Mutual Attention Learning Network for Domain Adaptive Person Re-identification

Older People Issues and Innovations in Care 3rd edition

Despite the great success of attention mechanisms on object recognition, scene recognition remains a challenging problem. The reason is that discriminative regions are not evident in a scene image. For example, a tree in an image can be a cue to recognize a scene, but the tree cannot be the only cue for recognizing the scene. That means several scene categories (e.g. mountain, marsh, and river) can contain a tree. Thus sometimes, overall regions, rather than specific regions, need to be considered for scene recognition. To solve the problem, we propose Spatial-Channel Transformer (SC-Transformer). The SC-Transformer is a simple yet effective module that uses a new attention mechanism by incorporating the importance between the spatial and the channel domain for a given scene image. If the given scene image should be considered only within some specific regions, SC-Transformer turns off the channel attention, and vice versa. Furthermore, the attention mechanism used in our proposed method is advanced from previous approaches. Previous spatial and channel attention mechanisms were designed in a sequential or parallel manner. These mechanisms eventually combine spatial and channel attention together, so spatial and channel attention may often interfere with each other. In contrast to the previous works, we present a new mechanism that simultaneously considers spatial and channel attentions. We validate our approach on a large-scale scene recognition dataset and outperform the previous state-of-the-art spatial-channel attention mechanism. Experimental results demonstrate the efficacy of our attention mechanism for scene recognition.

Recently, the attention mechanism combining spatial and channel information has been widely used in various deep convolutional neural networks (CNNs), proving its great potential in improving model performance. However, this usually uses 2D global pooling operations to compress spatial information or scaling methods to reduce the computational overhead in channel attention. These methods will result in severe information loss. Therefore, we propose a Spatial channel attention mechanism that captures cross-dimensional interaction, which does not involve dimensionality reduction and brings significant performance improvement with negligible computational overhead. The proposed attention mechanism can be seamlessly integrated into any convolutional neural network since it is a lightweight general module. Our method achieves a performance improvement of 2.08% on ResNet and 1.02% on MobileNetV2 in top-one error rate on the ImageNet dataset.

Small Target Spillage Detection Algorithm Based on Spatial Attention and Channel Attention Mechanism

It aimed to analyze the value of deep learning algorithm combined with magnetic resonance imaging (MRI) in the risk diagnosis and prognosis of endometrial cancer (EC). Based on the deep learning convolutional neural network (CNN) architecture residual network with 101 layers (ResNet-101), spatial attention and channel attention modules were introduced to optimize the model. A retrospective collection of MRI image data from 210 EC patients was used for model segmentation and reconstruction, with 140 cases as the test set and 70 cases as the validation set. The performance was compared with traditional ResNet-101 model, ResNet-101 model based on spatial attention mechanism (SA-ResNet-101), and ResNet-101 model based on channel attention mechanism (CA-ResNet-101), using accuracy (AC), precision (PR), recall (RE), and F1 score as evaluation metrics. Among the 70 cases in the validation set, there were 45 cases of low-risk EC and 25 cases of high-risk EC. Using ROC curve analysis, it was found that the area under the curve (AUC) for the diagnosis of high-risk EC of the proposed model in this article (0.918) was visibly larger as against traditional ResNet-101 (0.613), SA-ResNet-101 (0.760), and CA-ResNet-101 models (0.758). The AC, PR, RE, and F1 values of the proposed model for the diagnosis of EC risk were visibly higher (P < 0.05). In the validation set, postoperative recurrence occurred in 13 cases and did not occur in 57 cases. Using ROC curve analysis, it was found that the AUC for postoperative recurrence prediction of the patients by the proposed model (0.926) was visibly larger as against traditional ResNet-101 (0.620), SA-ResNet-101 (0.729), and CA-ResNet-101 models (0.767). The AC, PR, RE, and F1 values of the proposed model for postoperative recurrence prediction were visibly higher (P < 0.05). The proposed model in this article, assisted by MRI, presented superior performance in diagnosing high-risk EC patients, with higher sensitivity (Sen) and specificity (Spe), and also demonstrated excellent predictive AC in postoperative recurrence prediction.

Attention mechanism-based deep learning has played an important role in vision-based smoke detection in forest fire early warning systems. However, the lack of consideration of the specific characteristics of smoke may render existing attention mechanisms ineffective, e.g., in detecting small areas of smoke, particularly in the complex forest smoke scenes involving multiple points of ignition or varying scales. To enhance the accuracy and the interpretability of smoke detection, we propose a Spatial and Efficient Channel Attention mechanism, termed SECA, and integrate SECA into deep models to incorporate the characteristics of smoke diffusion. Technically, multi-kernel 1-dimensional (1D) convolution is utilized for multi-scale smoke-capturing, to replace single-kernel 2D or 3D convolution in existing methods. In implementation, our SECA mechanism can also be used as a common module and easily plugged into a backbone network. To accelerate our model, a DSConv-Haar Wavelet Downsampling technique called DHWD is also provided. Extensive experiments were conducted on public datasets and self-collected datasets. Compared to existing methods, our method can achieve a better or at least a comparable performance in smoke detection in terms of smoke detection accuracy, computational efficiency, and ease of use. For example, it surpasses baseline methods, demonstrating average improvements of 4.2% in mAP50 and of 3.7% in mAP50-95, respectively.

