Magnetic imaging and machine vision NDT for the on-line inspection of stainless steel strips

Abstract

An on-line inspection system for stainless steel strips has been developed on an annealing and pickling line at the Acciai Speciali Terni S.p.A. steel mill. Besides a machine vision apparatus, the system contextually exploits a magnetic imaging system designed and realized for the specific application. The main goal of the research is represented by the fusion of the information provided by the two apparatuses that can improve the detection and classification tasks by enlarging the set of detectable defects. In this paper, the development, the calibration and the characteristics of the magnetic imaging apparatus are detailed and experimental results obtained both in laboratory and in situ are reported. A comparative analysis of the performances of the two devices is also reported based on preliminary results and some conclusions and perspectives are drawn.

Research Progress of Automated Visual Surface Defect Detection for Industrial Metal Planar Materials

“…Consequently, the non-contact and high-security advantages of reflection-type measurement turn into the choice of most people, and outstanding results have been obtained in the surface detection of 3D objects. Non-optical measurement is a type of reflection measurement, from early radar and sonar to ultrasonic imaging [ 129 , 130 ], magnetic imaging [ 131 , 132 ], pulsed eddy current imaging [ 133 , 134 , 135 , 136 ], and so on. Non-optical measurement needs to be close to the detected surface and generates blind spots if the rough surface or noise interference occurs.…”

Automated Visual Defect Classification for Flat Steel Surface: A Survey

“…To reduce the impact of the noise on the obtained flat steel surface images, filters such as the median filter [46][47][48] (for impulsive noise), Gaussian smoothing filter [49][50][51] (for Gaussian white noise), mean filter [52], bilateral filter [45,53,54] and Wiener filter [55] [56] used a median filter to smooth the casting images first and then applied Wiener filter to improve these images and reduce the noise effects. Chu et al [39,57] improved the median filter and bilateral filter to remove salt-and-pepper noise and Gaussian noise, respectively.…”

“…The two basic image enhancement methods are based on spatial domain and frequency domain, respectively. For spatial domain methods, histogram equalization makes the image clearer by equalizing the histogram of the original image [55,61,62]. In addition, the grayscale intensity transformation enhances the image contrast by changing (i.e., stretching, compressing or transforming) the grayscale dynamic range of the image [18,48,63].…”

“…Otsu is a classical adaptive threshold method for separating defects from the background in flat steel images [59,69,70], which obtains a threshold value based on the characteristic of the large variance between the background and foreground. Different from the threshold methods, the methods based on edge detection use the first or second derivative to detect edge points by taking advantage of the property of discontinuous pixel values in adjacent regions, such as Robert [71], Sobel [72,73], Prewitt [74], Canny [55] and Kirsch [52,53]. The grayscale of steel strip images is ordinarily nonuniform, the gray value variation cross the background and the defect is sometimes gradual, and the size of defect area is very small, not easy to be recognized by the computer.…”

Research on Application of Wax Deposition Detection in the Nonmetallic Pipeline Based on Electrical Capacitance Tomography

“…With consideration to the operating environment of the pipeline, as well as the cost of detection, for practical application, nondestructivebased techniques are highly recommended. Many nondestructive techniques have achieved some degree of success in the pipeline detection field, for instance, the eddy current testing method [2,3], ultrasonic-based technique [4][5][6][7][8][9][10], radiological image method [11], and magnetic flux leakage detection method [12][13][14]. However, most of these techniques are applied to detect pipeline defects, and it is difficult to detect the sediment (i.e., paraffin wax) of the pipeline by using the above technologies, due to various reasons.…”