Yapeng Wu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Yapeng Wu has a total of 37 co-authors. Here are the top co-authors by number of shared publications:
Chen Yang
3 shared publicationsJing Liu
3 shared publicationsMin Yang
School of Mechanical Engineering and Automation, Beihang University, 100191, Beijing, People’s Republic of China
3 shared publications- Chao Hai
School of Mechanical Engineering and Automation, Beihang University, Beijing 100191, People’s Republic of China
2 shared publications Dalong Tan
2 shared publicationsHaoyan Guo
2 shared publications
Latest Publications
- 16 October 2026
- Published by Elsevier in Measurement
- Vol. 292
YOLOv11-Fire: an intelligent fire detection model based on visible and infrared fusion images
- 16 October 2026
- Published by Elsevier in Measurement
- Vol. 292
YOLOv11-Fire: an intelligent fire detection model based on visible and infrared fusion images
- Open Access
- 24 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 19 (19391404) , 15158-15184
EMSF-Net: An Efficient Multiscale Fusion Network With Lightweight Attention for Object Detection in Remote Sensing Images
Remote-sensing object detection is an important technique for addressing global challenges, including climate change and environmental pollution. However, existing methods still suffer from limited model adaptability, high miss-detection rates, and considerable deployment costs, particularly in UAV-based and edge-computing scenarios. To tackle these issues, an Efficient Multi-Scale Fusion network, termed EMSF-Net, is proposed by integrating the strengths of (Convolutional Neural Networks) CNNs and the Swin Transformer. The network incorporates an efficient multi-scale fusion module (MSF-Block) to achieve complementary integration of global semantic information and local fine-grained details, a lightweight detection head (LDetect) to improve classification and localization performance, two feature-enhancement modules-an enhanced feed-forward module (C2PSA-GDFN) and a depth-enhanced fusion module (C2D)-that jointly strengthen feature representation and fusion. In addition, a multi-scale optimization strategy based on the P2 prediction layer and depthwise convolution (DWConv) is constructed to reinforce fine-grained feature extraction and spatial localization. A customized SIoU loss function combined with a loss-weight optimization strategy is further designed to enhance bounding-box regression accuracy and training stability. Experimental results demonstrate that EMSF-Net achieves improvements of 8.4%, 3.3%, 8.8%, and 5.4% in mAP over the baseline model YOLOv11s across four mainstream datasets, while maintaining a favorable balance between detection accuracy and computational efficiency. Moreover, the reliability and generalization performance of the proposed model are further validated on the SIMD and VEDAI dataset, offering an effective and reliable approach for remote-sensing object detection in resource-limited settings.
- 24 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 19 (19391404) , 15158-15184
Open AccessEMSF-Net: An Efficient Multiscale Fusion Network With Lightweight Attention for Object Detection in Remote Sensing Images
Remote-sensing object detection is an important technique for addressing global challenges, including climate change and environmental pollution. However, existing methods still suffer from limited model adaptability, high miss-detection rates, and considerable deployment costs, particularly in UAV-based and edge-computing scenarios. To tackle these issues, an Efficient Multi-Scale Fusion network, termed EMSF-Net, is proposed by integrating the strengths of (Convolutional Neural Networks) CNNs and the Swin Transformer. The network incorporates an efficient multi-scale fusion module (MSF-Block) to achieve complementary integration of global semantic information and local fine-grained details, a lightweight detection head (LDetect) to improve classification and localization performance, two feature-enhancement modules-an enhanced feed-forward module (C2PSA-GDFN) and a depth-enhanced fusion module (C2D)-that jointly strengthen feature representation and fusion. In addition, a multi-scale optimization strategy based on the P2 prediction layer and depthwise convolution (DWConv) is constructed to reinforce fine-grained feature extraction and spatial localization. A customized SIoU loss function combined with a loss-weight optimization strategy is further designed to enhance bounding-box regression accuracy and training stability. Experimental results demonstrate that EMSF-Net achieves improvements of 8.4%, 3.3%, 8.8%, and 5.4% in mAP over the baseline model YOLOv11s across four mainstream datasets, while maintaining a favorable balance between detection accuracy and computational efficiency. Moreover, the reliability and generalization performance of the proposed model are further validated on the SIMD and VEDAI dataset, offering an effective and reliable approach for remote-sensing object detection in resource-limited settings.
