<abstract xmlns="http://www.w3.org/1999/xhtml">

<p>The Friction Stir Welding process usually produces weld members of good quality compared to composite weld made with a standard welding process. However, there is a possibility of the formation of various defects if the input parameters are not properly selected. In the recent case study, an image-based feature recognition system using the Fourier conversion method which is a computer visual recognition tool is developed. Five types of filters like Ideal Filter, Butterworth Filter, Low Filter, Gaussian Filter, and High Pass Filter. The results showed that the high pass filter has more ability to detect surface defects compared to the other four filters. It has also been observed that the Ideal filter has a lot of distortions compared to the Gaussian Filter and the Butterworth Filter.</p>
</abstract>

The Friction Stir Welding process usually produces weld members of good quality compared to composite weld made with a standard welding process. However, there is a possibility of the formation of various defects if the input parameters are not properly selected. In the recent case study, an image-based feature recognition system using the Fourier conversion method which is a computer visual recognition tool is developed. Five types of filters like Ideal Filter, Butterworth Filter, Low Filter, Gaussian Filter, and High Pass Filter. The results showed that the high pass filter has more ability to detect surface defects compared to the other four filters. It has also been observed that the Ideal filter has a lot of distortions compared to the Gaussian Filter and the Butterworth Filter.

Machine Learning Algorithm for Surface Quality Analysis of Friction Stir Welded Joint

Politecnico Di Milano, Department of Mechanical Engineering

Strojnícky časopis - Journal of Mechanical Engineering

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

The Friction Stir Welding process usually produces weld members of good quality compared to composite weld made with a standard welding process. However, there...

Machine Learning Algorithm for Surface Quality Analysis of Friction Stir Welded Joint

Data publikacji: 12 gru 2020

Zakres stron: 11 - 20

DOI: https://doi.org/10.2478/scjme-2020-0016

Słowa kluczowe
Machine Learning, Machine Vision, Friction Stir Welding, Surface Defects

© 2020 Akshansh Mishra, published by Sciendo

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

Machine Learning Algorithm for Surface Quality Analysis of Friction Stir Welded Joint

Data publikacji: 12 gru 2020

Zakres stron: 11 - 20

DOI: https://doi.org/10.2478/scjme-2020-0016

Słowa kluczoweMachine Learning, Machine Vision, Friction Stir Welding, Surface Defects

© 2020 Akshansh Mishra, published by Sciendo

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

Słowa kluczowe
Machine Learning, Machine Vision, Friction Stir Welding, Surface Defects