• Journal of Institute of Control, Robotics and Systems
• /
• v.22 no.4
• /
• pp.247-252
• /
• 2016

In this paper, we propose a new defect detection technology for micro-defect on the surface of steel products. Due to depth and size of microscopic defect, slop of surface and vibration of strip, the conventional optical method cannot guarantee the detection performance. To solve the above-mentioned problems and increase signal to noise ratio, a novel retro-schlieren method that consists of retro reflector and knife edge is proposed. Moreover dual switching lighting method is also applied to distinguish uneven micro defects and surface noise. In proposed method, defective regions are represented by a black and white pattern. This pattern is detected by a defect detection algorithm with Gabor filter. Experimental results by simulator for sample defects of cold rolled steel show that the proposed method is effective.

• International Journal of Internet, Broadcasting and Communication
• /
• v.14 no.2
• /
• pp.136-146
• /
• 2022

Some surface defects make the weak quality of steel materials. To limit these defects, we advocate a one-stage detector model RetinaNet among diverse detection algorithms in deep learning. There are several backbones in the RetinaNet model. We acknowledged two backbones, which are ResNet50 and VGG19. To validate our model, we compared and analyzed several traditional models, one-stage models like YOLO and SSD models and two-stage models like Faster-RCNN, EDDN, and Xception models, with simulations based on steel individual classes. We also performed the correlation of the time factor between one-stage and two-stage models. Comparative analysis shows that the proposed model achieves excellent results on the dataset of the Northeastern University surface defect detection dataset. We would like to work on different backbones to check the efficiency of the model for real world, increasing the datasets through augmentation and focus on improving our limitation.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.17 no.3
• /
• pp.129-138
• /
• 2022

Collecting and labeling sufficient training data, which is essential to deep learning-based visual inspection, is difficult for manufacturers to perform because it is very expensive. This paper presents a steel plate surface defect detection system with industrial-grade detection performance by training a small amount of steel plate surface images consisting of labeled and non-labeled data. To overcome the problem of lack of training data, we propose two data augmentation techniques: program-based augmentation, which generates defect images in a geometric way, and generative model-based augmentation, which learns the distribution of labeled data. We also propose a 4-step semi-supervised learning using pseudo labels and consistency training with fixed-size augmentation in order to utilize unlabeled data for training. The proposed technique obtained about 99% defect detection performance for four defect types by using 100 real images including labeled and unlabeled data.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.3
• /
• pp.19-26
• /
• 2020

Steel production requires high-quality surfaces with minimal defects. Therefore, the detection algorithms for the surface defects of steel strip should have good generalization performance. To meet the growing demand for high-quality products, the use of intelligent visual inspection systems is becoming essential in production lines. In this paper, we proposed a ShuffleDefectNet defect detection system based on deep learning. The proposed defect detection system exceeds state-of-the-art performance for defect detection on the Northeastern University (NEU) dataset obtaining a mean average accuracy of 99.75%. We train the best performing detection with different amounts of training data and observe the performance of detection. We notice that accuracy and speed improve significantly when use the overall architecture of ShuffleDefectNet.

• Proceedings of the KSME Conference
• /
• 2008.11a
• /
• pp.317-321
• /
• 2008

Piping in the Nuclear Power plants (NPP) are mostly consisted of carbon steel pipe. The wall thinning defect is mainly occurred by the affect of the flow accelerated corrosion (FAC) of fluid which flows in carbon steel pipes. This type of defect becomes the cause of damage or destruction of piping. Therefore, it is very important to measure defect which is existed not only on the welding partbut also on the whole field of pipe. Over the years, Infrared thermography (IRT) has been used as a non destructive testing methods of the various kinds of materials. This technique has many merits and applied to the industrial field but has limitation to the materials. Therefore, this method was combined with lock-in technique. So IRT detection resolution has been progressively improved using lock-in technique. In this paper, the quantitative analysis results of the location and the size of wall thinning defect that is artificially processed inside the carbon steel pipe by using IRT are obtained.

