Mixed Receptive Fields Augmented YOLO with Multi-Path Spatial Pyramid Pooling for Steel Surface Defect Detection

Xia, Kewen; Lv, Zhongliang; Zhou, Chuande; Gu, Guojun; Zhao, Zhiqiang; Liu, Kang; Li, Zelun

doi:10.3390/s23115114

Cited by 21 publications

(4 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Yang, L [10] and colleagues improved the detection precision of weak texture feature insulation defects on complex backgrounds by integrating Spatial Pyramid Pooling (SPP) with the MobileNet network. Xia, KW [11] and his team developed a unique reparameterized large-kernel C3 module specifically for weaktextured targets, combining adaptive receptive fields with multi-scale feature fusion to optimize detection of weak texture steel surface defects. Wang, Yong [12] and his group created various feature extraction modules that combine depth, shape, and texture characteristics of detection targets, input into a cooperative network to enhance detection accuracy.…”

Section: Related Workmentioning

confidence: 99%

A Recurrent Approach for Uninterrupted Tracking of Rotor Blades Using Kalman Filter

Xu,

Fu,

Peng

et al. 2024

IEEE Access

View full text Add to dashboard Cite

With the increasing maintenance demands for wind turbines, unmanned aerial vehicle (UAV) technology has become widely utilized for turbine inspections. However, wind turbine blades, which are thin and long, possess weak texture features that lead to target confusion when tracking specific parts of the dynamic blades. Additionally, wind turbine units, being large dynamic structures, often exceed the camera's field of view (FOV) and exhibit unique motion characteristics. These factors make the visual tracking of specific components unstable due to the lack of global motion information. To address these challenges and achieve consistent calibration of critical components under the motion conditions of the wind turbine, this study has integrated the Squeeze-and-Excitation Network (SEnet) into the backbone network of YOLOv5. Furthermore, two hyperparameters have been introduced into the existing loss function to control the weights of unbalanced samples, thereby enhancing detection accuracy. In the DeepSORT tracking algorithm, multiple Long Short-Term Memory (LSTM) units have been employed to predict the trajectory of the rotor blade's center point, and an optimized Kalman filter has been utilized to improve the system's adaptability and precision. The experimental results demonstrate that this method can accurately distinguish individual blades and specific sections of blades. The enhanced fusion model showed a 5.3% improvement in the mean average precision (mAP_0.5) evaluation metric. Moreover, the method achieved continuous and stable tracking of moving blades, even in scenarios where rotor blades frequently enter and exit the FOV. This approach strengthens the potential for widespread use of drones in wind turbine inspections.

show abstract

Section: Related Workmentioning

confidence: 99%

A Recurrent Approach for Uninterrupted Tracking of Rotor Blades Using Kalman Filter

Xu,

Fu,

Peng

et al. 2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…For example, the R-CNN [ 18 , 19 , 20 , 21 , 22 ] family of algorithms has high detection accuracy but is computationally intensive and inefficient. In contrast, single-stage algorithms such as the SSD [ 23 , 24 , 25 , 26 , 27 ] and YOLO [ 28 , 29 , 30 , 31 , 32 , 33 , 34 , 35 ] series can achieve significant detection speedups. Single-stage target detection algorithms have become preferred in industrial applications due to their ability to directly output information about the position and detection frame of the target to be detected.…”

Section: Related Workmentioning

confidence: 99%

MFF-YOLO: An Accurate Model for Detecting Tunnel Defects Based on Multi-Scale Feature Fusion

Жу

Wang

Xie

et al. 2023

Sensors

View full text Add to dashboard Cite

Tunnel linings require routine inspection as they have a big impact on a tunnel’s safety and longevity. In this study, the convolutional neural network was utilized to develop the MFF-YOLO model. To improve feature learning efficiency, a multi-scale feature fusion network was constructed within the neck network. Additionally, a reweighted screening method was devised at the prediction stage to address the problem of duplicate detection frames. Moreover, the loss function was adjusted to maximize the effectiveness of model training and improve its overall performance. The results show that the model has a recall and accuracy that are 7.1% and 6.0% greater than those of the YOLOv5 model, reaching 89.5% and 89.4%, respectively, as well as the ability to reliably identify targets that the previous model error detection and miss detection. The MFF-YOLO model improves tunnel lining detection performance generally.

show abstract

“…Point cloud processing techniques were combined to locate and identify surface defects on steel plates more precisely. Xia et al [27] proposed an improved YOLOv5s model. A large core C3 module that can be reparametrized is designed innovatively, which enhances the model's ability to perceive and extract features effectively in complex texture environments.…”

Section: Introductionmentioning

confidence: 99%

Steel Surface Defect Detection Algorithm Based on YOLOv8

Song,

Cao,

Zhang

et al. 2024

Electronics

View full text Add to dashboard Cite

To improve the accuracy of steel surface defect detection, an improved model of multi-directional optimization based on the YOLOv8 algorithm was proposed in this study. First, we innovate the CSP Bottleneck with the two convolutions (C2F) module in YOLOv8 by introducing deformable convolution (DCN) technology to enhance the learning and expression ability of complex texture and irregular shape defect features. Secondly, the advanced Bidirectional Feature Pyramid Network (BiFPN) structure is adopted to realize the weight distribution learning of input features of different scales in the feature fusion stage, allowing for more effective integration of multi-level feature information. Next, the BiFormer attention mechanism is embedded in the backbone network, allowing the model to adaptively allocate attention based on target features, such as flexibly and efficiently skipping non-critical areas, and focusing on identifying potentially defective parts. Finally, we adjusted the loss function from Complete-Intersection over Union (CIoU) to Wise-IoUv3 (WIoUv3) and used its dynamic non-monotony focusing property to effectively solve the problem of overfitting the low quality target bounding box. The experimental results show that the mean Average Precision (mAP) of the improved model in the task of steel surface defect detection reaches 84.8%, which depicts a significant improvement of 6.9% compared with the original YOLO8 model. The improved model can quickly and accurately locate and classify all kinds of steel surface defects in practical applications and meet the needs of steel defect detection in industrial production.

show abstract

Mixed Receptive Fields Augmented YOLO with Multi-Path Spatial Pyramid Pooling for Steel Surface Defect Detection

Cited by 21 publications

References 44 publications

A Recurrent Approach for Uninterrupted Tracking of Rotor Blades Using Kalman Filter

A Recurrent Approach for Uninterrupted Tracking of Rotor Blades Using Kalman Filter

MFF-YOLO: An Accurate Model for Detecting Tunnel Defects Based on Multi-Scale Feature Fusion

Steel Surface Defect Detection Algorithm Based on YOLOv8

Contact Info

Product

Resources

About