R Deivanathan

Bio: R Deivanathan is an academic researcher. The author has contributed to research in topics: Welding & Welding defect. The author has co-authored 1 publications.

Welding Defect Identification with Machine Vision System using Machine Learning

Abstract: Friction stir welding is a solid-state joining process to join similar or dissimilar metals, which uses the friction developed between the metals to join them. Friction stir welding is an efficient way to join the metals but the welding defects are a little difficult to find through naked eyes. Hence, there is a chance of it being unnoticed even in final inspection in industries. So the welded joints are inspected by machine vision, using camera and an intelligent system. After obtaining the images of the welded joints and processing them using MATLAB, they are differentiated according to the various defects in them, using machine learning technique. For this, the statistical features of the image are extracted and they are classified into different defects using classifiers like Decision Trees, Discriminant Analysis, Support Vector Machine and Nearest Neighbour.

Inline Defective Laser Weld Identification by Processing Thermal Image Sequences with Machine and Deep Learning Techniques

Abstract: The non-destructive testing methods offer great benefit in detecting and classifying the weld defects. Among these, infrared (IR) thermography stands out in the inspection, characterization, and analysis of the defects from the camera image sequences, particularly with the recent advent of deep learning. However, in IR, the defect classification becomes a cumbersome task because of the exposure to the inconsistent and unbalanced heat source, which requires additional supervision. In light of this, authors present a fully automated system capable of detecting defective welds according to the electrical resistance properties in the inline mode. The welding process is captured by an IR camera that generates a video sequence. A set of features extracted by such video feeds supervised machine learning and deep learning algorithms in order to build an industrial diagnostic framework for weld defect detection. The experimental study validates the aptitude of a customized convolutional neural network architecture to classify the malfunctioning weld joints with mean accuracy of 99% and median f1 score of 73% across five-fold cross validation on our locally acquired real world dataset. The outcome encourages the integration of thermographic-based quality control frameworks in all applications where fast and accurate recognition and safety assurance are crucial industrial requirements across the production line.

Visual Control of Weld Defects Using Computer Vision System on FANUC Robot

Abstract: The development of Industry 4.0 technology requires the automation of technological processes. Weld quality inspections are carried out predominantly manually and can combine different types of non-destructive testing depending on the structure's intended use. It significantly slows down the manufacturing process and negates the effect of robotized welding production. At the same time, when welding on a welding robot, it is possible to use the built-in vision system, which currently in its default configuration is not initially provided to inspect welds after welding. Typically, these systems are integrated with robot arms that locate the desired parts in a box or crate and feed them to the actuators, orienting them correctly for the task at hand. It should be noted that for visual inspection of welds, external specialized computer vision systems with pattern recognition systems are usually used. Thus, the actual task is to investigate the possibility of visual inspection of weld defects specific to surface defects such as pore, crack and undercut, using the integrated FANUC iRVision 3DL system. The visual inspection of weld defects using the iRVision 3DL computer vision system installed on the FANUC robot was established. The system has the highest sensitivity to pore-type defects. The developed technique allows the quality control of the weld in an automated mode. In this case, the quality assessment should be carried out primarily using a 3D Laser Vision Sensor and a photo of the weld.

Visual Control of Weld Defects Using Computer Vision System on FANUC Robot

Abstract: The development of Industry 4.0 technology requires the automation of technological processes. Weld quality inspections are carried out predominantly manually and can combine different types of non-destructive testing depending on the structure's intended use. It significantly slows down the manufacturing process and negates the effect of robotized welding production. At the same time, when welding on a welding robot, it is possible to use the built-in vision system, which currently in its default configuration is not initially provided to inspect welds after welding. Typically, these systems are integrated with robot arms that locate the desired parts in a box or crate and feed them to the actuators, orienting them correctly for the task at hand. It should be noted that for visual inspection of welds, external specialized computer vision systems with pattern recognition systems are usually used. Thus, the actual task is to investigate the possibility of visual inspection of weld defects specific to surface defects such as pore, crack and undercut, using the integrated FANUC iRVision 3DL system. The visual inspection of weld defects using the iRVision 3DL computer vision system installed on the FANUC robot was established. The system has the highest sensitivity to pore-type defects. The developed technique allows the quality control of the weld in an automated mode. In this case, the quality assessment should be carried out primarily using a 3D Laser Vision Sensor and a photo of the weld.

Algorithms for Vision-Based Quality Control of Circularly Symmetric Components

Abstract: Quality inspection in the industrial production field is experiencing a strong technological development that benefits from the combination of vision-based techniques with artificial intelligence algorithms. This paper initially addresses the problem of defect identification for circularly symmetric mechanical components, characterized by the presence of periodic elements. In the specific case of knurled washers, we compare the performances of a standard algorithm for the analysis of grey-scale image with a Deep Learning (DL) approach. The standard algorithm is based on the extraction of pseudo-signals derived from the conversion of the grey scale image of concentric annuli. In the DL approach, the component inspection is shifted from the entire sample to specific areas repeated along the object profile where the defect may occur. The standard algorithm provides better results in terms of accuracy and computational time with respect to the DL approach. Nevertheless, DL reaches accuracy higher than 99% when performance is evaluated targeting the identification of damaged teeth. The possibility of extending the methods and the results to other circularly symmetrical components is analyzed and discussed.

Development of the automatic method of detection and grouping of external welding imperfections

Abstract: The non-destructive testing of welded joints is a key issue in welding technology and processes. It is especially crucial for estimation of product quality. Among others methods, visual testing is the most fundamental. In most cases, it is made manually, which can introduce some problems resulting from the lack of objectivity and fatigue of the controlling person. This paper is an attempt of changing a mode of visual inspection. The proposed system is based on image analysis of welded joints. The welding imperfections were correlated with quality levels according to the ISO 5817 standard. For the original inspection system, six the most fundamental imperfections have been selected. The main idea of this system is a detection of selected welding imperfections and, based on acceptance criteria, assigning welded joints for the appropriate quality level. In order to validate the correct operation of the proposed system, the same welded joints have been subjected to conventional manual visual inspection. The compliance of the results was more than 90%, and the speed of visual examinations was more than 10 times higher than in the manual method.