Application of Shearlet transform to classification of surface defects for metals

Investigation of Different Algorithms for Surface Defects of Steel Sheet for Quality

“…This is performed through texture analysis using Neural Network and is increasing as a modeling tool [2,16]. Other algorithms have been applied for this purpose such as Gabor filters [13], wavelet packet transform [14] and Shearlet conversion employed by Shunhua Liu and his colleagues for classifying surface defects of metals [1]. Given that investigating metals surface automatically is a known issue which is being investigated, but there isn't a general method to detect failures automatically [12].…”

“…Textural defects like folding of sheet steel and dimensional defects like being convex or concave. In Figure 1, some of these defects have been shown [1]. In this paper different methods of classifying defects have been expressed.…”

“…It is a segmentation algorithm and quick search of defects and also has been used for diagnosing thin cracks and corners of the sheet by minimizing costs but isn't very sensitive to fault detection [1,11].…”

“…Therefore, product quality assurance is considered as the most significant industrial centers [6]. In recent decades, increased demand for higher quality product has caused to provide an optimal and efficient method for metal products which is one of the most important issues in international communities, and by early detection and accurate identification defects of steel sheets which is an important step in quality control produce higher quality products in a shorter time [1,2]. And in terms of steel industry to be able to compete with its competitors in the global markets given the recent developments, efficiency in the production of crude steel must be increased and given that visual inspection is time-consuming and does not have enough accuracy, it has led to apply automatic detection and classification method to minimize returned wastes caused by defects in products [3,4].…”

Learning a Class-specific and Shared Dictionary for Classifying Surface Defects of Steel Sheet

“…Most of existing works have focused on these two aspects to enhance the classification performance over the past few years. [9][10][11][12][13][14][15][16][17][18][19][20] A set of features are studied, including grayscale features, 9,10) geometric features, 9,10) shape features, [9][10][11] texture features, 10,12) Gabor filter output features, 13) Fourier spectral, 14) wavelet transform, [14][15][16][17] local binary pattern (LBP), 18,19) shearlet transform; 20) and the classifier of artificial neural network (ANN) 12,13,15) and support vector machine (SVM) 9,10,14,[17][18][19][20] is discussed. However, these approaches, which treat feature extraction and classifier training as two separating steps, are difficult to control the interaction of two steps.…”