http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
반도체 패키지의 마크문자 회전량 측정을 위한 고속 라돈 변환에 관한 연구
신균섭 ( Gyunseob Shin ),주효남 ( Hyonam Joo ),김상민 ( Sangmin Kim ),이정섭 ( Jung-seob Lee ) 한국정보처리학회 2010 한국정보처리학회 학술대회논문집 Vol.17 No.1
반도체 패키지 제조공정 중에는 제품에 일련번호를 인쇄하는 마킹공정이 있다. 마킹 공정에서 새겨진 문자는 해당 관리기준에 따라 관리되고 있는데 최근 반도체 패키지의 소형화에 따라 인쇄된 마크문자의 틀어짐 정도가 관리기준에 미달되는 문제가 발생되고 있다. 본 논문에서는 마크문자의 검사 항목 중 tilted mark(angle mark) 검사를 위한 회전량 측정방법으로 golden section searching 방법을 적용한 고속 라돈 변환(radon transform)방법을 제안한다. 실험에서는 제안한 방법이 일반적인 라돈 변환에 비해 최대 약 21 배의 회전량 측정속도가 향상되는 것을 확인하였다.
Fast Detection of Mura Defects Based on Modified Watershed Algorithm
Ye Jian Zhang(장엽검),Hyonam Joo(주효남),Joon Seek Kim(김준식) 제어로봇시스템학회 2017 제어·로봇·시스템학회 논문지 Vol.23 No.6
Many kinds of defects show up during the process of manufacturing display panels. However, mura defects are the most difficult to detect using the conventional image processing algorithms. Many factors cause mura defects to appear in display panels. When images are taken using cameras, mura defects normally show up as relatively dark or bright regions with no definite shape, no clear contours, and very low contrast against their surrounding background. When an imaged mura defect is relatively dark compared to its background, it can be considered a water catchment basin when the whole image is visualized in three dimensions (i.e., is topographically interpreted), and such catchment basins can be detected by watershed algorithms. In this paper, for the accurate segmentation of the mura region, the flooding step of the original watershed algorithm is carefully redesigned to detect the mura defect that exists both inside and at the boundary of an image. The depth of the catchment basins is recorded iteratively and then is used to segment the mura defects. The just noticeable difference (JND) technique is used to quantify the level of the mura defects. It is shown, by extensive experiments, that the proposed algorithm performs well, detecting very low-contrast mura defects, and quickly detects defects located anywhere in the image.
Low Contrast 특성을 갖는 LCD 편광필름 결함의 크기 자동 검출
박덕천(Duckchun Park),주효남(Hyonam Joo),류근호(Keun-Ho Rew) 제어로봇시스템학회 2008 제어·로봇·시스템학회 논문지 Vol.14 No.5
In this paper, segmenting and classifying low contrast defects on flat panel display is one of the key problems for automatic inspection system in practice. Problems become more complicated when the quality of acquired image is degraded by the illumination irregularity. Many algorithms are developed and implemented successfully for the defects segmentation. However, vision algorithms are inherently prone to be dependent on parameters to be set manually. In this paper, one morphological segmentation algorithm is chosen and a technique using frequency domain analysis of input images is developed for automatically selection the morphological parameter. An extensive statistical performance analysis is performed to compare the developed algorithms.
반동체 웨이퍼 고속 검사를 위한 GPU 기반 병열처리 알고리즘
박영대(Youngdae Park),김준식(Joon Seek Kim),주효남(Hyonam Joo) 제어로봇시스템학회 2013 제어·로봇·시스템학회 논문지 Vol.19 No.12
In a the present day, many vision inspection techniques are used in productive industrial areas. In particular, in the semiconductor industry the vision inspection system for wafers is a very important system. Also, inspection techniques for semiconductor wafer production are required to ensure high precision and fast inspection. In order to achieve these objectives, parallel processing of the inspection algorithm is essentially needed. In this paper, we propose the GPU (Graphical Processing Unit)-based parallel processing algorithm for the fast inspection of semiconductor wafers. The proposed algorithm is implemented on GPU boards made by NVIDIA Company. The defect detection performance of the proposed algorithm implemented on the GPU is the same as if by a single CPU, but the execution time of the proposed method is about 210 times faster than the one with a single CPU.
확장된 피사계 심도 알고리즘에서 엣지 정보 분석에 의한 3차원 깊이 정보 추출 방법
강선우(Sunwoo Kang),김준식(Joon Seek Kim),주효남(Hyonam Joo) 제어로봇시스템학회 2016 제어·로봇·시스템학회 논문지 Vol.22 No.2
Recently, popularity of 3D technology has been growing significantly and it has many application parts in the various fields of industry. In order to overcome the limitations of 2D machine vision technologies based on 2D image, we need the 3D measurement technologies. There are many 3D measurement methods as such scanning probe microscope, phase shifting interferometry, confocal scanning microscope, white-light scanning interferometry, and so on. In this paper, we have used the extended depth of focus (EDF) algorithm among 3D measurement methods. The EDF algorithm is the method which extracts the 3D information from 2D images acquired by short range depth camera. In this paper, we propose the EDF algorithm using the edge informations of images and the average values of all pixel on z-axis to improve the performance of conventional method. To verify the performance of the proposed method, we use the various synthetic images made by point spread function(PSF) algorithm. We can correctly make a comparison between the performance of proposed method and conventional one because the depth information of these synthetic images was known. Through the experimental results, the PSNR of the proposed algorithm was improved about 1 ~ 30 dB than conventional method.
