from CT scans of colorectal cancer using deep learning is essential for effective prognosis. However, existing convolutional neural networks struggle to capture long-range dependencies and contextual information, resulting in incomplete CT feature extraction. To address this, the PEDRA-EFB0 architecture integrates patch embeddings and a dual residual attention mechanism for enhanced feature extraction and survival prediction in colorectal cancer CT scans. A patch embedding method processes CT scans into patches, creating positional features for global representation and guiding spatial attention computation. Additionally, a dual residual attention mechanism during the upsampling stage selectively combines local and global features, enhancing CT data utilization. Furthermore, this paper proposes a feature selection algorithm that combines autoencoders and entropy technology, encoding and compressing high-dimensional data to reduce redundant information and using entropy to assess the importance of features, thereby achieving precise feature selection. Experimental results indicate the PEDRA-EFB0 model outperforms traditional methods on colorectal cancer CT metrics, notably in C-index, BS, MCC, and AUC, enhancing survival prediction accuracy. Our code is freely available at https://github.com/smile0208z/PEDRA .

In recent years, attention mechanisms have shown great potential in various computer vision tasks. However, most existing methods focus on developing more complex attention modules for better performance, which inevitably increases the complexity of the model. To overcome performance and complexity tradeoffs, this paper proposes efficient spatial and channel attention (ESCA), a symmetric, comprehensive, and efficient attention module. By analyzing squeeze-and-excitation (SE), convolutional block attention module (CBAM), coordinate attention (CA), and efficient channel attention (ECA) modules, we abandon the dimension-reduction operation of SE module, verify the negative impact of global max pooling (GMP) on the model, and apply a local cross-channel interaction strategy without dimension reduction to learn attention. We not only care about the channel features of the image, we also care about the spatial location of the target on the image, and we take into account the effectiveness of channel attention, so we designed the symmetric ESCA module. The ESCA module is effective, as demonstrated by its application in the ResNet-50 classification benchmark. With 26.26 M parameters and 8.545 G FLOPs, it introduces a mere 0.14% increment in FLOPs while achieving over 6.33% improvement in Top-1 accuracy and exceeding 3.25% gain in Top-5 accuracy. We perform image classification and object detection tasks on ResNet, MobileNet, YOLO, and other architectures on popular datasets such as Mini ImageNet, CIFAR-10, and VOC 2007. Experiments show that ESCA can achieve great improvement in model accuracy at a very small cost, and it performs well among similar models.

Lung disease diagnosis remains a significant clinical challenge due to the similarity in radiological features across various conditions such as COPD, pneumonia, tuberculosis, COVID‐19, and lung cancer. Manual interpretation of chest CT scans is time‐consuming and subject to inter‐observer variability, particularly in resource‐limited settings. To address these challenges, this study proposes a novel deep learning framework Multiscale Convolutional Neural Networks with Attention Mechanism (MCNN‐AM) for automated classification of lung diseases into six categories, including normal lungs. The model leverages multiscale convolutional layers to extract both localized and global features, enabling better discrimination between diseases with overlapping characteristics. A dynamic attention mechanism, comprising both spatial and channel attention modules, is integrated to emphasize disease‐relevant regions and suppress background noise, enhancing the model's diagnostic focus. Additionally, depthwise separable convolutions are utilized to reduce computational complexity while preserving feature richness. The MCNN‐AM model is trained and evaluated on publicly available datasets, comprising 6000 training images and 1200 testing images equally distributed across all classes. The model achieves a classification accuracy of 96.84%, outperforming state‐of‐the‐art models such as ResNet50, DenseNet121, and InceptionV3 in terms of precision, recall, F1‐score, sensitivity, and specificity. Ablation studies further validate the critical role of the attention modules in achieving high performance. These results demonstrate the potential of MCNN‐AM as a reliable, scalable tool for computer‐aided diagnosis of lung diseases.

Object tracking can be considered as an object detection task. However, in general, an object detector cannot differentiate among intraclass instances and cannot fully exploit the spatiotemporal clues between successive frames. In order to solve these problems, in this paper, we proposed a novel object tracking framework based on object detection. We name the new framework TSTrack. Firstly, transformer is used in the target guidance module to extract the object’s features to guide the backbone network to complete the positioning task, which can solve the problem of distinguishing the instances within the classification. Secondly, self-attention mechanism is introduced into the model, which is mainly completed by spatial attention module (SAM) and channel attention module (CAM). By using the characteristics of self-attention mechanism, the attention matrix is dynamically self-learned in two dimensions of space and channel to obtain a more robust object representation. The proposed model is evaluated on the OTB-2013, OTB-50, OTB-2015 and VOT-2016 benchmarks, and the experimental results demonstrate the excellent tracking performance and robustness of our model.

Clinical management and accurate disease diagnosis are evolving from qualitative stage to the quantitative stage, particularly at the cellular level. However, the manual process of histopathological analysis is lab-intensive and time-consuming. Meanwhile, the accuracy is limited by the experience of the pathologist. Therefore, deep learning-empowered computer-aided diagnosis (CAD) is emerging as an important topic in digital pathology to streamline the standard process of automatic tissue analysis. Automated accurate nucleus segmentation can not only help pathologists make more accurate diagnosis, save time and labor, but also achieve consistent and efficient diagnosis results. However, nucleus segmentation is susceptible to staining variation, uneven nucleus intensity, background noises, and nucleus tissue differences in biopsy specimens. To solve these problems, we propose Deep Attention Integrated Networks (DAINets), which mainly built on self-attention based spatial attention module and channel attention module. In addition, we also introduce a feature fusion branch to fuse high-level representations with low-level features for multi-scale perception, and employ the mark-based watershed algorithm to refine the predicted segmentation maps. Furthermore, in the testing phase, we design Individual Color Normalization (ICN) to settle the dyeing variation problem in specimens. Quantitative evaluations on the multi-organ nucleus dataset indicate the priority of our automated nucleus segmentation framework.