• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.9
• /
• pp.2189-2196
• /
• 2014

In this paper, a VLIW(Very Long Instruction Word) vision coprocessor which can efficiently accelerate computer vision algorithm for automotive is designed. The VLIW coprocessor executes four instructions per clock cycle via 8-stage pipelined structure and has 36 integer and floating-point instructions to accelerate computer vision algorithm for pedestrian detection. The processor has about 300-MHz operating frequency and about 210,900 gates under 45nm CMOS technology and its estimated performance is 1.2 GOPS(Giga Operations Per Second). The vision system composed of vision primitive engine and eight VLIW coprocessors can execute pedestrian detection at 25~29 frames per second(FPS). Because the VLIW coprocessor has high detection rate and loosely coupled interface with host processor, it can be efficiently applicable to a wide range of vision applications.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.12
• /
• pp.468-474
• /
• 2019

Korea had the highest accident rate among OECD countries in 2018, with 8.4 per 100,000 population, ranking 4th among 35 countries. In addition, the accident rate of traffic with children and the elderly was also high. This study reviewed the relevant literature and analyzed the traffic-accident analysis system. Customized traffic safety facilities were developed. In addition, by measuring the visibility of the traffic safety facilities by installing a test bed, this study measured the forward driving frequency and vehicle driving speed while driving. As a result of applying the "pedestrian pedestrian model" collision test model, the possibility of serious injury after installing the facility was reduced greatly to 4.6%. In this study, the visibility of traffic safety facilities and the effect of reducing the traffic speed were verified through test beds. Recognizing traffic safety facilities will reduce traffic accidents.

• Journal of Korean Society of Transportation
• /
• v.21 no.5
• /
• pp.119-133
• /
• 2003

도로는 교통기능 외에 여러 시설물을 수용할 수 있는 사회 인프라구축의 공간으로 활용될 수 있기 때문에 서울시뿐만 아니라 각 지방자치단체에서는 이러한 공간을 활용하기 위한 도로점용공사는 매년 전국적으로 수만 건에 달하고 있다. 따라서. 매년 발생하는 도로점용공사의 차로과다 점용, 교통시설물 설치 부적절, 보행자 안전무시, 공사안내표지 부적절, 공사구간에서의 불법 주.정차 등의 문제점이 계속하여 발생.지적되고 있으나 이에 따른 관련 규정이 미흡하여 도로점용공사 시행자의 불법을 초래하고 있는 실정이다. 또한 서울시의 경우 시스템적 접근방법인 관련조례에 의해서 수행되기보다는, 교통자문회의로 운영되고 있으며, 관련 인원 및 법적 제재조치가 미약하여 도로점용공사의 효과적인 관리가 어려운 실정이다. 따라서 서울시 및 각 지방자치단체에서는 도로점용공사로 인한 보행자와 운전자의 통행권 확보와 안전상의 문제 등을 하루빨리 인식하여 도로점용공사에 따른 도로공사장 교통관리 규정을 법규화 하여 체계적으로 관리할 필요가 있다. 이에 따라, 본 논문에서는 도로점용공사에 따른 관련법 및 적용대상, 관련보고서 검토, 서울시를 사례로 한 공사중 교통처리방안 운영실태 및 문제점의 진단을 통하여 향후 도로공사장 교통관리 규정의 운영적 및 제도적 측면의 개선방안과 향후 추진방향을 제시하고자 한다.

• Journal of Biomedical Engineering Research
• /
• v.38 no.3
• /
• pp.129-136
• /
• 2017

The exoskeleton robot is a technology developed to be used in various fields such as military, industry and medical treatment. The exoskeleton robot works by sensing the movement of the wearer. By recognizing the wearer's daily activities, the exoskeleton robot can assist the wearer quickly and efficiently utilize the system. In this study, LDA, QDA, and kNN are used to classify gait types through kinetic data obtained from subjects. Walking was selected from general walking and stair walking which are mainly performed in daily life. Seven IMUs sensors were attached to the subject at the predetermined positions to measure kinematic factors. As a result, LDA was classified as 78.42%, QDA as 86.16%, and kNN as 87.10% ~ 94.49% according to the value of k.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.12 no.4
• /
• pp.533-538
• /
• 2017

In this thesis, the ambulatory aid mechanism was implemented so that blind people could be safer at risk of walking outdoors. Using ultrasonic sensors, the obstacles can be detected when the distance between the obstacle is within 50 cm of the obstacle. If the light sensor becomes less than 25 lux, the LED will automatically turn on and help the safety of the visually impaired and the security of sight of the peripheral walkers. Color recognition sensors increase the rate of recognition of yellow color by the detection distance is 1cm, it vibrated when yellow light was detected. Using GPS with 7.3 m of error range, the guardian was able to check the location of the visually impaired.

