Defect detection of steel surface using entropy segmentation

Abstract

Today, because of high manufacturing speed in steel industry, there is a need of fast and accurate detection of steel defect for quality assurance of product. Unlike other papers on defect detection of steel surface based on entropy this paper presents a new pre-processing and processing algorithm. The method presented here overcomes the limitations of traditional segmentation method or adaptive segmentation method like Otsu's method. This paper presents a new defect detection algorithm based on entropy. As a pre-processing step illumination compensation of image has been introduced using inverse illumination to remove non-uniformity of light intensity in the image. In the second part Local entropy of image has been used to detect the region of defect. The paper also suggests the concept of dynamic updation which helps to find a good background i.e. ideal steel surface and provides an effective method to classify the defects in its initial stage into defective and non-defective image. Background subtraction method is then used to extract the defective portion of image from the entropy image by comparing the entropy of image with the entropy of background image. Histogram thresholding method has been introduced to separate the background and defective portion in the background subtracted image to get the segmented image. The method was successfully tested on three kinds of defect on steel surface i.e. water droplet, blister and scratch.Index Termssteel defect, entropy, image processing.

Automated Visual Defect Detection for Flat Steel Surface: A Survey

“…The dynamical thresholding procedure can then discriminately separate true defects from random noise. Further, Nand et al [22] calculated the local entropy of defective and defect -free images respectively and extracted defective region of image by using background subtraction method by comparing their entropy, it is reported to perform better on detecting defective blocks of Fig. 3.…”

Uneven Illumination Surface Defects Inspection Based on Saliency Detection and Intrinsic Image Decomposition

“…The second term is an L1-norm regularization term that guarantees the smoothness constrain of the shading image. B represents the background map weight, which can be obtained through the saliency map as B = 1 − D, The traditional fidelity term 1 2 (S − I) 2 2 , which constrains the similarity of the whole image region. We introduce background similarity prior into fidelity terms.…”

“…Directly calculating the matrices F ∇ T ∇ is computationally expensive. According to [1], the matrix F ∇ T ∇ is equivalent to 2 cos(2πu/p) + 2cos(2πv/q) − 4, where m and n represent the image width and height, respectively, and u ∈ [0, p) and v ∈ [0, q) are the frequencies in the frequency domain.…”

“…The statistical-based approaches are usually applied to detect defects in images with uneven illumination by utilizing the histogram and texture statistics information. Nand et al [1] compare the entropy difference between defective images and defect-free images to identify the defective regions of steel images. Doo-Chul et al [2] adopt morphological features to detect pinhole defects of steel slab images.…”

An Efficient Quality Control System by Machine Learning for Surface Defects

“…AN entropy segmentation approach is discussed in [10] for steel surface defects detection. The non-uniformity in the input image is removed at first and teh local entropy feature is utilized for the segmentation of defect region.…”