Abstract
Manual recognition and classification of weld joints in real-time using welding images is idiomatic, takes skill, and might be prejudiced. Also, because most welding robot applications are taught and played, they must be reconfigured each time they engage in a new duty. This takes time, and welding settings must be improvised for each new weld job. Hence, this study addresses these concerns by proposing an alternate way of automatically recognizing weld joint types. This paper suggests an effective way to classify the weld joint type using the feature extraction technique. This research aims to create a fusion model that uses sophisticated Convolutional Neural Networks (CNN) and Support Vector Machines (SVM) to recognize Welding joints from a dataset. The suggested hybrid model incorporates the essential characteristics of both; the CNN and the SVM classifier. In this fusion model, CNN is an automated feature extractor, while SVM serves as a classifier. The model is trained and tested using the Kaggle Weld joint dataset (for Butt and Tee Joint) and an in-house dataset (for Vee and lap weld joint). The collection comprises a variety of weld joint photos captured from various perspectives. CNN’s receptive field aids in the automated extraction of the most distinguishing aspects of these images. The experimental findings show that the suggested framework is successful, with a recognition accuracy of 99.7% over the mentioned dataset. Accuracy is determined using the k-fold cross-validation method, where k = 10.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Reisgen, U., Mann, S., Middeldorf, K., Sharma, R., Buchholz, G., Willms, K.: Connected, digitalized welding production—industrie 4.0 in gas metal arc welding. Welding in the World 63(4), 1121–1131 (2019). https://doi.org/10.1007/s40194-019-00723-2
Mahadevan, R., Jagan, A., Pavithran, L., Shrivastava, A., Selvaraj, S.K.: Intelligent welding by using machine learning techniques. Mater. Today Proc. 46, 74027410 (2021). https://doi.org/10.1016/j.matpr.2020.12.1149
Xiuping, W., Fan, X., Ying, F.: Recognition of the Type of Welding Joint Based on Line Structured Light Vision, pp. 4403–4406 (2015)
Chen, X., Chen, S., Lin, T., Lei, Y.: Practical method to locate the initial weld position using visual technology. Int. J. Adv. Manuf. Technol. 30(7–8), 663–668 (2006). https://doi.org/10.1007/s00170-005-0104-z
Hong, T.S., Ghobakhloo, M., Khaksar, W.: Robotic Welding Technology 6. Elsevier (2014). https://doi.org/10.1016/B978-0-08-096532-1.00604-X
Zhang, Y.M., Feng, Z., Chen, S.: Trends in intelligentizing robotic welding processes. J. Manuf. Process. 63, 1 (2021). https://doi.org/10.1016/j.jmapro.2020.11.012
Zeng, J., Cao, G.Z., Peng, Y.P., Huang, S.D.: A weld joint type identification method for visual sensor based on image features and SVM. Sensors (Switzerland) 20(2), 471 (2020). https://doi.org/10.3390/s20020471
Fan, J., Jing, F., Fang, Z., Tan, M.: Automatic recognition system of welding seam type based on SVM method. Int. J. Advanced Manufacturing Technol. 92(1–4), 989–999 (2017). https://doi.org/10.1007/s00170-017-0202-8
Tian, Y., et al.: Automatic identification of multi-type weld seam based on vision sensor with silhouette-mapping. IEEE Sens. J. 21(4), 5402–5412 (2021). https://doi.org/10.1109/JSEN.2020.3034382
Wang, Z., Jing, F., Fan, J.: Weld seam type recognition system based on structured light vision and ensemble learning. In: Proceedings 2018 IEEE International Conference Mechatronics Autom. ICMA 2018, no. 61573358, pp. 866–871 (2018). https://doi.org/10.1109/ICMA.2018.8484570
Shah, H.N.M., Sulaiman, M., Shukor, A.Z., Kamis, Z., Rahman, A.A.: “Butt welding joints recognition and location identification by using local thresholding,” robot. Comput. Integr. Manuf. 51, 181–188 (2018). https://doi.org/10.1016/j.rcim.2017.12.007
Li, Y., Xu, D., Tan, M.: Welding joints recognition based on Hausdorff distance. Gaojishu Tongxin/Chinese High Technol. Lett. 16(11), 1129–1133 (2006)
Fan, J., Jing, F., Yang, L., Long, T., Tan, M.: A precise seam tracking method for narrow butt seams based on structured light vision sensor. Opt. Laser Technol. 109, 616–626 (2019). https://doi.org/10.1016/j.optlastec.2018.08.047
Zou, Y., Chen, T.: Laser vision seam tracking system based on image processing and continuous convolution operator tracker. Opt. Lasers Eng. 105(January), 141–149 (2018). https://doi.org/10.1016/j.optlaseng.2018.01.008
Chen, S., Liu, J., Chen, B., Suo, X.: Universal fillet weld joint recognition and positioning for robot welding using structured light. Robot. Comput. Integr. Manuf., 74, 102279 (2021). https://doi.org/10.1016/j.rcim.2021.102279
Tang, Y.: Deep Learning using Linear Support Vector Machines (2013). https://doi.org/10.48550/ARXIV.1306.0239
Agarap, A.F.: An Architecture Combining Convolutional Neural Network (CNN) and Support Vector Machine (SVM) for Image Classification, pp. 5–8 (2017). http://arxiv.org/abs/1712.03541
Jiang, S., Hartley, R., Fernando, B.: Kernel support vector machines and convolutional neural networks. 2018 Int. Conf. Digit. Image Comput. Tech. Appl. DICTA, pp. 1–7 (2019). https://doi.org/10.1109/DICTA.2018.8615840
Ahlawat, S., Choudhary, A.: Hybrid CNN-SVM classifier for handwritten digit recognition. Procedia Comput. Sci. 167(2019), 2554–2560 (2020). https://doi.org/10.1016/j.procs.2020.03.309
Kaggle: Rectified Linear Units (ReLU) in Deep Learning. https://www.kaggle.com/dansbecker/rectified-linear-units-relu-in-deep-learning
Hantos, N., Iván, S., Balázs, P., Palágyi, K.: Binary image reconstruction from a small number of projections and the morphological skeleton. Ann. Math. Artif. Intell. 75(1–2), 195–216 (2014). https://doi.org/10.1007/s10472-014-9440-8
MathWorks: ResNet-18 convolutional neural network - MATLAB resnet18 - MathWorks India. https://in.mathworks.com/help/deeplearning/ref/resnet18.html
Cortes, C., Vapnik, V.: Support-vector network. IEEE Expert. Syst. their Appl. 7(5), 63–72 (1992). https://doi.org/10.1109/64.163674
MathWorks: ClassificationECOC. https://in.mathworks.com/help/stats/classificationecoc.html
"Weld-Joint-Segments | Kaggle. https://www.kaggle.com/datasets/derikmunoz/weld-joint-segments Accessed 1 Apr 2022
MathWorks Inc.: Fit multi-class models for support vector machines or other classifiers (2018). https://in.mathworks.com/help/stats/fitcecoc.html Accessed 1 Apr 2022
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Sonwane, S., Chiddarwar, S., Rahul, M.R., Dalvi, M. (2023). Pre-trained CNN Based SVM Classifier for Weld Joint Type Recognition. In: Arai, K. (eds) Proceedings of the Future Technologies Conference (FTC) 2022, Volume 1. FTC 2022 2022. Lecture Notes in Networks and Systems, vol 559. Springer, Cham. https://doi.org/10.1007/978-3-031-18461-1_12
Download citation
DOI: https://doi.org/10.1007/978-3-031-18461-1_12
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-18460-4
Online ISBN: 978-3-031-18461-1
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)