A5082921939- Industrial Vision Systems and Defect Detection
- High Temperature Alloys and Creep
- Image Processing Techniques and Applications
- Optical measurement and interference techniques
- High-Temperature Coating Behaviors
- Anomaly Detection Techniques and Applications
- High Entropy Alloys Studies
- Surface Roughness and Optical Measurements
- Non-Destructive Testing Techniques
- Infrastructure Maintenance and Monitoring
- Image and Object Detection Techniques
- Advanced Neural Network Applications
- Intermetallics and Advanced Alloy Properties
- Integrated Circuits and Semiconductor Failure Analysis
- Additive Manufacturing Materials and Processes
- Advanced Vision and Imaging
- Prenatal Screening and Diagnostics
- Robotics and Sensor-Based Localization
- Video Surveillance and Tracking Methods
- Reproductive Biology and Fertility
- Advanced Image and Video Retrieval Techniques
- Nuclear Materials and Properties
- Advanced Measurement and Metrology Techniques
- Robot Manipulation and Learning
- Aluminum Alloy Microstructure Properties
Chinese Academy of Sciences
2015-2025
Shandong Institute of Automation
2016-2025
Institute of Automation
2016-2024
Guiyang Medical University
2024
Beijing Academy of Artificial Intelligence
2020-2023
University of Chinese Academy of Sciences
2020-2023
Binzhou Technician College
2023
University of Science and Technology of China
1994-2021
Shenyang University of Technology
2017-2018
General Electric (India)
2010
As the failure of power line insulators leads to transmission systems, an insulator inspection system based on aerial platform is widely used. Insulator defect detection performed against complex backgrounds in images, presenting interesting but challenging problem. Traditional methods, handcrafted features or shallow learning techniques, can only localize and detect faults under specific conditions, such as when sufficient prior knowledge available, with low background interference, at...
Automatic metallic surface defect inspection has received increased attention in relation to the quality control of industrial products. Metallic detection is usually performed against complex scenarios, presenting an interesting but challenging problem. Traditional methods are based on image processing or shallow machine learning techniques, these can only detect defects under specific conditions, such as obvious contours with strong contrast and low noise, at certain scales, illumination...
Currently, deep learning-based visual inspection has been highly successful with the help of supervised learning methods. However, in real industrial scenarios, scarcity defect samples, cost annotation, and lack a priori knowledge defects may render supervised-based methods ineffective. In recent years, unsupervised anomaly localization algorithms have become more widely used tasks. This paper aims to researchers this field by comprehensively surveying achievements images using learning. The...
Unsupervised anomaly detection in real industrial scenarios is challenging since the small amount of defect-free images contain limited discriminative information, and defects are unpredictable. Although nowadays image reconstruction-based methods widely being used various applications, they cannot effectively learn semantic representation, which leads to imperfect reconstruction. In this article, formulated as a joint problem feature reconstruction inpainting dual-siamese framework. The...
The coexistence of subtle and long-range anomalies in real-world industrial applications brings significant challenges for anomaly localization. Existing methods typically train deep models by utilizing the multilevel patches or layers-fusion approaches learning global-local distribution; however, these do not consider local global features simultaneously, which suffer from inaccurate localization results. To this end, a hybrid transformer model, ViTALnet, is proposed here, established based...
As a critical electrical connector component in the modern industrial environment, spring-wire sockets and their manufacture quality are closely relevant to equipment safety. These types of defects difficult properly distinguish due defect similarity diversity. In such cases, can only be determined using cumbersome human visual inspection. To satisfy requirements control, machine vision apparatus for inspection is presented this paper. With brief description system design, our emphasis put...
Surface defects on precision optical elements must be carefully inspected since they impact the normal operation of an system. It is a challenge to inspect large-aperture using imaging system because efficiency and accuracy. This paper designs novel effective inspection instrument with two systems for elements. They are dark-field (DFIS) line scan camera in 10-μm resolution bright-field (BF) microscopic 0.85-μm resolution. To keep clarity DFIS large-scope quickly scanning, adaptive path...
The force-based control algorithm of robotic multiple peg-in-hole assembly is a challenge. For the difficulty low adaptability model-based algorithms and learning efficiency model-free algorithms, goal-based hierarchical policy (HPL) that combines conventional demonstration (DL) proposed to learn skill. First, HPL adds goal as new variable action value function. Multiple states reached in each episode are randomly selected subgoals improve distribution positive rewards. Second, an initial DL...
Scratches are one of the most common defects in industrial manufacturing. Weak scratches environment have an ambiguous edge, low contrast, large span, and unfixed shape, which bring difficulty for automatic defect detection. Recently, many existing visual inspection methods based on deep learning cannot completely effectively inspect weak due to lack discriminative features sufficient spatial detail. In this article, a novel DeepScratchNet is proposed scratch detection by aggregating rich...
Scratches as the major defects in precision optical components are caused inevitably manufacturing process, which is harmful to whole system. Most scratches on surface of weak with low contrast and uneven distribution gray scale, poses a significant problem for inspection. In this article, an end-to-end scratch inspection method based novel scratch-enhancement methods convolutional neural network (CNN) proposed components. To enhance scratches, local maximum index (LMI) module...
The detection of conductive particle images is an important part the chip on glass (COG) process and can be used to ensure performance electrical connections. segmentation particles essential but a difficult task, since scale edge imaging effect are different. In recent years, methods based deep learning have become representative method image segmentation. However, currently existing cannot fully consider characteristics high model complexity. this article, multi-frequency feature...
High-cycle fatigue tests of Pt–Al bond coated and bare single crystal superalloys were conducted at 700 °C 800 °C, aiming investigating the effect coat on high-cycle properties a superalloy. The impact was studied merely tensile creep this superalloy, but not found behavior alloy. experimental results showed that to be beneficial or under low stresses owing fact cracks retarded by fine grains in IDZ, P-type rafted γ′ phases substrate coat/substrate interface, respectively. However,...
The design of a high-precision robot assembly system is great challenge. In this article, robotic developed to assemble two components with six degree-of-freedoms in three-dimensional space. It consists manipulators, structured light camera which mounted on the end-effector aside component A measure pose B. Firstly, features irregular are extracted based U-NET network training few labeled images. Secondly, an algorithm proposed calculate B image and corresponding coordinates its ellipse...