Spatial and Channel Dimensions Attention Feature Transfer for Better Convolutional Neural Networks

Tang, Jialiang; Liu, Mingjin; Jiang, Ning; Yu, Wenxin; Yang, Changzheng

doi:10.1109/iscas51556.2021.9401381

Cited by 4 publications

(3 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recent studies have shown that optimizing spatial dimension information can improve the performance of the network [43]. The SE attention module added allows the network to calibrate the importance of the channel adaptively.…”

Section: Extended Se Attention Modulementioning

confidence: 99%

MR-YOLO: An Improved YOLOv5 Network for Detecting Magnetic Ring Surface Defects

Lang

Ren

Wan

et al. 2022

Sensors

View full text Add to dashboard Cite

Magnetic rings are widely used in automotive, home appliances, and consumer electronics. Due to the materials used, processing techniques, and other factors, there will be top cracks, internal cracks, adhesion, and other defects on individual magnetic rings during the manufacturing process. To find such defects, the most sophisticated YOLOv5 target identification algorithm is frequently utilized. However, it has problems such as high computation, sluggish detection, and a large model size. This work suggests an enhanced lightweight YOLOv5 (MR-YOLO) approach for the identification of magnetic ring surface defects to address these issues. To decrease the floating-point operation (FLOP) in the feature channel fusion process and enhance the performance of feature expression, the YOLOv5 neck network was added to the Mobilenetv3 module. To improve the robustness of the algorithm, a Mosaic data enhancement technique was applied. Moreover, in order to increase the network’s interest in minor defects, the SE attention module is inserted into the backbone network to replace the SPPF module with substantially more calculations. Finally, to further increase the new network’s accuracy and training speed, we substituted the original CIoU-Ioss for SIoU-Loss. According to the test, the FLOP and Params of the modified network model decreased by 59.4% and 47.9%, respectively; the reasoning speed increased by 16.6%, the model’s size decreased by 48.1%, and the mAP only lost by 0.3%. The effectiveness and superiority of this method are proved by an analysis and comparison of examples.

show abstract

Section: Extended Se Attention Modulementioning

confidence: 99%

MR-YOLO: An Improved YOLOv5 Network for Detecting Magnetic Ring Surface Defects

Lang

Ren

Wan

et al. 2022

Sensors

View full text Add to dashboard Cite

show abstract

“…It has been proved that correlation between channels and spatial dimensions in the feature map can be completely decoupled [ 42 ]. Thus, separating ordinary convolutions can significantly reduce parameters and training time.…”

Section: Methodsmentioning

confidence: 99%

Evaluation of Application Effect of Self-Made Compression Cold Therapy in Postoperative Rehabilitation of Patients with Orthopedic Dyskinesia

Wan

Zhao

et al. 2022

Computational and Mathematical Methods in Medicine

View full text Add to dashboard Cite

With the accelerated aging of the population, orthopedic injuries have become more collective. Among them, the incidence of ankle fractures remains high. Surgery is an effective way to treat ankle fractures by utilizing special surgical site, complex anatomical structure, and specific surgical methods. With surgical approach, it is easy for basis postoperative blood loss, pain, swelling, and other problems. After surgery, most patients suffer from symptoms of fear, increased pain sensitivity, and excessive irrational concerns about physical movement or activity. Compression cold therapy combines cold therapy with air pressure therapy to ease local exudation, constrict blood vessels, improve circulation, relieve pain, and control inflammation through the effects of low temperature and pressure. Application during the rehabilitation period can prevent joint swelling, reduce muscle soreness, and promote the functional recovery of limbs, which provides an effective guarantee for postoperative rehabilitation of patients with orthopedic dyskinesia. Based on this, it is very important to evaluate the application and effect of self-made compression cold therapy in postoperative rehabilitation of patients with orthopedic dyskinesia. This work proposes a one-dimensional deep convolutional neural network-based method; DenseNet for analyzing the rehabilitation effect of patients with orthopedic dyskinesia after ankle fracture surgery. The approach is to evaluate the rehabilitation effect of self-made compression cold therapy from the perspectives of feature reuse, attention mechanism, and feature decoupling. Experiments on the dataset show that the proposed neural network has better efficacy evaluation performance. The proposed systematic assessment based on the emerging deep learning network has great significance in healthcare domain, particularly in assessing applicability, side effects, and noninvasiveness of treatment methods.

show abstract

“…It selects the key area of the image and gears the focus of the model. CBAM is a feed-forward model whereby the starfish images are analysed from two dimensions namely channel and spatial [34]. The spatial attention is used to search the area where the attention needs to be focussed in the starfish feature map.…”

Section: Proposed Optimised Cnn Modelsmentioning

confidence: 99%

Identification of Crown of Thorns Starfish (COTS) using Convolutional Neural Network (CNN) and attention model

et al. 2023

View full text Add to dashboard Cite

Coral reefs play important roles in the marine ecosystem, from providing shelter to aquatic lives to being a source of income to others. However, they are in danger from outbreaks of species like the Crown of Thorns Starfish (COTS) and the widespread coral bleaching from rising sea temperatures. The identification of COTS for detecting outbreaks is a challenging task and is often done through snorkelling and diving activities with limited range, where strong currents result in poor image capture, damage of capturing equipment, and are of high risks. This paper proposes a novel approach for the automatic detection of COTS based Convolutional Neural Network (CNN) with an enhanced attention module. Different pre-trained CNN models, namely, VGG19 and MobileNetV2 have been applied to our dataset with the aim of detecting and classifying COTS using transfer learning. The architecture of the pre-trained models was optimised using ADAM optimisers and an accuracy of 87.1% was achieved for VGG19 and 80.2% for the MobileNetV2. The attention model was developed and added to the CNN to determine which features in the starfish were influencing the classification. The enhanced model attained an accuracy of 92.6% while explaining the causal features in COTS. The mean average precision of the enhanced VGG-19 with the addition of the attention model was 95% showing an increase of 2% compared to only the enhanced VGG-19 model.

show abstract

Spatial and Channel Dimensions Attention Feature Transfer for Better Convolutional Neural Networks

Cited by 4 publications

References 13 publications

MR-YOLO: An Improved YOLOv5 Network for Detecting Magnetic Ring Surface Defects

MR-YOLO: An Improved YOLOv5 Network for Detecting Magnetic Ring Surface Defects

Evaluation of Application Effect of Self-Made Compression Cold Therapy in Postoperative Rehabilitation of Patients with Orthopedic Dyskinesia

Identification of Crown of Thorns Starfish (COTS) using Convolutional Neural Network (CNN) and attention model

Contact Info

Product

Resources

About