• Proceedings of the Korean Society of Computer Information Conference
• /
• 2020.07a
• /
• pp.23-26
• /
• 2020

본 논문에서는 CCTV를 통해 얻은 영상에서 얼굴을 인식하여 자동으로 출석 여부를 체크하는 시스템을 소개한다. 이 시스템은 CNN을 바탕으로 RetinaFace 모델을 사용하여 얼굴을 탐지하고, 탐지된 얼굴을 ArcFace 모델로 R⁵¹²의 목표 공간으로 임베딩한다. 기존 데이터베이스에 등록된 얼굴과 CCTV를 통해 얻은 얼굴들의 임베딩 벡터 사이의 Angular Cosine Distance를 측정하여 동일 인물인지 판단하는 매칭 알고리즘을 제안한다. 실험을 통해 두 모델을 동시에 사용할 최적의 환경을 파악하고, 마스크 착용으로 얼굴의 하단부가 가려지는 폐색 문제에 더욱 효과적으로 대응하여 매칭 성능을 높이는 방법을 제안한다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.12
• /
• pp.2341-2346
• /
• 2017

The relative position of the eyeballs in the orbit can be a criterion for estimating multiple pathological conditions. Especially, it is useful to diagnose orbital fracture, thyroid eye disease, orbital tumor, and to evaluate the result of drug and surgical treatment. The Hertel and Naugle exophthalmometer, which are the most commonly used measuring instruments for measuring the prominence of the eye, are different from each other. Even if the same examiner repeatedly measures, it is inevitable. Also, even if the same exophthalmometer is different from the manufacturing company, the design of the fixed part of the orbit is different, and a measurement error is caused by the inspectors. In this paper, we propose a method of automatic measurement that can increase the accuracy and repeatability of measurement of exophthalmos using white light scanning interferometer, which is a 3D image measurement technique.

• Convergence Security Journal
• /
• v.19 no.5
• /
• pp.75-84
• /
• 2019

Recently, the value of video as an important data of medical information technology is increasing due to the feature of rich clinical information. On the other hand, video is also required to be de-identified as a medical image, but the existing methods are mainly specialized in the stereotyped data and still images, which makes it difficult to apply the existing methods to the video data. In this paper, we propose an automated system to index candidate elements of personal identification information on a frame basis to solve this problem. The proposed system performs indexing process using text and person detection after preprocessing by scene segmentation and color knowledge based method. The generated index information is provided as metadata according to the purpose of use. In order to verify the effectiveness of the proposed system, the indexing speed was measured using prototype implementation and real surgical video. As a result, the work speed was more than twice as fast as the playing time of the input video, and it was confirmed that the decision making was possible through the case of the production of surgical education contents.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.45 no.6
• /
• pp.149-156
• /
• 2008

SLAM(Simultaneous Localization and Mapping) system is to find a global position and build a map with sensing data when an unmanned-robot navigates an unknown environment. Two kinds of system were developed. One is used distance measurement sensors such as an ultra sonic and a laser sensor. The other is used stereo vision system. The distance measurement SLAM with sensors has low computing time and low cost, but precision of system can be somewhat worse by measurement error or non-linearity of the sensor In contrast, stereo vision system can accurately measure the 3D space area, but it needs high-end system for complex calculation and it is an expensive tool. In this paper, we implement the SLAM system using a single camera image and a PSD sensors. It detects obstacles from the front PSD sensor and then perceive size and feature of the obstacles by image processing. The probability SLAM was implemented using the data of sensor and image and we verify the performance of the system by real experiment.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.7 no.7
• /
• pp.1551-1555
• /
• 2003

This paper describes the extraction method for cerebro-region using cellular automata from the cerebrum MR images. The cerebro-region extracted from applying the cellular automata rules obtained from histogram distribution analysis after removing the background image from the cerebro-region by determining the critical value. The experiment results showed that PSNR is 42㏈ on the image quality and the correlation factor is estimated 98%. And the result can be used as the medical auto-diagnostics system of cerebrum.

• Journal of the Institute of Convergence Signal Processing
• /
• v.15 no.4
• /
• pp.148-154
• /
• 2014

This paper suggests a new system based on image-processed information which is designed to control traffic light by measuring two-way traffic at the same time with a surveillance camera. With this system, the existing way of using one camera for one lane or loop coil will be improved to the more effective way of using only one camera to monitor a two-way, 8 lane roads statistically. Car images can be detected clearly even in irregular condition because of the background updating in real time. In addition, more accurate measurement is possible to users by selecting extra attention-needed regions. The automatic traffic light controlling algorithm, suggested in this paper, will prevent users and drivers from wasting their time and energy by controlling the number of traffic in advance.

