Attention-based bi-directional refinement network for salient object detection

Yuan, Junbin; Wei, Jinhui; Wattanachote, Kanoksak; Zeng, Kun; Luo, Xiangfeng; Xu, Qingzhen; Gong, Yongyi

doi:10.1007/s10489-021-03040-8

Cited by 9 publications

(5 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The formula for recall is shown in Equation (10), which indicates that the detection box that is correctly divided into positive samples accounts for the proportion of all positive sample detection boxes.…”

Section: ) Precision Recall Pr Curve and Mapmentioning

confidence: 99%

“…Because the instrument is analog instrument cannot complete the automatic acquisition and transmission work, so in the substation over again need to set a specific person to read and record the instrument work. In recent years, as the smart grid construction work is gradually promoted, the substation gradually realize the intelligence, its instrument type and quantity also increased greatly, and the traditional manual meter reading method has low reading efficiency, easily affected by the surrounding environment factors, human resources consumption, easy to be affected by human subjective factors and other problems, so the manual reading has been unable to meet China's current smart grid development needs 8–13 . At the same time, the intelligent robot technology gradually becomes mature, the substation inspection robot is also developed, and the manual reading of instruments and meters is gradually replaced by the substation inspection robot, and now has achieved certain results 14–17 .…”

Section: Introductionmentioning

confidence: 99%

“…In recent years, as the smart grid construction work is gradually promoted, the substation gradually realize the intelligence, its instrument type and quantity also increased greatly, and the traditional manual meter reading method has low reading efficiency, easily affected by the surrounding environment factors, human resources consumption, easy to be affected by human subjective factors and other problems, so the manual reading has been unable to meet China's current smart grid development needs. [8][9][10][11][12][13] At the same time, the intelligent robot technology gradually becomes mature, the substation inspection robot is also developed, and the manual reading of instruments and meters is gradually replaced by the substation inspection robot, and now has achieved certain results. [14][15][16][17] Therefore, how to realize the intelligence of instrument readings, and apply the object detection algorithm to the inspection robot to achieve instrument target detection in the substation is the current research hotspot.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Substation instrumentation target detection based on multi‐scale feature fusion

Feng

Huang

Sun

et al. 2022

Concurrency and Computation

View full text Add to dashboard Cite

SUMMARY With the promotion of smart grid construction work, the use of high‐precision and high‐efficiency substation inspection robot has become the development trend of substation inspection. A multi‐scale feature fusion meter target detection algorithm is proposed to address the problems of low efficiency and susceptibility to surrounding environmental factors by the traditional manual meter reading method. Kinecct is used to acquire color images of substation meters with different backgrounds, light intensities, and angles to build a substation meter dataset. Based on the complementarity and correlation of multi‐scale features, an SSD target detection model with multi‐scale feature fusion is established, and the performance of the algorithm is tested on the constructed dataset, and comparative experiments are conducted to verify the effectiveness of the algorithm for target detection accuracy improvement.

show abstract

Section: ) Precision Recall Pr Curve and Mapmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Substation instrumentation target detection based on multi‐scale feature fusion

Feng

Huang

Sun

et al. 2022

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

“…In the field of salient object detection, attention mechanisms have also been shown to be effective. It has been implemented into a variety of salient object detection methods and significantly enhanced the performance [54], [55].…”

Section: Attention Mechanismsmentioning

confidence: 99%

CoGANet: Co-Guided Attention Network for Salient Object Detection

Zhao

Song

et al. 2022

IEEE Photonics J.

View full text Add to dashboard Cite

Recent salient object detection methods are mainly based on Convolutional Neural Networks (CNNs). Most of them adopt a U-shape architecture to extract and fuse multi-scale features. The coarser-level semantic information is progressively transmitted to finer-level layers through continuous upsampling operations, and the coarsest-level features will be diluted, resulting in the salient object boundary being blurred. On the other hand, the hand-craft feature has the advantage of being purposive and easy to calculate, in which the edge density feature may help improve the sharpness of the salient object boundary by the rich edge information. In this paper, we propose a Co-Guided Attention Network (CoGANet). On the base of the Feature Pyramid Network (FPN), our model implements a coguided attention mechanism between the image itself and its edge density feature. In the bottom-up pathway of FPN, two streams separately work, taking the original image and its edge density feature as inputs, and each producing five feature maps. Then the last feature map in each stream generates a set of attention maps through a Multi-scale Spatial Attention Module (MSAM). In the top-down pathway, the attention maps of one stream are directly delivered to each stage in the other stream. These attention maps are fused with the feature maps by an Attention-based Feature Fusion Module (AFFM). Finally, an accurate saliency map is produced by fusing the finest-level outputs of the two streams. Experimental results on five mainstream benchmark datasets demonstrate our model is superior to 13 state-of-the-art methods in terms of four evaluation metrics.

show abstract

“…Hence we are to pursue an adaptive weight assignment strategy to obtain the more reasonable confidence scores automatically for predicted bounding boxes, as shown in Figure 1b. The attention mechanism [18,19] offers a good option for learning weights adaptively, and has been widely utilized in various tasks such as image classification, detection and segmentation [20,21]. There are commonly two types of attention, i.e., the spatial attention and channel attention.…”

Section: Introductionmentioning

confidence: 99%

Improved Dual Attention for Anchor-Free Object Detection

Xiang

Zhao

et al. 2022

Sensors

View full text Add to dashboard Cite

In anchor-free object detection, the center regions of bounding boxes are often highly weighted to enhance detection quality. However, the central area may become less significant in some situations. In this paper, we propose a novel dual attention-based approach for the adaptive weight assignment within bounding boxes. The proposed improved dual attention mechanism allows us to thoroughly untie spatial and channel attention and resolve the confusion issue, thus it becomes easier to obtain the proper attention weights. Specifically, we build an end-to-end network consisting of backbone, feature pyramid, adaptive weight assignment based on dual attention, regression, and classification. In the adaptive weight assignment module based on dual attention, a parallel framework with the depthwise convolution for spatial attention and the 1D convolution for channel attention is applied. The depthwise convolution, instead of standard convolution, helps prevent the interference between spatial and channel attention. The 1D convolution, instead of fully connected layer, is experimentally proved to be both efficient and effective. With the adaptive and proper attention, the correctness of object detection can be further improved. On public MS-COCO dataset, our approach obtains an average precision of 52.7%, achieving a great increment compared with other anchor-free object detectors.

show abstract

Attention-based bi-directional refinement network for salient object detection

Cited by 9 publications

References 48 publications

Substation instrumentation target detection based on multi‐scale feature fusion

Substation instrumentation target detection based on multi‐scale feature fusion

CoGANet: Co-Guided Attention Network for Salient Object Detection

Improved Dual Attention for Anchor-Free Object Detection

Contact Info

Product

Resources

About