• Journal of the HCI Society of Korea
• /
• v.14 no.2
• /
• pp.41-47
• /
• 2019

In this study, a smart assistive device is designed to recognize pedestrian signal and to provide audio instructions for visually impaired people in crossing streets safely. Walking alone is one of the biggest challenges to the visually impaired and it deteriorates their life quality. The proposed device has a camera attached on a pair of glasses which can detect traffic lights, recognize pedestrian signals in real-time using a machine learning algorithm on GPU board and provide audio instructions to the user. For the portability, the dimension of the device is designed to be compact and light but with sufficient battery life. The embedded processor of device is wired to the small camera which is attached on a pair of glasses. Also, on inner part of the leg of the glasses, a bone-conduction speaker is installed which can give audio instructions without blocking external sounds for safety reason. The performance of the proposed device was validated with experiments and it showed 87.0% recall and 100% precision for detecting pedestrian green light, and 94.4% recall and 97.1% precision for detecting pedestrian red light.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.45 no.6
• /
• pp.11-17
• /
• 2008

This paper proposes the USN node location recognition method with low power, which is the modified centroid method. The conventional Zigbee node location recognition is used to the three RSSI values from three beacons, respectively. When a person move with node, the moving speed of USN node doesn't fast. Therefore, one among three positions used to the value that it is a previous value. This method doesn't large the error in terms of the exact position. Using hand-off method, we get about 30% power advantage than the conventional centroid method. And Our monitoring system add the function that it is possible to search for the node region by color.

• Journal of Intelligence and Information Systems
• /
• v.15 no.4
• /
• pp.53-64
• /
• 2009

This paper presents a performance evaluation of shape descriptors for gait analysis in case of silhouette sequence images. We used moment descriptors(MD), Fourier descriptors(FD) and Zernike descriptors(ZD) as a shape descriptor. To evaluate their performance, we firstly defined the performance index, that is, AI(asymmetry index) and PI(periodic index) based on the periodic property of the gait images. This is why they are represented by periodic parameters due to periodic gait images. This index means that how the shape is represented periodically. According to these indexes, we evaluated the data sets with periodic images, downloaded from internet. The results showed that Zernike descriptors had better performance of AI = 1.09 and PI = 2.21 than others. And in case of FD and ZD, it's efficient to implement the gait analysis with 5~10 parameters.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.20 no.7
• /
• pp.1269-1276
• /
• 2016

Recently, natural disasters including earthquakes, tsunamis, floods, and snowstorms, in addition to disasters of human origin such as arson, and acts of terror, have caused numerous injuries and fatalities around the world. During such disasters, victims need to obtain information such as the exact location of the disaster and appropriate evacuation routes in order to relocate to safe areas. In this study, We propose the algorithm for Emergency Rescue Evacuation Support System(ERESS). In case a emergency disaster occurs, ERESS is possible to detect it quickly using through the movement of people. The mobile terminal analyzes behavior and location of indoor pedestrian. And it sends the result to the server. The server determines whether an emergency situation occurred or not based on the received transmission information. When an emergency situation occurs, the server will notify it to the mobile terminal. Then, indoor pedestrian conduct emergency evacuation using mobile terminal.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.23 no.1
• /
• pp.35-45
• /
• 2013

A autonomous vehicle requires improved and robust perception systems than conventional perception systems of intelligent vehicles. In particular, single sensor based perception systems have been widely studied by using cameras and laser radar sensors which are the most representative sensors for perception by providing object information such as distance information and object features. The distance information of the laser radar sensor is used for road environment perception of road structures, vehicles, and pedestrians. The image information of the camera is used for visual recognition such as lanes, crosswalks, and traffic signs. However, single sensor based perception systems suffer from false positives and true negatives which are caused by sensor limitations and road environments. Accordingly, information fusion systems are essentially required to ensure the robustness and stability of perception systems in harsh environments. This paper describes a perception system for autonomous vehicles, which performs information fusion to recognize road environments. Particularly, vision and laser radar sensors are fused together to detect lanes, crosswalks, and obstacles. The proposed perception system was validated on various roads and environmental conditions with an autonomous vehicle.