- 29 January 2026
- Published by Elsevier in Infrared Physics & Technology
- Vol. 154
DualWaveTransnet: A wavelet convolution Transformer Dual-Branch model for Dual-Spectral image fusion
- 29 January 2026
- Published by Elsevier in Infrared Physics & Technology
- Vol. 154
DualWaveTransnet: A wavelet convolution Transformer Dual-Branch model for Dual-Spectral image fusion
- 18 February 2025
- Published by IOP Publishing in Engineering Research Express
- Vol. 7 (1) , 015255
Flame recognition and location algorithm based on YOLO v5s and binocular vision
An improved YOLO v5s framework in conjunction with binocular vision is developed to fulfill the stringent requirements for precise identification and accurate localization of flames. An Efficient Multi-scale Attention (EMA) mechanism is seamlessly integrated into the backbone network of the YOLO v5s model, thereby significantly improving the capacity to focus on and interpret critical target features. To bolster the multi-scale feature fusion capability, the Bidirectional Feature Pyramid Network (BiFPN) is strategically introduced at the fusion layer within the object detection architecture. The Semi-Global Block Matching (SGBM) algorithm is deployed to synchronize the binocular images, complemented by the least squares curve fitting method to adjust location inaccuracies, enhancing the precision of flame positioning. A binocular vision system is meticulously constructed to facilitate a series of rigorous experiments on flame detection and location. The experimental results demonstrate that the improved YOLO v5s network model achieves an exceptional flame recognition rate of 96.25%, which represents a noteworthy increase of 0.98% in accuracy in comparison to the original YOLO v5s model. Additionally, within a distance range of 3 to 10 meters, the flame ranging error is consistently less than 0.1 meters, successfully realizing the objective of high-precision flame detection and localization.
- 18 February 2025
- Published by IOP Publishing in Engineering Research Express
- Vol. 7 (1) , 015255
Flame recognition and location algorithm based on YOLO v5s and binocular vision
An improved YOLO v5s framework in conjunction with binocular vision is developed to fulfill the stringent requirements for precise identification and accurate localization of flames. An Efficient Multi-scale Attention (EMA) mechanism is seamlessly integrated into the backbone network of the YOLO v5s model, thereby significantly improving the capacity to focus on and interpret critical target features. To bolster the multi-scale feature fusion capability, the Bidirectional Feature Pyramid Network (BiFPN) is strategically introduced at the fusion layer within the object detection architecture. The Semi-Global Block Matching (SGBM) algorithm is deployed to synchronize the binocular images, complemented by the least squares curve fitting method to adjust location inaccuracies, enhancing the precision of flame positioning. A binocular vision system is meticulously constructed to facilitate a series of rigorous experiments on flame detection and location. The experimental results demonstrate that the improved YOLO v5s network model achieves an exceptional flame recognition rate of 96.25%, which represents a noteworthy increase of 0.98% in accuracy in comparison to the original YOLO v5s model. Additionally, within a distance range of 3 to 10 meters, the flame ranging error is consistently less than 0.1 meters, successfully realizing the objective of high-precision flame detection and localization.
