• 제어로봇시스템학회논문지
• /
• 제3권4호
• /
• pp.406-414
• /
• 1997

A real-time inspection system is developed using line scan cameras. Several improved algorithms are proposed for real-time detection of defects in this automated inspection system. The major improved algorithms include the preprocessing, the threshold decision, and the clustering algorithms. The preprocessing algorithms are for exact binarization and the threshold decision algorithm is for fast detection of defects in 1-D binary images. The clustering algorithm is also developed for fast classifying of the defects. The system is applied to PCBs(Printed Circuit Boards) inspection. The typical defects in PCBs are pits, dent, wrinkle, scratch, and black spots. The results show that most defects are detected and classified successfully.

• 반도체디스플레이기술학회지
• /
• 제10권2호
• /
• pp.97-102
• /
• 2011

Lots of 2D vision algorithms have been applied for inspection. However, these 2D vision algorithms have limitation in inspection applications which require 3D information data such as the height of semiconductor chips. Stereo vision is a well known method to measure the distance from the camera to the object to be measured. But it is difficult to apply for inspection directly because of its measurement error. In this paper, we propose two adjustment methods to reduce the error of the measured height data for stereo vision. The weight value based model is used to minimize the mean squared error. The average value based model is used with simple concept to reduce the measured error. The effect of these algorithms has been proved through the experiments which measure the height of semiconductor chips.

• 반도체디스플레이기술학회지
• /
• 제9권3호
• /
• pp.59-63
• /
• 2010

This paper proposes computer visual inspection algorithms for various LCD defects which are found in a manufacturing process. Modular vision processing steps are required in order to detect different types of LCD defects. Those key modules include RGB filtering for pixel defects, gray-scale morphological processing and Hough transform for line defects, and adaptive threshold for spot defects. The proposed algorithms can give users detailed information on the type of defects in the LCD panel, the size of defect, and its location. The machine vision inspection system is implemented using C language in an embedded Linux system for a high-speed real-time image processing. Experiment results show that the proposed algorithms are quite successful.

• 전기학회논문지P
• /
• 제59권1호
• /
• pp.89-95
• /
• 2010

A new live-line insulator tester was developed to detect faulty insulators in 154kV power transmission lines. This paper is the second part of the two-part paper and deals with its inspection algorithm development. Unlike normal condition with low pollution and low humidity, the inspection data measured in the field under high pollution or high humidity showed that the voltage distribution of an insulator string has offsets in comparison with those of others and its insulation resistances are greatly decreased, which leads to wrong results of the existing inspection algorithms under such conditions. To solve this problem, we propose new diagnosis algorithms that can exactly detect faulty insulators from measured data regardless of environmental conditions. Its effectiveness was validated by live-line field tests in actual power lines.

• 전기학회논문지
• /
• 제67권7호
• /
• pp.865-870
• /
• 2018

A machine vision based industrial inspection includes defects detection and classification. Fast inspection is a fundamental problem for many applications of real-time vision systems. It requires little computation time and localizing defects robustly with high accuracy. Deep learning technique have been known not to be suitable for real-time applications. Recently a couple of fast region-based CNN algorithms for object detection are introduced, such as Faster R-CNN, and YOLOv2. We apply these methods for an industrial inspection problem. Three CNN based detection algorithms, VOV based CNN, Faster R-CNN, and YOLOv2, are experimented for defect detection on metal surface. The results for inspection time and various performance indices are compared and analysed.

• 제어로봇시스템학회논문지
• /
• 제9권8호
• /
• pp.640-646
• /
• 2003

We presents an edge-based point correlation algorithm for fast and stable visual inspection system. Conventional algorithms based on NGC(normalized gray-level correlation) have to overcome some difficulties in applying automated inspection systems to real factory environment. First of all, NGC algorithms involve highly complex computation and thus require high performance hardware for realtime process. In addition, lighting condition in realistic factory environments is not stable and therefore intensity variation from uncontrolled lights gives many troubles for applying NGC directly as pattern matching algorithm. We propose an algorithm to solve these problems, using thinned and binarized edge data, which are obtained from the original image. A point correlation algorithm with the thinned edges is introduced with image pyramid technique to reduce the computational complexity. Matching edges instead of using original gray-level image pixels overcomes problems in NGC method and pyramid of edges also provides fast and stable processing. All proposed methods are proved by the experiments using real images.

• 한국구조물진단유지관리공학회지
• /
• 제18권3호
• /
• pp.62-67
• /
• 2014

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2002년도 춘계학술대회 논문집
• /
• pp.173-176
• /
• 2002

In this paper, we present the combined 2-D and 3-D algorithms for automatic solder paste inspection. For automatic inspection, optical system for the combined inspection and driving unit is made. One-pass run length algorithm that has fast and efficient memory space is applied to the input image fur extracting solder paste patterns. The path of probe movement is then calculated for an automatic inspection. For a fast 3-D inspection, the phase shift algorithm based on Moire interferometry is also used. In addition, algorithms used in this paper are coded by $MMX^{TM}$. A probe system is manufactured to simultaneously inspect 2-D and 3-D for 10mm$\times$10mm field of view, with resolutions of 10 $\mu\textrm{m}$for both x, y axis and 17 $\mu\textrm{m}$for z axis, and then, experiments on several PCBs are conducted. The processing times of 2-D and 3-D, excluding an image capturing, is 0.039 sec and 0.047 sec, respectively. The credible result with $\pm$ 1$\mu\textrm{m}$uncertainty can be also achieved.

• 한국해양공학회지
• /
• 제30권3호
• /
• pp.194-200
• /
• 2016

Recently, UUVs (unmanned underwater vehicles) have increasingly been applied in various science and engineering applications. In-water inspection, which used to be performed by human divers, is a potential application for UUVs. In particular, the operational safety and performance of in-water inspection missions can be greatly improved by using an underwater robotic vehicle. The capabilities of hovering maneuvers and automatic image mosaicking are essential for autonomous underwater visual inspection. This paper presents the development of a hover-capable autonomous underwater vehicle system for autonomous in-water inspection, which includes both a hardware platform and operational software algorithms. Some results from an experiment in a model basin are presented to demonstrate the feasibility of the developed system and algorithms.

• 제어로봇시스템학회논문지
• /
• 제15권2호
• /
• pp.169-175
• /
• 2009

A new wire detection algorithm for power line inspection by a mobile robot has been proposed in this paper. There have been a lot of studies in order to support the high-quality electric power. For the high-quality power supply, it is necessary to investigate the power lines and insulators before the lines or insulators were disconnected or damaged. Although Korea Electric power Corp. has made many efforts for the quality improvement, it is not enough to inspect all the power lines by human inspectors. According to this problem, it is decided to replace the human operators by the power line inspection robot. When the robots are used for the inspection, there could be several advantages, for example, the working efficiency and the prevention of accident. And also the shortage of human power for dangerous jobs can be resolved. In this paper, as a part of the development of power line inspection robot, DICRO, the sensor fusion and fuzzy control algorithms are developed to detect the wire and slope of the wire. The effectiveness of the proposed algorithms is proved by the real experiments with DICRO which is under development so far.