• Steel and Composite Structures
• /
• v.19 no.1
• /
• pp.131-152
• /
• 2015

Easy detection and evaluation of defect in the tube structure is a continuous problem and remains a significant demand in tube inspection technologies. This study is aimed to automate defect detection using the pattern recognition approach based on the classification of high frequency stress wave signals. The stress wave signals from vibrational impact excitation on several tube conditions were captured to identify the defect in ASTM A179 seamless steel tubes. The variation in stress wave propagation was captured by a high frequency sensor. Stress wave signals from four tubes with artificial defects of different depths and one reference tube were classified using the autoregressive (AR) algorithm. The results were demonstrated using a dendrogram. The preliminary research revealed the natural arrangement of stress wave signals were grouped into two clusters. The stress wave signals from the healthy tube were grouped together in one cluster and the signals from the defective tubes were classified in another cluster. This approach was effective in separating different stress wave signals and allowed quicker and easier defect identification and interpretation in steel tubes.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.23 no.3
• /
• pp.270-279
• /
• 2003

Parametric studies have been conducted into the variability of the factors affecting the ultrasonic testing applied to weldments. The influence of ultrasonic equipment, transducer parameters, test technique, job parameters, defect type and characteristics on reliability far defect detection and sizing was investigated by experimentation. The investigation was able to build up substantial bank of information on the reliability of manual ultrasonic method for testing weldments. The major findings of the study separate into two parts, one dealing with correlation between ultrasonic techniques, equipment and defect parameters and inspection performance effectiveness and other with human factors. Defect detection abilities are dependent on the training, experience and proficiency of the UT operators, the equipment used, the effectiveness of procedures and techniques.

• Journal of Information Processing Systems
• /
• v.19 no.6
• /
• pp.745-755
• /
• 2023

The wire rope is an indispensable production machinery in coal mines. It is the main force-bearing equipment of the underground traction system. Accurate detection of wire rope defects and positions exerts an exceedingly crucial role in safe production. The existing defect detection solutions exhibit some deficiencies pertaining to the flexibility, accuracy and real-time performance of wire rope defect detection. To solve the aforementioned problems, this study utilizes the camera to sample the wire rope before the well entry, and proposes an object based on YOLOv5. The surface small-defect detection model realizes the accurate detection of small defects outside the wire rope. The transfer learning method is also introduced to enhance the model accuracy of small sample training. Herein, the enhanced YOLOv5 algorithm effectively enhances the accuracy of target detection and solves the defect detection problem of wire rope utilized in mine, and somewhat avoids accidents occasioned by wire rope damage. After a large number of experiments, it is revealed that in the task of wire rope defect detection, the average correctness rate and the average accuracy rate of the model are significantly enhanced with those before the modification, and that the detection speed can be maintained at a real-time level.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.4
• /
• pp.577-586
• /
• 2022

Detection and classification of steel surface defects are critical for product quality control in the steel industry. However, due to its low accuracy and slow speed, the traditional approach cannot be effectively used in a production line. The current, widely used algorithm (based on deep learning) has an accuracy problem, and there are still rooms for development. This paper proposes a method of steel surface defect detection combining EfficientNetV2 for image classification and YOLOv5 as an object detector. Shorter training time and high accuracy are advantages of this model. Firstly, the image input into EfficientNetV2 model classifies defect classes and predicts probability of having defects. If the probability of having a defect is less than 0.25, the algorithm directly recognizes that the sample has no defects. Otherwise, the samples are further input into YOLOv5 to accomplish the defect detection process on the metal surface. Experiments show that proposed model has good performance on the NEU dataset with an accuracy of 98.3%. Simultaneously, the average training speed is shorter than other models.

• Journal of Welding and Joining
• /
• v.16 no.3
• /
• pp.64-73
• /
• 1998

This paper is concerned with defects detection and evaluation of heat affected zone (HAZ) in austenitic stainless steel type 304 by ultrasonic wave and neural network. In experiment, the reflected ultrasonic defect signals from artificial defects (side hole, vertical hole, notch) of HAZ appears as beam distance of prove-defect, distance of probe-surface, depth of defect-surface on CRT. For defect classification simulation, neural network system was organized using total results of ultrasonic experiment. The organized neural network system was learned with the accuracy of 99%. Also it could be classified with the accuracy of 80% in side hole, and 100% in vertical hole, 90% in notch about ultrasonic pattern recognition. Simulation results of neural network agree fairly well with results of ultrasonic experiment. Thus were think that the constructed system (ultrasonic wave - neural network) in this work is useful for defects dection and classification such as holes and notches in HAZ of austenitic stainless steel 304.