Object tracking is one of the important research topics in the direction of computer vision. And occlusion, deformation and scale change are the most challenging issues in the realm of tracking. In an effort to address the issue that the Siamese network-based object tracking algorithm’s resilience and success rate are both low and weak in complicated circumstances such as similar object interference and object occlusion, a Siamese network Object tracking algorithm combined with attention mechanism based on the existing Siamese network algorithm is proposed. Firstly, in order to extract the features of the template picture and the sample picture, the modified AlexNet convolutional neural network <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">[1]</sup> is utilized as the feature extraction network. Secondly, in order to increase the model’s precision and increase tracking’s rate of success, the module for the attention mechanism, which consists of the channel attention mechanism and the spatial attention mechanism, is introduced. The channel attention mechanism can focus on the information useful for tracking the Object, suppress unimportant background information, and the spatial attention mechanism is capable of picking up on the object’s abundant spatial information. Experiments show that the proposed Siamese network target tracking algorithm combined with attention mechanism achieves contestable results on the OTB100 test dataset, and the enhanced method outperforms the original approach in terms of performance.

Driver distraction is considered a main cause of road accidents, every year, thousands of people obtain serious injuries, and most of them lose their lives. In addition, a continuous increase can be found in road accidents due to driver’s distractions, such as talking, drinking, and using electronic devices, among others. Similarly, several researchers have developed different traditional deep learning techniques for the efficient detection of driver activity. However, the current studies need further improvement due to the higher number of false predictions in real time. To cope with these issues, it is significant to develop an effective technique which detects driver’s behavior in real time to prevent human lives and their property from being damaged. In this work, we develop a convolutional neural network (CNN)-based technique with the integration of a channel attention (CA) mechanism for efficient and effective detection of driver behavior. Moreover, we compared the proposed model with solo and integration flavors of various backbone models and CA such as VGG16, VGG16+CA, ResNet50, ResNet50+CA, Xception, Xception+CA, InceptionV3, InceptionV3+CA, and EfficientNetB0. Additionally, the proposed model obtained optimal performance in terms of evaluation metrics, for instance, accuracy, precision, recall, and F1-score using two well-known datasets such as AUC Distracted Driver (AUCD2) and State Farm Distracted Driver Detection (SFD3). The proposed model achieved 99.58% result in terms of accuracy using SFD3 while 98.97% accuracy on AUCD2 datasets.

Real-time semantic segmentation is an important field in computer vision. It is widely employed in real-world scenarios such as mobile devices and autonomous driving, requiring networks to achieve a trade-off between efficiency, performance, and model size. This paper proposes a lightweight network with multi-scale information interaction attention (MSIANet) to solve this issue. Specifically, we designed a multi-scale information interaction module (MSI) is the main component of the encoder and is used to densely encode contextual semantic features. Moreover, we designed the multi-channel attention fusion module (MAF) in the decoder part, thereby realizing multi-scale information fusion through channel attention mechanism and spatial attention mechanism. We verify our method through numerous experiments and prove that our network possesses fewer parameters and faster inference speed compared to most of the existing real-time semantic segmentation methods in multiple datasets.

Track-by-attention model, such as TrackFormer, have been one of the most popular deep models for multi-object tracking (MOT). However, one of their core limitations is image recognition accuracy and local feature attention due to the long sequence attention mechanism. To remedy this problem, MEFormer, a multi-scale feature enhanced model is proposed, in which both the channel attention and spatial attention mechanisms are leveraged for feature enhanced. MEFormer can effectively capture discriminative representation for MOT with attention and aggregation. Experiments on the MOT17 dataset indicate that the proposed MEFormer is able to achieve competitive performance on the task of multi-object tracking with 56.0% MOTA and 2400 ID switches.

MiRNA has distinct physiological functions at various cellular locations. However, few effective computational methods for predicting the subcellular location of miRNA exist, thereby leaving considerable room for improvement. Accordingly, our study proposes the MGFmiRNAloc simplified molecular input line entry system (SMILES) format as a new approach for predicting the subcellular localization of miRNA. Additionally, the graphical convolutional network (GCN) technique was employed to extract the atomic nodes and topological structure of a single base, thereby constructing RNA sequence molecular map features. Subsequently, the channel attention and spatial attention mechanisms (CBAM) were designed to mine deeper for more efficient information. Finally, the prediction module was used to detect the subcellular localization of miRNA. The 10-fold cross-validation and independent test set experiments demonstrate that MGFmiRNAloc outperforms the most sophisticated methods. The results indicate that the new atomic level feature representation proposed in this study could overcome the limitations of small samples and short miRNA sequences, accurately predict the subcellular localization of miRNAs, and be extended to the subcellular localization of other sequences.

