• Journal of Information Processing Systems
• /
• 제12권1호
• /
• pp.100-108
• /
• 2016

In this paper, we propose a method to achieve improved number plate detection for mobile devices by applying a multiple convolutional neural network (CNN) approach. First, we processed supervised CNN-verified car detection and then we applied the detected car regions to the next supervised CNN-verifier for number plate detection. In the final step, the detected number plate regions were verified through optical character recognition by another CNN-verifier. Since mobile devices are limited in computation power, we are proposing a fast method to recognize number plates. We expect for it to be used in the field of intelligent transportation systems.

• Journal of Information Processing Systems
• /
• 제13권1호
• /
• pp.174-183
• /
• 2017

In this paper, we propose a novel feature for recognizing handwritten Odia numerals. By using polygonal approximation, each numeral is segmented into segments of equal pixel counts where the centroid of the character is kept as the origin. Three primitive contour features namely, distance (l), angle (${\theta}$), and arc-tochord ratio (r), are extracted from these segments. These features are used in a neural classifier so that the numerals are recognized. Other existing features are also considered for being recognized in the neural classifier, in order to perform a comparative analysis. We carried out a simulation on a large data set and conducted a comparative analysis with other features with respect to recognition accuracy and time requirements. Furthermore, we also applied the feature to the numeral recognition of two other languages-Bangla and English. In general, we observed that our proposed contour features outperform other schemes.

• 전기학회논문지
• /
• 제57권6호
• /
• pp.944-950
• /
• 2008

When small cogenerators are operated in connected with power system, there are many positive aspects such as the reduction of power plant construction, making a improvement of power security etc. At the same time, there are some negative effects or difficulties such as we should make sure of protective coordination, especially, Actually these are not fault, but it is likely to consider the current as the fault. This is one of major causes of malfunctions for protective relays in power distribution system which is including interconnection point. Thus, in this paper, We showed that the directional protection is necessary to the dispersed generation system which is including connection point. We also executed contingency analysis to find out the magnitude of fault current and direction which are classified by fault points, length of line and kinds of faults using ETAP power system analysis program.

• 대한전기학회논문지
• /
• 제41권9호
• /
• pp.953-959
• /
• 1992

As the TC(time-current) curve corresponding to symmetrical time invariant RMS value has been traditionally chosen and used for setting the relay, it has caused the misoperation errors on relay coordination because of CT secondary current being actually an asymmetrical time varying RMS value. In this paper, an algorithm for calculating the relay operating time is developed to study the asymmetrical effect using the step-by-step method. We represent the relay operating time errors between with and without DC offset versus PSM(plug setting multiplier), TMS(time multiplier setting) and X/R ratio. And also we present the correction factor. Finally we confirm the validity of this technique through the case study.

• 한국경영과학회:학술대회논문집
• /
• 한국경영과학회/대한산업공학회 2003년도 춘계공동학술대회
• /
• pp.1020-1027
• /
• 2003

The development of an automatic delivery sorter requires enormous cost and time, and actually it is a key business which gains control of the future of postal service for national economy. In this paper, we select executive alternatives considering the restrictive conditions both in physically and technically such as installation location, OCR location, and the types of VCS. Then, we evaluated selected alternatives through the analysis of a cost and a performance. And we also performed economy analysis reflecting on the qualitative factors.

• 전기학회논문지P
• /
• 제55권4호
• /
• pp.167-175
• /
• 2006

This paper presents a method to segment semiconductor wafer ID on poor quality images. The method is based on multiple templates and normalized gray-level correlation (NGC) method. If the lighting condition is not so good and hence, we can not control the image quality, target image to be inspected presents poor quality ID and it is not easy to identify and then recognize the ID characters. Conventional several method to segment the interesting ID regions fails on the bad quality images. In this paper, we propose a multiple template method, which uses combinational relation of multiple templates from model templates to match several characters of the inspection images. To find out the optimal solution of multiple template model in ID regions, we introduce newly-developed snake algorithm. Experimental results using images from real FA environment are presented.

