Search

Copyright of Journal of Zhengzhou University (Natural Science Edition) is the property of Journal of Zhengzhou University (Natural Science Edition) Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

In this paper, a dual-camera-guided weld start point identification and weld tracking method based on K-means passive vision is proposed to improve the problem that structured light visual guidance is not applicable in stir friction welding due to the severe reflection on the surface of narrow weld aluminum alloy. This method is mainly applied to narrow weld seam stir friction welding where no bright light or smoke is generated, because it has many advantages over methods that use structured light to guide the welding process: the ability to extract more information about the weld seam and the ability to identify narrower weld seams. By pre-processing operations such as noise reduction and filtering of the target image, K-means is used to obtain information on the start point of the weld seam. The weld seam is tracked in real time after the coordinates are converted and the device is controlled to reach the target point. The experimental results show that the method can effectively identify the start point and path of the weld, and has high accuracy for tracking complex narrow weld seams, and has great potential to improve production efficiency. [ABSTRACT FROM AUTHOR]

This work addresses the online extraction of the geometry characteristics (width and centerline) of deposited beads with monocular cameras for wire arc additive manufacturing (WAAM). To enable online measurement and feature extraction from captured images, an adaptive threshold is used for segmentation, a Canny algorithm for edge detection, a Hough-line transform for feature identification of the bead edges, and a filtering step to attenuate the low signal-to-noise ratios of deposition processes. Online measurements are performed in single-bead and layer (multi-bead) scenarios. The proposed vision-based solution is experimentally implemented in a WAAM robotic system composed of a welding torch, a Kuka KR90 robot arm, a power source, wire feeder, and a passive monocular camera. Experimental results illustrate the performance and effectiveness of the proposed visual-based methodology. [ABSTRACT FROM AUTHOR]

Vision-based welding seam tracking is an important and unique branch of welding automation. Active vision seam tracking systems achieve accurate feature extraction by using an auxiliary light source, but this will introduce extra costs and the real-time performance will be affected. In contrast, passive vision systems achieve better real-time performance and their structure is relatively simple. This paper proposes a passive vision welding seam tracking system in Plasma Arc Welding (PAW) based on semantic segmentation. The BiseNetV2 network is adopted in this paper and online hard example mining (OHEM) is used to improve the segmentation effect. This network structure is a lightweight structure allowing effective image feature extraction. According to the segmentation results, the offset between the welding seam and the welding torch can be calculated. The results of the experiments show that the proposed method can achieve 57 FPS and the average error of the offset calculation is within 0.07 mm, meaning it can be used for real-time seam tracking. [ABSTRACT FROM AUTHOR]

Seam tracking is of great importance in automatic welding. Currently, the laser tracking algorithm has reached an advanced stage of development, with several products available in the market. However, it is expensive to deploy a laser sensor and unsuitable for none-groove butt welds with narrow gap. By contrast, passive vision-based tracking algorithm can retrench the spending and is suitable for none-groove seam, but unlike laser tracking algorithm which can project artificial feature onto weldments, passive vision can only extract the features naturally generated by welding environment, which brings a great challenge to its accuracy and stability. In this paper, to tackle the problems of passive vision-based seam tracking caused by different groove types, an interactive segmentation network is used to locate the welding seam area. Compared to normal segmentation network, interactive segmentation can be fed by artificial information before welding, so it is more tolerant to different types of groove(0.108 mIoU better than normal segmentation network). Then a shortest path planning algorithm with a mixture of Hessian and Canny gradient cost map is proposed to extract the seam edge. The proposed algorithm is successfully verified stable and accurate by experiments of different groove type, and the mean absolute error is 0.25mm and the maximum standard deviation is 0.28mm. • The factors that affect the accuracy and stability of passive vision seam tracking are analyzed. • An interactive segmentation network is used in welding seam recognition. • A dynamic programming-based welding seam recognition algorithm is designed. [ABSTRACT FROM AUTHOR]

