Article
Computer Science, Artificial Intelligence
Du-Ming Tsai, Po-Hao Jen
Summary: This paper evaluates the unsupervised autoencoder learning method for automated defect detection in manufacturing, and proposes a new CAE model with regularizations that significantly improves the detection performance based on the center of defect samples.
ADVANCED ENGINEERING INFORMATICS
(2021)
Article
Computer Science, Artificial Intelligence
Iker Pastor-Lopez, Borja Sanz, Alberto Tellaeche, Giuseppe Psaila, Jose Gaviria de la Puerta, Pablo G. Bringas
Summary: The study proposed a method called BoDoC for multi-objective recognition in image classification problems. By creating a new dataset and using a series of techniques, the classification results were successfully improved, achieving high precision.
Article
Energy & Fuels
Yi Li, Minzhe Ni, Yanfeng Lu
Summary: This paper introduces an image enhancement method based on illumination correction and compensation to address the issues of uneven illumination, low contrast, and poor details display in outdoor images. Additionally, a real-time one-step detection model based on YOLOv5 is proposed for detecting insulator defects. Evaluation results demonstrate that the proposed method achieves competitive results while maintaining real-time performance.
Article
Construction & Building Technology
Mingzhu Wang, Srinath Shiv Kumar, Jack C. P. Cheng
Summary: This study proposes a framework based on deep learning defect detection and metric learning for tracking multiple sewer defects in CCTV videos. Experimental results show that tracking performance can be influenced by detection accuracy and configurations of the metric learning module.
AUTOMATION IN CONSTRUCTION
(2021)
Article
Engineering, Multidisciplinary
Junjie Xing, Minping Jia
Summary: An automatic detection method based on convolutional neural networks is proposed in this paper and its detection performance is evaluated and compared with other models, showing that the method has better performance in real-time automatic detection of workpiece surface defects.
Article
Engineering, Multidisciplinary
Chaobo Zhang, Chih-chen Chang, Maziar Jamshidi
Summary: The study introduces a novel fully convolutional model for detecting and grouping image pixels for individual defects, demonstrating good robustness and accuracy. Compared with existing instance segmentation and semantic segmentation networks, the proposed model has a distinct advantage in boundary delineation and accuracy.
STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL
(2021)
Article
Materials Science, Textiles
Zebin Su, Hao Zhang, Pengfei Li, Huanhuan Zhang, Yanjun Lu
Summary: A lightweight model of digital printing fabric defect detection based on YOLOX is proposed in this paper. By introducing the SE attention module to enhance features, it improves the accuracy of diversified defect detection and solves the influence of small target detection accuracy.
JOURNAL OF ENGINEERED FIBERS AND FABRICS
(2023)
Article
Engineering, Electrical & Electronic
Du-Ming Tsai, Shu-Kai S. Fan, Yi-Hsiang Chou
Summary: This study proposes a deep learning scheme for automatic defect detection in material surfaces, using CycleGAN to automatically generate defect annotations without manual work, resulting in high accuracy and efficiency in training.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2021)
Article
Automation & Control Systems
Yuechu Mao, Julong Yuan, Yongjian Zhu, Yingguang Jiang
Summary: Due to the difficulty and variability of smartphone glass detection, the accuracy of detection results is easily affected by the environment. This study proposes a new network called Dy-YOLO v5s, which incorporates an attention module, cross-scale and cross-layer connections, and a dynamic detection framework to improve feature extraction and information exchange capabilities. Experimental results show that Dy-YOLO v5s achieves high precision and recall rates, and outperforms other deep-learning algorithms in terms of overall accuracy and real-time performance.
INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY
(2023)
Article
Computer Science, Information Systems
Zhiyan Zhong, Hongxin Wang, Dan Xiang
Summary: Surface defect detection is critical in improving production yield of magnetic tiles. Existing methods struggle to accurately locate and segment small defects due to their low proportions and difficult visual identification. To overcome these challenges, we propose an effective algorithm that decomposes the image, estimates possible defect regions, locates defects, enhances contrast, and precisely segments defect areas. Experimental results demonstrate the algorithm's superiority and potential for industrial applications.
Review
Engineering, Industrial
Yongbing Zhou, Minghao Yuan, Jian Zhang, Guofu Ding, Shengfeng Qin
Summary: This paper systematically reviews the methods and development trends in PCB defect detection. It summarizes the latest image processing and machine learning methods used in PCB defect detection, discusses commonly used evaluation indicators and data sets, as well as feedback and optimization processes in visual inspection and manufacturing systems. It proposes an intelligent PCB defect visual detection approach as a future research direction.