- 17 January 2024
- Published by Elsevier in NDT & E International
- Vol. 143
An enhancement algorithm based on multi-grayscale fusion and edge-weight for low contrast X-ray image
- 17 January 2024
- Published by Elsevier in NDT & E International
- Vol. 143
An enhancement algorithm based on multi-grayscale fusion and edge-weight for low contrast X-ray image
- 10 November 2022
- Published by IOP Publishing in Measurement Science and Technology
- Vol. 34 (2) , 025402
Deformation measurement within lithium-ion battery using sparse-view computed tomography and digital image correlation
Electrode deformation can cause high local strain and serious capacity degradation in lithium-ion batteries (LIBs) during cycling. Risk reduction in many applications requires an understanding of the effects of the charging/discharging rate on the electrode structure during the battery life cycle. Cyclic charging/discharging experiments of wound 18 650 cylindrical LIBs were conducted at different charging/discharging rates (1C and 2C) to determine the effect of rate on electrode deformation. The charging/discharging capacity as well as battery voltage and time were analyzed during cycling. To acquire electrode deformation images and meet the requirements of computed tomography (CT) within 2 min during the charging/discharging process, sparse-view CT was performed at fixed cycle intervals. Subsequently, a sparse-view reconstruction network was proposed to generate a slice image. Finally, the electrode displacement and strain fields were calculated using the augmented Lagrangian digital image correlation algorithm. The causes of electrode deformation were analyzed and discussed from the perspective of molecular and macroscopic structure. Experimental results show that the structural similarity, peak signal-to-noise ratio and root mean square error in the reconstructed image of the axial section within the battery obtained via the proposed network were 0.9616, 38.7411 dB and 0.0108, respectively, which were better than the other methods of comparison. After 100 cycles, the capacity decay of the battery at 2C was 9.23-fold higher than that at 1C. After 100 cycles at 2C, the maximum displacement of the electrode reached 0.46 mm along the x-direction. The electrode structural deformation of the battery can be intuitively understood at different rates, which facilitates reasonable utilization and structural optimization of the battery.
- 10 November 2022
- Published by IOP Publishing in Measurement Science and Technology
- Vol. 34 (2) , 025402
Deformation measurement within lithium-ion battery using sparse-view computed tomography and digital image correlation
Electrode deformation can cause high local strain and serious capacity degradation in lithium-ion batteries (LIBs) during cycling. Risk reduction in many applications requires an understanding of the effects of the charging/discharging rate on the electrode structure during the battery life cycle. Cyclic charging/discharging experiments of wound 18 650 cylindrical LIBs were conducted at different charging/discharging rates (1C and 2C) to determine the effect of rate on electrode deformation. The charging/discharging capacity as well as battery voltage and time were analyzed during cycling. To acquire electrode deformation images and meet the requirements of computed tomography (CT) within 2 min during the charging/discharging process, sparse-view CT was performed at fixed cycle intervals. Subsequently, a sparse-view reconstruction network was proposed to generate a slice image. Finally, the electrode displacement and strain fields were calculated using the augmented Lagrangian digital image correlation algorithm. The causes of electrode deformation were analyzed and discussed from the perspective of molecular and macroscopic structure. Experimental results show that the structural similarity, peak signal-to-noise ratio and root mean square error in the reconstructed image of the axial section within the battery obtained via the proposed network were 0.9616, 38.7411 dB and 0.0108, respectively, which were better than the other methods of comparison. After 100 cycles, the capacity decay of the battery at 2C was 9.23-fold higher than that at 1C. After 100 cycles at 2C, the maximum displacement of the electrode reached 0.46 mm along the x-direction. The electrode structural deformation of the battery can be intuitively understood at different rates, which facilitates reasonable utilization and structural optimization of the battery.
- Open Access
- 19 November 2021
- Published by MDPI AG in Applied Sciences
- Vol. 11 (22) , 10942
Sparse-View Neutron CT Reconstruction Using a Modified Weighted Total Difference Minimization Method
Indirect neutron imaging is an effective method for nondestructive testing of spent nuclear fuel elements. Considering the difficulty of obtaining experimental data in a high-radiation environment and the characteristic of high noise of neutron images, it is difficult to use the traditional FBP algorithm to recover the complete information of the sample based on the limited projection data. Therefore, it is necessary to develop the sparse-view CT reconstruction algorithm for indirect neutron imaging. In order to improve the quality of the reconstruction image, an iterative reconstruction method combining SIRT, MRP, and WTDM regularization is proposed. The reconstruction results obtained by using the proposed method on simulated data and actual neutron projection data are compared with the results of four other algorithms (FBP, SIRT, SIRT-TV, and SIRT-WTDM). The experimental results show that the SIRT-MWTDM algorithm has great advantages in both objective evaluation index and subjective observation in the reconstruction image of simulated data and neutron projection data.