In this research, a moving target position measurement system based on vision sensors is developed, which can be used for remote monitoring of UAV refueling docking under complex environment. The developed system uses two vision sensors and embedded devices placed on a fixed point of the UAV to realize a synchronous acquisition of given drogue images, and transmits it to the computer to calculate the drogue position in real time. In this paper, an adaptive processing algorithm for abnormal regions based on stereo vision is proposed, which can determine the illumination inconsistent regions that affect the matching in the image pair according to the statistical distribution. A fast stereo rectification algorithm for image pair is proposed to deal with sensor micro jitter. A multi-cost fusion robust matching algorithm based on “ship” search is also proposed to further improve the position measurement accuracy. The robustness and effectiveness of the measurement system in moving target remote measurements are verified by simulation flight and field flight tests. [ABSTRACT FROM AUTHOR]

Vision-based welding seam tracking is an important and unique branch of welding automation. Active vision seam tracking systems achieve accurate feature extraction by using an auxiliary light source, but this will introduce extra costs and the real-time performance will be affected. In contrast, passive vision systems achieve better real-time performance and their structure is relatively simple. This paper proposes a passive vision welding seam tracking system in Plasma Arc Welding (PAW) based on semantic segmentation. The BiseNetV2 network is adopted in this paper and online hard example mining (OHEM) is used to improve the segmentation effect. This network structure is a lightweight structure allowing effective image feature extraction. According to the segmentation results, the offset between the welding seam and the welding torch can be calculated. The results of the experiments show that the proposed method can achieve 57 FPS and the average error of the offset calculation is within 0.07 mm, meaning it can be used for real-time seam tracking.

We present a literature review to analyze the state of the art in the area of road detection based upon frontal images. For this purpose, a systematic literature review (SLR) was conducted that focuses on analyzing region-based works, since they can adapt to different surface types and do not depend on road geometry or lane markings. Through the comprehensive study of publications in a 11-year time frame, we analyze the methods that are being used, on which types of surface they are applied, whether they are adaptive in relation to surface changes, and whether they are able to distinguish possible faults or changes in the road, such as potholes, shadows, and puddles. [ABSTRACT FROM AUTHOR]

In recent years, passive vision-based methods in the weld seam tracking system are becoming more and more popular due to their higher accuracy and robustness to various interferences. In this paper, we present a novel robust weld seam tracking system based on passive vision. A robust discriminative correlation filter is proposed to locate the seam accurately by using weld seam images and background images as positive and negative samples,respectively. Additionally, a new online updating filter strategy is proposed to reduce the effect of varying illumination or changing structure and appearance. The two-scale searching window mode and the fast Fourier transform are combined to increase the speed and precision of the weld seam tracking. The extensive experimental results indicate that the proposed method is simple and fast, and also, its accuracy is high in tracking weld seams. The proposed system has been applied successfully to industrial production lines. [ABSTRACT FROM AUTHOR]

A passive vision system based on image processing technology for automatic welding system is researched in order to enhance the efficiency for the tube-to-tubesheet welding of heat exchanger in this paper. To this end, the detection and extraction methods of the welding path are studied in tubesheet welding through the passive vision system. A calibration procedure is proposed for the vision system, and the positional relationships between the cameras and the robot are obtained. The Hough transform is used to extract the tubes' position roughly through a camera that is fixed above the workpiece, and another camera is used to extract the feature of the tubes accurately one by one. A multiple iteration method of least squares is also proposed in this paper to remove the background noises. The edge of the inside tube circle is accurately extracted, and the welding path is obtained. Validation experiments are finally conducted to confirm the reliability and accuracy of this method. [ABSTRACT FROM AUTHOR]