JOURNAL OF MANUFACTURING SYSTEMS
(2023)
Article
Computer Science, Information Systems
Zhengrui Peng, Xinyi Gong, Bengang Wei, Xiangyi Xu, Shixiong Meng
Summary: This paper introduces an unsupervised learning method based on self-feature comparison for accurately locating and segmenting anomalies in fabric texture images. Compared to traditional methods, this approach performs better in locating anomalies on fiber texture surfaces.
Article
Engineering, Multidisciplinary
Tongjia Zhang, Chengrui Zhang, Yanjie Wang, Xiaofu Zou, Tianliang Hu
Summary: This paper proposes a vision-based fusion method to automatically and reliably detect defects on the spiral cutting edge. By using improved Yolov3-tiny to extract the target cutting edge region and traditional image processing method to detect and evaluate defects, the method improves the detection accuracy and evaluation precision of defects, as well as the robustness of illumination compared to using deep learning alone. The case study shows that the proposed method can effectively detect and evaluate small defects on the spiral cutting edges illumination insensitively with high detection accuracy.
Article
Engineering, Electrical & Electronic
Jiawei Zhang, Yu Liu, Yuan Li, Jinyong Yu
Summary: The proposed framework in this article provides a unified and generalized detection method for pointer meters by using techniques such as background modeling, similarity measurement, and point-to-point distance fitting to enhance images, locate pointers and scale marks, and calculate indicator values.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2022)
Article
Automation & Control Systems
Xinyue Zhao, Quanzhi Li, Menghan Xiao, Zaixing He
Summary: In this paper, a novel method based on 3D machine vision is proposed to detect small 3D printing defects. The method includes two steps, potential defect region extraction and accurate defect detection. Experiments prove that the proposed method is more accurate and robust than other methods for both potential defect region extraction and precise defect detection.
INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY
(2023)
Article
Computer Science, Information Systems
Yali Yuan, Christian Melching, Yachao Yuan, Dieter Hogrefe
Article
Computer Science, Information Systems
Yali Yuan, Liuwei Huo, Yachao Yuan, Zhixiao Wang
INTERNATIONAL JOURNAL OF DISTRIBUTED SENSOR NETWORKS
(2019)
Article
Engineering, Electrical & Electronic
Yali Yuan, Robert Tasik, Sripriya Srikant Adhatarao, Yachao Yuan, Zheli Liu, Xiaoming Fu
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
(2020)
Article
Chemistry, Physical
Yang Liu, Yachao Yuan, Cristhian Balta, Jing Liu
Article
Computer Science, Information Systems
Yachao Yuan, Md Saiful Islam, Yali Yuan, Shengjin Wang, Thar Baker, Lutz Maria Kolbe
Summary: In this article, the EcRD framework is proposed for road damage detection and warning, leveraging the advantages of edge and cloud computing. It includes a simple and efficient road segmentation algorithm, a light-weighted road damage detector, and a multitypes road damage detection model for accurate and rapid detection. The proposed approach is significantly faster than cloud-based methods, with improved accuracy and low storage and labeling costs.
IEEE INTERNET OF THINGS JOURNAL
(2021)
Article
Computer Science, Theory & Methods
Yachao Yuan, Yali Yuan, Thar Baker, Lutz Maria Kolbe, Dieter Hogrefe
Summary: In this paper, we propose FedRD, a novel privacy-preserving edge-cloud and Federated learning-based framework for intelligent hazardous Road Damage detection and warning. FedRD achieves high detection performance and provides fast responses with accurate warning information covering a wider area while preserving users' privacy, even when some edges have limited data.
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE
(2021)
Article
Engineering, Multidisciplinary
Yang Liu, Yachao Yuan, Jing Liu
Summary: This paper proposes a novel deep learning model ImDeep for imbalanced multi-label surface defect classification, which integrates three key techniques to improve classification performance and reduce model complexity and latency on small datasets.
MEASUREMENT SCIENCE AND TECHNOLOGY
(2022)
Article
Computer Science, Information Systems
Yachao Yuan, Yali Yuan, Parisa Memarmoshrefi, Thar Baker, Dieter Hogrefe
Summary: This article introduces a novel load-balanced secure and private EV charging framework that ensures reliable and efficient charging services through lightweight encryption techniques and online pricing strategies. It also balances energy load consumption and preserves users' privacy.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Proceedings Paper
Computer Science, Hardware & Architecture
Yali Yuan, Sripriya Srikant Adhatarao, Mingkai Lin, Yachao Yuan, Zheli Liu, Xiaoming Fu
IEEE INFOCOM 2020 - IEEE CONFERENCE ON COMPUTER COMMUNICATIONS
(2020)