SDDet: An Enhanced Encoder–Decoder Network With Hierarchical Supervision for Surface Defect Detection

Wang, Junpu; Xu, Guili; Li, Chunlei; Gao, Guangshuai; Wu, Quan Wei

doi:10.1109/jsen.2022.3229031

Cited by 17 publications

(6 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Numerous current studies focus on improving the accuracy of detection by increasing the scale of network [ 28 , 29 ], resulting in slower detection speed and heavier computing resources, which is difficult to be applied in the practical industry production process. Inspired by the ResNet [ 30 ], which is widely used in various industry detection scenarios [ 31 , 32 ] due to its efficient feature extraction capacity, an improved backbone based on ResNet34 is designed to constitute the encoders of the proposed method for efficient feature extraction. The details of the encoders are shown in Table 1 , in which the “Number” denotes that the times to repeat the corresponding operation.…”

Section: Methodsmentioning

confidence: 99%

A Multi-Source Data Fusion Network for Wood Surface Broken Defect Segmentation

Zhu,

Xu,

Lin

et al. 2024

Sensors

View full text Add to dashboard Cite

Wood surface broken defects seriously damage the structure of wooden products, these defects have to be detected and eliminated. However, current defect detection methods based on machine vision have difficulty distinguishing the interference, similar to the broken defects, such as stains and mineral lines, and can result in frequent false detections. To address this issue, a multi-source data fusion network based on U-Net is proposed for wood broken defect detection, combining image and depth data, to suppress the interference and achieve complete segmentation of the defects. To efficiently extract various semantic information of defects, an improved ResNet34 is designed to, respectively, generate multi-level features of the image and depth data, in which the depthwise separable convolution (DSC) and dilated convolution (DC) are introduced to decrease the computational expense and feature redundancy. To take full advantages of two types of data, an adaptive interacting fusion module (AIF) is designed to adaptively integrate them, thereby generating accurate feature representation of the broken defects. The experiments demonstrate that the multi-source data fusion network can effectively improve the detection accuracy of wood broken defects and reduce the false detections of interference, such as stains and mineral lines.

show abstract

Section: Methodsmentioning

confidence: 99%

A Multi-Source Data Fusion Network for Wood Surface Broken Defect Segmentation

Zhu,

Xu,

Lin

et al. 2024

Sensors

View full text Add to dashboard Cite

show abstract

“…The Bi-directional Feature Pyramid Network (BIFPN) is a feature fusion mechanism designed to enhance the performance of object detection models [29]. Its principle involves multi-scale feature fusion in both top-down and bottom-up directions, facilitating the integration of semantic and spatial information across different levels of feature maps.…”

Section: Bifpn Modelmentioning

confidence: 99%

YOLOv8-BCC: Lightweight Object Detection Model Boosts Urban Traffic Safety

Jun,

Lai,

et al. 2024

Preprint

View full text Add to dashboard Cite

With the rapid development of urbanization, the role of urban transportation systems has become increasingly prominent. However, traditional methods of traffic management are struggling to cope with the growing demands of traffic and the complexity of urban environments. In response to this situation, we propose the YOLOv8-BCC algorithm to address existing shortcomings. Leveraging advanced technologies such as CFNet, CBAM attention modules, and BIFPN structure, our algorithm aims to enhance the accuracy, real-time performance, and adaptability of urban traffic intelligent detection systems. Experimental results demonstrate significant improvements in detection accuracy and real-time performance compared to traditional methods. The introduction of the YOLOv8-BCC algorithm provides a robust solution for enhancing urban traffic safety and intelligent management.

show abstract

“…2. The tooth groove widths of both convex and concave teeth are equal, and the width at the top of the grinding wheel used to machine these teeth is also identical [16]. The mechanical model for the centering of circular end teeth used in the study belongs to a model based on the principle of elastic averaging.…”

Section: Design Of Connection Structure Between Arc End Teeth and Dou...mentioning

confidence: 99%

Optimization of electronic sensors based on the connection structure of double rotors with circular arc end teeth

2024

J. vibroeng.

View full text Add to dashboard Cite

To analyze the mechanical performance of the arc end tooth connection structure, the optimization of electronic sensors based on the connection structure of double rotors with circular arc end teeth was studied. By designing and optimizing the structural dimensions of the arc end tooth, the relative position of the grinding wheel and the workpiece was determined, and the optimal electromagnetic actuator size, as well as the number and installation position of the optimized sensors, were obtained. The experimental results showed that the relative error of the rigid vibration mode of the arc end tooth double rotor connection structure was controlled within 10 %. At high speeds, the electromagnetic actuator had a significant control effect on the vibration of the centrifugal machine shaft. It diminished the vibration amplitudes in the X and Y directions by 37.6 % and 30.2 %, respectively. By optimizing the double rotor connection structure with circular end teeth and electronic sensors, effective control of the relative error of rigid vibration modes has been achieved, leading to significant reduction of the centrifugal shaft's vibration at high speeds. In addition, the sensor layout was optimized to enable real-time equipment vibration monitoring, ensuring equipment safety and stability. These achievements not only enhance the mechanical performance of the rotor system, but also provide important guarantees for the stability of the mechanical system.

show abstract

SDDet: An Enhanced Encoder–Decoder Network With Hierarchical Supervision for Surface Defect Detection

Cited by 17 publications

References 45 publications

A Multi-Source Data Fusion Network for Wood Surface Broken Defect Segmentation

A Multi-Source Data Fusion Network for Wood Surface Broken Defect Segmentation

YOLOv8-BCC: Lightweight Object Detection Model Boosts Urban Traffic Safety

Optimization of electronic sensors based on the connection structure of double rotors with circular arc end teeth

Contact Info

Product

Resources

About