Associated to the weld quality, the weld bead geometry is one of the most important parameters in welding processes. It is a significant requirement in a welding project, especially in automatic welding systems where a specific width, height, or penetration of weld bead is needed. This paper presents a novel technique for real-time measuring of the width and height of weld beads in gas metal arc welding (GMAW) using a single high-speed camera and a long-pass optical filter in a passive vision system. The measuring method is based on digital image processing techniques and the image calibration process is based on projective transformations. The measurement process takes less than 3 milliseconds per image, which allows a transfer rate of more than 300 frames per second. The proposed methodology can be used in any metal transfer mode of a gas metal arc welding process and does not have occlusion problems. The responses of the measurement system, presented here, are in a good agreement with off-line data collected by a common laser-based 3D scanner. Each measurement is compare using a statistical Welch’s t-test of the null hypothesis, which, in any case, does not exceed the threshold of significance level α = 0.01, validating the results and the performance of the proposed vision system. [ABSTRACT FROM AUTHOR]

Abstract: A passive vision sensor system has been developed to monitor the nozzle to the top surface distance (NTSD) in layer additive manufacturing (LAM) using gas metal arc welding (GMAW). The deviations in the NTSD are compensated by the movement of the working flat, and the adjustment of the deposition rate on next deposition layer. After simplification of the controlled process into a linear system, an adaptive control system has been designed to keep the NTSD constant. The effectiveness of the controller is evaluated through deposition of single-bead multi-layer walls, and the experimental results confirm that the process stability can be improved when applying the developed controller. [Copyright &y& Elsevier]

Welding robots have been widely used in manufacturing process to substitute for human welders. However, most of them are rigid and cannot adjust to variations in the weld seam positions caused by natural welding environmental factors. To address this problem, this paper presents a passive vision-based robotic welding system, which can realize the seam tracking function for pulse-MAG welding. In this paper, the light spectrum of the welding process is analyzed to determine the optical filters used during the image capture. Then, a robust image processing method is proposed to extract the offset from the image which contains much noise. The transformation formula is calibrated to obtain the relationship between the image coordinate system and the robot coordinate system. The tracking strategy is designed to improve the tracking precision and the stability of the welding process. Finally, experiments are conducted on straight line and curved line seam to verify the feasibility of the developed system. [ABSTRACT FROM AUTHOR]

Acquire spatial information of a 3D model is very useful in many applications in the computer science field. A device broadly used as acquisition hardware of models is the 3D scanner. With a 3D scanner is possible to obtain the geometric structure based on points and triangles of a model, and use it on several computing applications. According the hardware and algorithms employed, the scanners can be classified as passive and active vision. The passive vision scanners are lower cost devices which require geometrics algorithms to obtain a 3D reconstruction. In this paper, we present a design and implementation of a simple 3D scanner of lower cost based on passive vision. The scanner consists in the design of hardware and algorithms where two webcam are placed on a controlled environment. Several experiments demonstrate excellent results on the geometric reconstruction of the scanned objects. [ABSTRACT FROM PUBLISHER]

A passive vision sensor is added to a teaching and playback welding robot for pulsed gas tungsten arc welding, which is normally used in the welding of thin plate closed-gap butts. This paper presents a seam tracking method based on this sensor through period visual measurement of the offset between the torch and the seam center. A robust image processing algorithm is developed to extract the seam center. A kind of ARX(auto-regressive with exogenous input) model is studied to describe the relationship between the rectifying voltage and the offset. The fuzzy PID(proportional-integral-derivative) seam tracking controller is analyzed and designed in consideration of various offsets based on the ARX model. The experimental results on straight line weld display a good seam tracking capability of the proposed method. [ABSTRACT FROM AUTHOR]

The automatic control in square-wave alternating current (AC) gas tungsten arc welding (GTAW) is significant for a “teach and playback” robot to overcome the variation of the seam trajectory and the seam gap in the welding process. This paper presents a welding robot system based on the real-time visual measurement in the different levels of the welding current. The primary objective is to measure the offset of the torch to the seam center and the size of seam gap by passive vision, track the seam and control the weld pool in real time. A novel visual image analysis algorithm was developed for seam tracking and seam gap measuring, free from robot calibration. The control algorithm based on the knowledge base was established to control the weld formation by regulating the welding current and wire feed rate. The welding practice for the rocket storage tank demonstrates the efficiency of the proposed system. [ABSTRACT FROM AUTHOR]