4.5 Article

A weld feature points detection method based on improved YOLO for welding robots in strong noise environment

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 Engineering, Electrical & Electronic

Image Denoising of Seam Images With Deep Learning for Laser Vision Seam Tracking

Lei Yang et al.

Summary: This paper proposes a novel image denoising method for automatic laser stripe extraction in intelligent robot welding applications. By introducing deep convolutional neural networks and attention mechanism to extract and accumulate features, and using bi-directional ConvLSTM blocks to learn multi-scale and long-range contexts, the proposed method achieves high detection precision and good robustness against strong image noise interference.

IEEE SENSORS JOURNAL (2022)

Article Engineering, Mechanical

Robotic seam tracking system combining convolution filter and deep reinforcement learning

Yanbiao Zou et al.

Summary: In this paper, a robust and accurate seam tracking system is designed using a six-axis welding robot, a line laser sensor, and an industrial computer. By combining convolution filter and deep reinforcement learning, a two-stage weld feature point localization method is proposed to accurately determine the weld feature points in real time during welding. The experiments conducted show that the system has a mean tracking error of 0.189 mm, demonstrating its ability to meet actual welding requirements with smooth torch movement and minimal interference.

MECHANICAL SYSTEMS AND SIGNAL PROCESSING (2022)

Article Engineering, Electrical & Electronic

Lane Detection Based on Two-Stage Noise Features Filtering and Clustering

Yuanlong Wang et al.

Summary: This paper presents a new multi-lane detection method that utilizes an object detection neural network to remove vehicles from road images and accurately extract lane line features. The method utilizes binarization, object detection, and clustering to perform lane detection and demonstrates good real-time performance and robustness in various complex scenarios.

IEEE SENSORS JOURNAL (2022)

Article Engineering, Electrical & Electronic

Automatic Identification of Multi-Type Weld Seam Based on Vision Sensor With Silhouette-Mapping

Yingzhong Tian et al.

Summary: This study uses silhouette-mapping as an intermediary for weld seam identification and introduces a multi-type weld seam automatic identification system based on vision sensor, showing improved accuracy and robustness in weld seam identification.

IEEE SENSORS JOURNAL (2021)

Article Computer Science, Artificial Intelligence

AP-Loss for Accurate One-Stage Object Detection

Kean Chen et al.

Summary: This paper introduces a novel framework using average-precision loss for ranking task in one-stage object detectors, addressing the extreme foreground-background class imbalance issue. Experimental results demonstrate significant improvement in solving the imbalance problem and achieving outstanding performance on various standard benchmarks.

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE (2021)

Article Environmental Sciences

A Real-Time Apple Targets Detection Method for Picking Robot Based on Improved YOLOv5

Bin Yan et al.

Summary: This study proposed an improved YOLOv5s algorithm for apple target recognition in apple picking robots. Experimental results show that the algorithm successfully differentiates between occluded graspable and ungraspable apples, with improved performance and speed.

REMOTE SENSING (2021)

Article Automation & Control Systems

Additive seam tracking technology based on laser vision

Zhuang Zhao et al.

Summary: This paper studies the additive weld seam tracking technology based on laser vision and designs a welding seam tracking system, demonstrating high robustness and strong adaptability to meet practical welding requirements.

INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY (2021)

Article Computer Science, Information Systems

Intelligent Guidance Programming of Welding Robot for 3D Curved Welding Seam

Bo Zhou et al.

Summary: This paper proposes a welding seam perception method based on a line structure-light sensor, realizes 3D modeling perception of welding parts through the robot's eye-in-hand system, and solves the problem of offline programming system being unable to be applied. Algorithms for extracting various types of 3D space curve welding seams are proposed, which can extract welds with high precision and resolution and are less affected by the welding environment.

IEEE ACCESS (2021)

Article Engineering, Multidisciplinary

Real-time spatial intersecting seam tracking based on laser vision stereo sensor

Zhenwei Jia et al.

MEASUREMENT (2020)

Article Computer Science, Artificial Intelligence

Squeeze-and-Excitation Networks

Jie Hu et al.

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE (2020)

Article Automation & Control Systems

A weld line detection robot based on structure light for automatic NDT

Zhaoxuan Dong et al.

INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY (2020)

Article Automation & Control Systems

Automatic seam detection and tracking system for robots based on laser vision

Yanbiao Zou et al.

MECHATRONICS (2019)

Article Automation & Control Systems

Strong noise image processing for vision-based seam tracking in robotic gas metal arc welding

Rongqiang Du et al.

INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY (2019)

Article Engineering, Multidisciplinary

A seam tracking system based on a laser vision sensor

Yanbiao Zou et al.

MEASUREMENT (2018)

Article Computer Science, Artificial Intelligence

Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks

Shaoqing Ren et al.

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE (2017)

Proceedings Paper Computer Science, Artificial Intelligence

Xception: Deep Learning with Depthwise Separable Convolutions

Francois Chollet

30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017) (2017)