Jong Pil Yun

Real-time vision-based defect inspection for high-speed steel products

In the steel-making industry, both the quality and quantity of the products are critical. This work presents a real-time defect detection method for high-speed steel bar in coil (BIC). For good performance characteristics, the detection algorithm must be robust to problems associated with the cylindrical shape of BICs, the presence of noise and nonuniform brightness distribution of images, the various types of defects, and so on. Furthermore, because the target speed is very high, it should have a fast processing time. Therefore, a defect detection algorithm should satisfy the two conflicting requirements of reducing the processing time and improving the efficiency of defect detection. This work proposes an effective real-time defect detection algorithm that can satisfy these conditions. Moreover, to reduce cost, the proposed algorithm is implemented on a PC-based real-time defect detection system without a professional digital signal processing (DSP) board. Experimental results show that the proposed algorithm guarantees both real-time processing and accurate detection.

Vision-based defect detection of scale-covered steel billet surfaces

Vision-based inspection systems have been widely investigated for the detection and classification of defects in various industrial product. We present a new defect detection algorithm for scale-covered steel billet surfaces. Because of the availability of various kinds of steel, presence of scales, and manufacturing conditions, the features of billet surface images are not uniform. In particular, scales severely change the properties of defect-free surfaces. Moreover, the various kinds of possible defects make their detection difficult. In order to resolve these problems and to improve the detection performance, two methods are proposed. First, undecimated wavelet transform and vertical projection profile are presented. Second, a method for detecting the variations in the block features along the vertical direction is proposed. The former method can effectively detect vertical line defects, and the latter can efficiently detect the remaining defects, except the vertical line defects. The experimental results conducted on billet surface images obtained from actual steel production lines show that the proposed algorithm is effective for defect detection of scale-covered steel billet surfaces.

Torque Ripples Minimization in PMSM using Variable Step-Size Normalized Iterative Learning Control

Periodic torque ripples exist due to non-perfect sinusoidal flux distribution, cogging torque and current measurement errors in permanent magnet synchronous motor (PMSM). These ripples are reflected as periodic oscillations in the motor speed and deteriorate the performance of application of PMSM as a high-precision tracking applications. In this paper, we propose a variable step-size normalized iterative learning control (VSS-NILC) scheme to reduce periodic torque ripples. VSS-NILC is combined to existing PI current controller and generates compensated reference current iteratively from cycle to cycle so as to minimize the mean square torque error. VSS-NILC scheme alters the step-size of the update equation to reduce the conflict between speed of convergence and minimum mean square error (MSE). Consequently VSS-NILC scheme has faster convergence rate and lower mean square torque error. Simulation results show significant improvements in the steady-state torque response and the effectiveness in minimizing torque ripples

Development of Real-time Defect Detection Algorithm for High-speed Steel Bar in Coil(BIC)

In steel manufacturing industry, as many advanced technologies increase manufacturing speed, fast and exact products inspection gets more important. This paper deals with a real-time defect detection algorithm for high-speed steel bar in coil (BIC). To get good performance, this algorithm has to solve several difficult problems such as cylindrical shape of a BIC, influence of light, many kinds of defects. Additionally, it should process quickly the large volumes of image for real-time processing since a steel bar moves at high speed. Therefore defect detection algorithm should satisfy two conflicting requirements of reducing the processing time and improving the efficiency of defect detection. This paper proposes an effective real-time defect detection algorithm that can solve above problems. And the algorithm is implemented by a high speed image processing system and will be applied to a practical manufacturing line. Finally, the performance of the proposed algorithm is demonstrated by experiment results

Defect detection for corner cracks in steel billets using a wavelet reconstruction method

Presently, automatic inspection algorithms are widely used to ensure high-quality products and achieve high productivity in the steelmaking industry. In this paper, we propose a vision-based method for detecting corner cracks on the surface of steel billets. Because of the presence of scales composed of oxidized substances, the billet surfaces are not uniform and vary considerably with the lighting conditions. To minimize the influence of scales and improve the accuracy of detection, a detection method based on a visual inspection algorithm is proposed. Wavelet reconstruction is used to reduce the effect of scales. Texture and morphological features are used to identify the corner cracks among the defective candidates. Finally, the experimental results show that the proposed algorithm is effective in detecting corner cracks on the surfaces of the steel billets.

Pinhole detection in steel slab images using Gabor filter and morphological features

Presently, product inspection for quality control is becoming an important part in the steel manufacturing industry. In this paper, we propose a vision-based method for detection of pinholes in the surface of scarfed slabs. The pinhole is a very tiny defect that is 1-5 mm in diameter. Because the brightness in the surface of a scarfed slab is not uniform and the size of a pinhole is small, it is difficult to detect pinholes. To overcome the above-mentioned difficulties, we propose a new defect detection algorithm using a Gabor filter and morphological features. The Gabor filter was used to extract defective candidates. The morphological features are used to identify the pinholes among the defective candidates. Finally, the experimental results show that the proposed algorithm is effective to detect pinholes in the surface of the scarfed slab.

Detection of line defects in steel billets using undecimated wavelet transform

In this paper, we present a new detection algorithm for line defects of scale-covered steel billets. Because of the presence of scales on the billet surface, features of surface images such as brightness and textures are non-uniform. To minimize the influence of scales and to improve the accuracy of detection, a new detection method based on undecimated wavelet transform is proposed. The vertical projection profile of subimage with high-frequency information produced by undecimated wavelet transform is used to detect the line defects. Experimental results conducted on billets surface image from actual steel production line show that the proposed algorithm is capable of detecting line defects on billet surface.

Localizing slab identification numbers in factory scene images

This paper deals with algorithms for text localization and character segmentation in images for process automation in the steel-making industry. Each character which comprises slab identification numbers may be corrupted severely before it is captured by network cameras. Therefore, proper processing is required to localize the target texts successfully. In this paper, we propose (1) a method to evaluate the closeness of an edge patch to the form of a closed contour, (2) an edge inspection method to determine character colors and estimate font thickness, and (3) three reasonable binarization methods to increase the performance of the algorithm for the detection of the left and right boundaries of the text rectangle. The experimental results show that the proposed algorithms are reliable.

Text localization and character segmentation algorithms for automatic recognition of slab identification numbers

This paper describes application-oriented text localization and character segmentation algorithms in images. The target text in our application includes many unclear characters due to poor environment as well as the fact that their positions are variable in the images. Consequently, it is difficult to expect a high success rate when using existing text localization algorithms that have been developed for generic texts. Therefore, it is necessary to develop a new text localization algorithm. We propose (1) a coarse algorithm for detecting top and bottom boundaries, (2) a fitness function that is used to decide the true text among the text candidates, (3) two kinds of presegmentation algorithms for calculating the fitness function, and (4) a blank-detecting algorithm that determines whether the text is upside down or not. By the proposed algorithms, input upside-down text is rotated automatically without using any supervised or unsupervised learning methods; further, character segmentation can be done in the process of selecting the true text. To evaluate the algorithms, image data captured by the installed recognition system at Pohang Steel Company (POSCO) are used, and experimental results show that the proposed algorithms are fast and reliable.

Edge-based text localization and character segmentation algorithms for automatic slab information recognition

Content-based image indexing is to label image based on its content such as color, texture, shape, face, text, and etc. Because the text can be easily localized and recognized compared to other image contents, many researchers have studied about the general text localization in images actively. Generally, it is known that localizing scene text is more difficult than localizing caption text, and still it is not easy. However, although our target texts are scene texts, it is clear that they are not general, that is, the properties of target texts are fixed. In this paper, we propose an edge-based text localization and segmentation algorithms for automatic slab information recognition system.