AN INDUSTRIAL INSPECTION APPROACH FOR WELD DEFECTS USING MACHINE LEARNING ALGORITHM

John Aravindar D,Kalai Selvi V

doi:10.29284/ijasis.5.1.2019.15-21

Abstract

The weld defects are formed due to the incorrect welding patterns or wrong welding process. The defects in the weld may vary from size, shape and their projected quality. The most common weld defects occur during welding process is slag inclusions, porosity, lack of fusion and incomplete penetration. In this study, an effective method for weld defect classification using machine learning algorithm is presented. The system uses Speeded-up Robust Features (SURF) for feature extraction and one of the machine learning algorithms called Auto-Encoder Classifier (AEC) for classification. Initially, the features that distinguish weld defects and no defects in the weld image are extracted by SURF. Then, AEC is analyzed for weld image classification using different number of neurons in different hidden layers (2 and 3 hidden layers). The performance of the system is evaluated by GD X-ray weld image database. The results show that the weld images are correctly classified with 98% accuracy using SURF and AEC.

Highlights

Welding image classification in low rank representation using spatial pyramid matching is discussed in [1]
The GD X-ray image database is used for weld image classification
Features are extracted by using Speeded-up Robust Features (SURF) technique which is a local descriptor

Summary

Introduction

Welding image classification in low rank representation using spatial pyramid matching is discussed in [1]. The input weld image features are extracted by Scale Invariant Feature Transform (SIFT). The encoding is made by vector quantization, sparse representation and low rank representation. The classification is made by linear and non-linear Support Vector Machine (SVM). Gaussian kernel and Convolutional Neural Network (CNN) based welding defect classification is discussed in [2]. Gaussian kernel is used to blur the image. Features are extracted and classified by CNN in two stages

Methods

Results

Conclusion