• 대한전기학회논문지:전력기술부문A
• /
• 제52권3호
• /
• pp.143-150
• /
• 2003

In this paper, we developed the window based monitoring system. This system offers the analysis of accidents in the power plants, the operating status of relays and the historical data with which the experts can offer the idea about the operation of the power system. And we can also monitor the status of the ACB and relay at a long distance by means of the function of communication in the digital protective relay itself. By means of the above functions, we developed the system, which is run after the diagnosis and maintenance of the electrical machines and an amount of data are realized by graphical method fitting the operating convenience of the users.

• 제어로봇시스템학회논문지
• /
• 제10권8호
• /
• pp.726-733
• /
• 2004

This paper presents a method to segment object ID(identification) marks on poor quality images under uncontrolled lighting conditions of automated inspection process. The method is based on dynamic programming using multiple templates and normalized gray-level correlation (NGC) method. If the lighting condition is not good and hence, we can not control the image quality, target image to be inspected presents poor quality ID marks and it is not easy to identify and recognize the ID characters. Conventional several methods to segment the interesting ID mark regions fail on the bad quality images. In this paper, we propose a multiple template method, which uses combinational relation of multiple templates from model templates to match several characters of the inspection images. To increase the computation speed to segment the ID mark regions, we introduce the dynamic programming based algorithm. Experimental results using images from real factory automation(FA) environment are presented.

• 융합정보논문지
• /
• 제12권1호
• /
• pp.163-170
• /
• 2022

4차산업 혁명이 도래함에 따라 AI 기술을 적용하는 솔루션 개발이 활발하게 이루어지고 있다. 2017년도부터 금융권, 보험사를 중심으로 AI 기반 RPA(Robotic Process Automation)을 이용한 업무 자동화 솔루션 도입이 이루어지기 시작했으며, 최근에는 RPA 솔루션 도입 단계를 지나 확산하는 시기로 진입하고 있다. 이러한 RPA 솔루션을 이용한 업무 자동화 중에서 각 종 문서들을 이용한 업무 자동화에는 문서내의 문자 정보를 얼마나 정확하게 인식하는지가 매우 중요하다. 이러한 문자 인식은 최근 딥러닝 기술을 도입함으로써 그 정확도가 많이 높아졌지만, 여전히 완벽한 인식 정확도 갖는 인식 모델은 존재하지 않는다. 따라서, 본 논문에서는 딥러닝 기반 문자 인식 엔진에 전/후 처리기 기술을 적용할 경우 얼마나 정확도가 향상되는지를 확인하고 RPA 인식 엔진과 연계 기술을 구현하였다.

• 예술인문사회 융합 멀티미디어 논문지
• /
• 제6권6호
• /
• pp.103-110
• /
• 2016

최근 의료데이터의 유출사고가 빈번히 발생하여 환자의 프라이버시 침해 및 의료기관의 피해가 날로 증가하고 있다. 정부에서는 개인정보보호법등과 같은 법규를 제정하여 이러한 피해사례 예방하고 있다. 이중 의료기관 및 의료데이타에 대한 가이드라인은 보건복지부에서 발표한 '국내 의료기관 개인정보보호 가이드라인' 정도만 발표되어 있다. 환자개인의 민감정보를 포함한 의료데이타를 타의료기관 또는 제3의 연구기관등에 전달이 필요한 경우가 발생한다. 전달하고자 하는 의료 이미지 데이터를 일반적인 이미지파일 (JPG, JPEG, TIFF)의 포맷으로 자료의 교환이 이루어지고 있다. 이와같이 일반적인 이미지 포맷의 파일은 아무런 보호조치가 되어 있지 않아 외부로 유출시에는 파일내에 포함된 환자의 주요 식별정보가 노출되는 위험성이 존재한다. 본 연구에서는 이미지 파일에 대한 광학문자판독기술(OCR)을 적용하고 민감정보가 포함된 이미지파일에 암호화된 모자이크기술을 이용한 마스킹 기법을 도입하여 이러한 위험성을 해결하기 위한 이미지 비식별화 방안을 제시한다.