- 19 November 2021
- Published by MDPI AG in Applied Sciences
- Vol. 11 (22) , 10942
Open AccessSparse-View Neutron CT Reconstruction Using a Modified Weighted Total Difference Minimization Method
Indirect neutron imaging is an effective method for nondestructive testing of spent nuclear fuel elements. Considering the difficulty of obtaining experimental data in a high-radiation environment and the characteristic of high noise of neutron images, it is difficult to use the traditional FBP algorithm to recover the complete information of the sample based on the limited projection data. Therefore, it is necessary to develop the sparse-view CT reconstruction algorithm for indirect neutron imaging. In order to improve the quality of the reconstruction image, an iterative reconstruction method combining SIRT, MRP, and WTDM regularization is proposed. The reconstruction results obtained by using the proposed method on simulated data and actual neutron projection data are compared with the results of four other algorithms (FBP, SIRT, SIRT-TV, and SIRT-WTDM). The experimental results show that the SIRT-MWTDM algorithm has great advantages in both objective evaluation index and subjective observation in the reconstruction image of simulated data and neutron projection data.
- 17 February 2021
- Published by IOP Publishing in Measurement Science and Technology
- Vol. 32 (4) , 045405
Pole-piece position distance identification of cylindrical lithium-ion battery through x-ray testing technology
In the preparation process of cylindrical lithium-ion batteries, a rigorous manufacturing process demands that the position distances between positive and negative pole-pieces must be kept within a reasonable range of variation. Otherwise, a too small position distance may cause safety problems, such as short circuits and thermal runaway. To inspect the position distances between positive and negative pole-pieces automatically, and to decrease the risk of safety and economic losses during the subsequent use, this paper proposes a method to identify the position distance defects of a cylindrical lithium-ion battery on the base of x-ray digital radiography (DR) images. According to this method, the DR image is firstly enhanced by the GPU-accelerated homomorphic filtering algorithm to intensify its contrast and detailed information. Then through the Shi-Tomasi corner detection algorithm, corners of all the positive and negative pole-pieces are preliminarily detected in a region of interest which is defined in advance. To delete the false corners and find the lost corners, one-dimension region growing and curve-fitting methods are adopted. Finally, the minimum position distance, repeated accuracy and alignment metrics are calculated at the base of the detected corners. The experimental results show that the corner positions of five 26 650 cylindrical lithium-ion batteries with different pole-piece structural characteristics can be effectively identified by the proposed method, which provides a useful approach to filtrate unqualified batteries during the process of manufacture.
- 17 February 2021
- Published by IOP Publishing in Measurement Science and Technology
- Vol. 32 (4) , 045405
Pole-piece position distance identification of cylindrical lithium-ion battery through x-ray testing technology
In the preparation process of cylindrical lithium-ion batteries, a rigorous manufacturing process demands that the position distances between positive and negative pole-pieces must be kept within a reasonable range of variation. Otherwise, a too small position distance may cause safety problems, such as short circuits and thermal runaway. To inspect the position distances between positive and negative pole-pieces automatically, and to decrease the risk of safety and economic losses during the subsequent use, this paper proposes a method to identify the position distance defects of a cylindrical lithium-ion battery on the base of x-ray digital radiography (DR) images. According to this method, the DR image is firstly enhanced by the GPU-accelerated homomorphic filtering algorithm to intensify its contrast and detailed information. Then through the Shi-Tomasi corner detection algorithm, corners of all the positive and negative pole-pieces are preliminarily detected in a region of interest which is defined in advance. To delete the false corners and find the lost corners, one-dimension region growing and curve-fitting methods are adopted. Finally, the minimum position distance, repeated accuracy and alignment metrics are calculated at the base of the detected corners. The experimental results show that the corner positions of five 26 650 cylindrical lithium-ion batteries with different pole-piece structural characteristics can be effectively identified by the proposed method, which provides a useful approach to filtrate unqualified batteries during the process of manufacture.