With the continuous development of deep learning technology, object detection has received extensive attention across various computer fields as a fundamental task of computational vision. Effective detection of objects in remote sensing images is a key challenge, owing to their small size and low resolution. In this study, a remote sensing image detection (RSI-YOLO) approach based on the YOLOv5 target detection algorithm is proposed, which has been proven to be one of the most representative and effective algorithms for this task. The channel attention and spatial attention mechanisms are used to strengthen the features fused by the neural network. The multi-scale feature fusion structure of the original network based on a PANet structure is improved to a weighted bidirectional feature pyramid structure to achieve more efficient and richer feature fusion. In addition, a small object detection layer is added, and the loss function is modified to optimise the network model. The experimental results from four remote sensing image datasets, such as DOTA and NWPU-VHR 10, indicate that RSI-YOLO outperforms the original YOLO in terms of detection performance. The proposed RSI-YOLO algorithm demonstrated superior detection performance compared to other classical object detection algorithms, thus validating the effectiveness of the improvements introduced into the YOLOv5 algorithm.

ECG is a frequently preferred method to monitor heart health. The signals produced through ECG are analyzed by experts to detect cardiovascular diseases. However, the manual analysis process does not provide an effective evaluation process due to the limited time to experts, the difficulty in signal morphology, and the difference in experience between evaluators. However, early diagnosis is important. Because cardiovascular diseases are one of the leading causes of death worldwide. For this, the diagnostic process should be strengthened by computer-aided studies. In this study, training was performed with MobileNetV2, a lightweight architecture for the detection of heart rhythms in scope to S, V, F, Q, M and normal categories. Then, to improve the performance of this fully convolutional neural network, the MobileNetV2 architecture was restructured in 3 different formats. The Improved MobileNetv2 Version 3 architecture, restructured with scSE and ECA integrations, demonstrated maximum performance. Improved MobileNetv2 Version 3 architecture achieved approximately 9% accuracy improvement compared to the original MobileNetv2 architecture for the detection of normal and non-normal heart rhythms with the publicly available dataset provided from the Kaggle platform. This improvement indicates that the proposed combination is successful.

With the development and progress of medical imaging technology, the resolution of medical images has been increasing, and a variety of high-definition imaging modalities such as CT, PET-CT and MRI have emerged. U-Net network has the advantages of simple network topology and small training set data requirement; thus, the field of medical image segmentation uses it extensively. However, U-Net also has some problems, such as edge loss of segmentation results, long training time and single application scenario. For the medical image segmentation problem, this paper proposes a method that combines channel attention and spatial attention and uses an improved join strategy to join the network structure. To address the problem of insufficient data volume of medical images, this paper performs a data augmentation operation on the dataset with elastic deformation. In addition, we use a local-global training strategy to further improve the performance of training on medical images. When compared to the original U-Net, the Dice coefficient and IOU metrics are significantly better when utilizing the method suggested in this work. After extensive experiments, the strategy proposed in this study can achieve good outcomes when facing medical image segmentation problems and has great potential.

Wheat head identification and localization are crucial for field management and yield estimation. Traditional wheat head detection relies on labor-intensive manual counting. To address this issue, this study establishes a densely populated wheat head dataset and correspondingly proposes an efficient, rapid real-time model of CB-YOLO, which incorporates a Convolutional Block Attention Module (CBAM) mechanism (a sophisticated fusion of spatial attention mechanism and channel attention mechanism) so as to integrate channel and spatial features. Also, with an incorporated Bidirectional Feature Pyramid Network (BiFPN) to the neck, crowded wheat head detection was enhanced by establishing connections from the backbone to the bottom-up path. This enhancement surpasses the performance of the basic YOLOv5L model, achieving [Formula: see text] of 94.3% and [Formula: see text] of 53.9%. Among 10 classical detection models, CB-YOLO has a relatively small parameter size of 44[Formula: see text]Mb and high computing speed of 108[Formula: see text]G FLOPs. Additionally, ablation experiments demonstrate that compared to YOLOv5L without CBAM and BiFPN, the feature fusion module improves [Formula: see text] from 93.2% to 94.3%, indicating its effectiveness in enhancing detection performance. In conclusion, CB-YOLO exhibits outstanding identification performance, showing its feasibility in practical agricultural applications, and future work will focus on improving its localization ability.