• The Journal of the Korea Contents Association
• /
• v.14 no.11
• /
• pp.1-6
• /
• 2014

Emotion recognition is defined as that machines automatically recognize human emotions. Because the human emotions is very subjective, it is impossible to measure objectively. Therefore, the goal of emotion recognition is to obtain a measure that is agreed by as many people as possible. Emotion recognition in a image is implemented as the method that matches human emotions to the various features of the image. In the paper, we propose an emotion recognition system using color features of clothing image based on the Kobayashi's image scale. The proposed system stores colors of image scale into a database. And extracted major colors from a input clothing image are compared with those in the database. The proposed system can obtain three emotions maximally. In order to evaluate the system performance 70 observers are tested. The test results shows that recognized emotions of the proposed system are very similar to the observers emotions.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.11
• /
• pp.1384-1390
• /
• 2019

This paper proposes a technological solution for improving the recognition direction of target objects for automatic vision inspection by machine vision. This paper proposes a technological solution for improving the recognition direction of target objects for automatic vision inspection by machine vision. This enables the automatic machine vision inspection to detect the image of the inspection object regardless of the position and orientation of the object, eliminating the need for a separate inspection jig and improving the automation level of the inspection process. This study develops the technology and method that can be applied to the wire harness manufacturing process as the inspection object and present the result of real system. The results of the system implementation was evaluated by the accredited institution. This includes successful measurement in the accuracy, detection recognition, reproducibility and positioning success rate, and achievement the goal in ten kinds of color discrimination ability, inspection time within one second and four automatic mode setting, etc.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2018.05a
• /
• pp.273-273
• /
• 2018

하천유량을 산정하는 전통적인 방법은 수위-유량관계(stage-discharge relation)를 이용하는 것이며, 최근에는 하천특성에 따라 연속적으로 유속을 직접 측정하고 지표유속법(index velocity method) 또는 유속분포법(velocity profile method)을 이용하는 방법도 적용되고 있다. 연속으로 유속을 측정하는 방법에는 전자파 또는 영상이미지를 이용하여 표면유속을 측정하는 방법과 수중에 초음파 유속계를 설치하여 수중의 유속을 측정하는 방법이 있으며, 국내에서는 초음파 유속계를 이용하는 방법이 주로 활용되고 있다. 본 연구에서는 전자파표면유속계(MU2720)로 측정된 표면유속과 지표유속법을 이용하여 하천유량 산정방법을 검토하였다. 본 연구는 홍천강의 홍천군(반곡교)와 홍천군(홍천교), 한탄강의 철원군(한탄대교) 수위관측소에서 2016년도에 전자파표면유속계로 측정된 43개의 자료(평균유속 0.61~2.56m/s)를 이용하였다. 전자파표면유속계에 의한 유속측정은 한 지점에 고정하여 측정하지 않고 여러 개의 측선을 분할하여 표면유속을 측정하였으며, 단면 평균유속(Vm)과 가장 상관성이 높은 측선(지점)의 유속을 지표유속으로 하여 지표유속(vi)과 평균유속과의 관계식을 작성하고, 이를 통해 유량을 산정하였다. 여기서 평균유속은 전자파표면유속계로 측정한 유속을 이용하여 산정한 값이다. 그 결과 지표유속과 평균유속관계식의 결정계수($R^2$)는 0.9874~0.9976이었으며, 측정 평균유속과 지표유속관계식으로 산정된 평균유속과의 평균 편차율은 2.26~3.80%로 표면유속과 지표유속법을 이용한 유량산정 방법의 적정성을 확인할 수 있었다. 따라서 현재 국내에서 자동유량측정시스템에 초음파유속계만 상용화되고 있으나 향후에는 하천특성에 따라 전자파표면유속계를 이용하는 방법도 고려해 볼 필요가 있을 것으로 판단된다.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.25 no.5
• /
• pp.427-436
• /
• 2007

Because the slope made up the construction of any other facilities, has many risks of a collapse, existing inspection methods to collect information for a construction site of slope bring up a long time of inspection period, cost and approach for a measuring instrument and it presents the critical point of collecting materials. For getting images to use zoom lens in any positions this study will use free zoomer constructed values of data classified by the focal length develop Image Loader system to make it load not only camera information but also camera test data values of the focal length took a photograph automatically if it measure to use a variety of cameras or other lens. Also, as it constructs three dimensions spatial image information from images of obtained objects this study presents effective basic materials of slope surveying and inspection and it shows exact surveying methods for dangerous slope not to access.