相关参考文献
注意:仅列出部分参考文献,下载原文获取全部文献信息。
Article
Forestry
Rijun Wang et al.
Summary: This article introduces a new method for detecting wood surface defects, which improves detection capability and performance by incorporating a dynamic convolution coordinate attention mechanism and an efficient feature extraction module. Experimental results demonstrate that the method achieves significant effectiveness in wood defect detection.
Proceedings Paper
Computer Science, Artificial Intelligence
Jierun Chen et al.
Summary: To design faster neural networks, this paper proposes a novel partial convolution (PConv) method and a new family of neural networks called FasterNet. By reducing redundant computation and memory access, PConv can extract spatial features more efficiently, while FasterNet achieves higher running speed on various devices without sacrificing accuracy.
2023 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Chien-Yao Wang et al.
Summary: Real-time object detection is an important research topic in computer vision, and the development of new approaches in architecture optimization and training optimization has led to two related research topics. To address these topics, a trainable solution combining flexible and efficient training tools, proposed architecture, and compound scaling method is proposed. YOLOv7 outperforms all known object detectors in terms of speed and accuracy, achieving the highest AP accuracy of 56.8% among real-time object detectors with 30 FPS or higher on GPU V100.
2023 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR
(2023)
Article
Computer Science, Information Systems
Zhiqiang Kang et al.
Summary: This study proposes a novel DME-YOLO model that addresses the dilemma of detection accuracy versus speed in traditional YOLO models. By establishing a backbone based on YOLOv7 and Dense blocks and applying feature multiplexing, the model achieves accelerated defect detection. The addition of a multi-source attention mechanism module called MSAM compensates for the loss of spatial information and improves the detection accuracy of small target defects while ensuring real-time detection.
Article
Materials Science, Ceramics
Guang Wan et al.
Summary: In this study, a detection method based on improved YOLOv5s for ceramic tile detection was proposed. By deepening the network layer, adding attention mechanism module, introducing small-scale detection layer, enhancing network feature fusion, and replacing convolution, the problems caused by difficulty in extracting texture features and small defects were effectively solved.
CERAMICS INTERNATIONAL
(2022)
Article
Chemistry, Analytical
Jianfeng Zheng et al.
Summary: This paper proposes an improved YOLOv7 model for detecting insulator-defect targets. The model clusters the target boxes of the insulator dataset to generate more suitable anchor boxes for detection. It introduces the CoordAtt module and HorBlock module to enhance the feature-extraction process and weaken ineffective features. The model uses SIoU and focal loss functions to accelerate convergence and improves the NMS method for better detection performance.
Article
Automation & Control Systems
Zhaohui Zheng et al.
Summary: The proposed CIoU loss and Cluster-NMS approach, which incorporates geometric factors, significantly improve average precision and average recall in object detection and instance segmentation, with notable gains without sacrificing inference efficiency.
IEEE TRANSACTIONS ON CYBERNETICS
(2022)
Article
Chemistry, Multidisciplinary
Xin Li et al.
Summary: This paper proposes a surface defect detection model based on an improved YOLOv5 object detection algorithm to address the low efficiency and poor accuracy issues in conventional methods. The model recalculates the parameters of preset anchors using a k-means clustering algorithm, adds an attention mechanism, and improves the neck network structure. Experimental results show that the proposed model achieves higher accuracy and lower inference time compared to other object detection algorithms, making it efficient and suitable for practical applications.
APPLIED SCIENCES-BASEL
(2022)
Article
Chemistry, Multidisciplinary
Haijiang Zhu et al.
Summary: This paper proposes an improved IA-Mask R-CNN detection method for surface defect detection on automotive engine parts. By establishing an image dataset and analyzing labeled data, suitable anchor scales for surface defect detection are determined to improve the anchor design in Mask R-CNN. The experimental results show that the proposed method outperforms other detection methods in detecting both minor and larger defects.
APPLIED SCIENCES-BASEL
(2022)
Article
Computer Science, Artificial Intelligence
Yi-Fan Zhang et al.
Summary: This paper focuses on the crucial step of bounding box regression in object detection and proposes two improvements, named EIOU loss and Focal-EIOU loss, to address the issues with previous loss functions. Experimental results demonstrate notable superiority in convergence speed and localization accuracy compared to other methods.
Article
Computer Science, Artificial Intelligence
Zhaowei Cai et al.
Summary: In object detection, the commonly used IoU threshold of 0.5 can lead to noisy detections, and performance may degrade for larger thresholds. The Cascade R-CNN architecture addresses this issue by training detectors sequentially with increasing IoU thresholds and eliminating quality mismatches at inference, resulting in state-of-the-art performance and significant improvement in high-quality detection across various datasets. The model is also generalized to instance segmentation, achieving nontrivial improvements over existing methods.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2021)
Article
Computer Science, Information Systems
Zhihua Xu et al.
Summary: A multi-stage balanced R-CNN (MSB R-CNN) for defect detection is proposed in this paper based on Cascade R-CNN, which uses deformable convolution, balanced feature pyramid, and balanced loss to improve the accuracy and convergence effect of the model.
Article
Computer Science, Software Engineering
Yibin Huang et al.
Article
Computer Science, Artificial Intelligence
Shaoqing Ren et al.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2017)
Article
Computer Science, Artificial Intelligence
Saeed Hosseinzadeh Hanzaei et al.
PATTERN RECOGNITION
(2017)
Proceedings Paper
Computer Science, Artificial Intelligence
Ross Girshick
2015 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV)
(2015)