QFN 패키지의 Resin Bleed와 Melting 검출 알고리즘
왕명걸(Mingjie Wang),박덕천(Duckchun Park),주효남(Hyonam Joo),김준식(Joon-Seek Kim) 제어로봇시스템학회 2009 제어·로봇·시스템학회 논문지 Vol.15 No.9
There are many different types of surface defects on semiconductor Integrated Chips (IC’s) caused by various factors during manufacturing process, such as Scratch, Flash, Resin bleed, and Melting. These defects must be detected and classified by an inspection system for productivity improvement and effective process control. Among defects, in particular, Resin bleed and Melting are the most difficult ones to classify accurately. The brightness value and the shape of Resin bleed and Melting defects are so similar that normally it is difficult to classify the Resin bleed and Melting. In this paper, we propose a segmenting method and a set of features for detecting and classifying the Resin bleed and Melting defects.
반도체 부품 마크 미세 결함 검사를 위한 패턴 영역 분할 및 인식 방법
장유정(Yuting Zhang),이정섭(Jung-Seob Lee),주효남(Hyonam Joo),김준식(Joon-seek Kim) 제어로봇시스템학회 2010 제어·로봇·시스템학회 논문지 Vol.16 No.9
To inspect the defects of printed markings on the surface of IC package, the OCV (Optical Character Verification) method based on NCC (Normalized Correlation Coefficient) pattern matching is widely used. In order to detect the micro pattern defects appearing on the small portion of the markings, a Partitioned NCC pattern matching method was proposed to overcome the limitation of the NCC pattern matching. In this method, the reference pattern is first partitioned into several blocks and the NCC values are computed and are combined in these small partitioned blocks, rather than just using the NCC value for the whole reference pattern. In this paper, we proposed a method to decide the proper number of partition blocks and a method to inspect and combine the NCC values of each partitioned block to identify the defective markings.
지성철(Seong Cheol Ji),강선우(Sun Woo Kang),김준식(Joon Seek Kim),주효남(Hyonam Joo) 제어로봇시스템학회 2015 제어·로봇·시스템학회 논문지 Vol.21 No.12
A camera-based real-time hand posture and gesture recognition system is proposed for controlling various devices inside automobiles. It uses an imaging system composed of a camera with a proper filter and an infrared lighting device to acquire images of hand-motion sequences. Several steps of pre-processing algorithms are applied, followed by a background normalization process before segmenting the hand from the background. The hand posture is determined by first separating the fingers from the main body of the hand and then by finding the relative position of the fingers from the center of the hand. The beginning and ending of the hand motion from the sequence of the acquired images are detected using pre-defined motion rules to start the hand gesture recognition. A set of carefully designed features is computed and extracted from the raw sequence and is fed into a decision tree-like decision rule for determining the hand gesture. Many experiments are performed to verify the system. In this paper, we show the performance results from tests on the 550 sequences of hand motion images collected from five different individuals to cover the variations among many users of the system in a real-time environment. Among them, 539 sequences are correctly recognized, showing a recognition rate of 98%.
박제성(Je Sung Park),이원우(Won Woo Lee),Zhangye Jian,주효남(Hyonam Joo),김준식(Joon Seek Kim) 제어로봇시스템학회 2016 제어·로봇·시스템학회 논문지 Vol.22 No.5
When Flat Panel Display (FPD) is made with backlight module, such as LED TV, it inherently suffers from the non-uniform backlight luminance problem that results in un-even brightness distribution throughout the TV screen. If the luminance of each pixel location of a TV screen as a function of the driving voltage can be measured, it can be used to compensate the non-uniformity of the backlight module. We use a carefully calibrated imaging system to take pictures of a TV screen at different levels of brightness and generate the compensation functions for the driving circuitry to correct the luminance level at each pixel location. Making use of the fact that the luminance of the screen is normally brightest at around the center of the screen and gradually decreases toward the border of the screen, the luminance of the whole TV screen is approximated by a mathematical function of the pixel locations. The parameters of the function are computed in the least square sense by the values of both the pixel luminance sent from the driving circuit and the grayscale value measured from the image taken by the imaging system. To justify the correction system, a simple second order polynomial function is used to approximate the luminance across the screen. When the driving circuit voltage is corrected according to the measured function, the variance of the screen luminance is reduced to one tenth of the one measured from the un-corrected TV screen.