• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.07a
• /
• pp.17-20
• /
• 2021

본 연구는 인원 계수에 딥러닝 알고리즘을 적용한다. 인원 계수는 안전 관리 분야, 상업 분야에 적용될 수 있다. 예를 들어, 건물 내 화재 발생 시, 계수된 인원을 활용하여 인명 피해를 최소화할 수 있다. 다른 예로, 유동인구 데이터를 기반으로 상권을 분석하여 경제적 효율성을 극대화할 수 있다. 이처럼 인원 데이터의 중요성이 증가함에 따라 인원 계수 연구도 활발하다. 그 예로, 객체 탐지(Object Detection) 같은 딥러닝 기반 인원 계수, 센서 기반 인원 계수 등이 있다. 본 연구에선 딥러닝 알고리즘인 VGGNet을 사용하여 인원을 계수했다. 결과로 Mean Absolute Percentage Error(이하 MAPE)는 약 5.9%의 오차율을 보였다. 결과 확인 방법으로는 설명 가능한 인공지능(XAI) 알고리즘 중 하나인 Grad-CAM을 적용했다.

• Journal of the Health Care and Life Science
• /
• v.8 no.2
• /
• pp.121-125
• /
• 2020

People counting has always been a method of interest for maximizing energy saving by identifying the congestion level or amount of use of a specific facility to efficiently manage the facility, or automatically implementing a power saving function by identifying the number of people entering and exiting a specific place such as a toilet. The method of counting people by image processing is very expensive and has the disadvantage of being severely affected by the surrounding environment of the lighting. In the case of the area sensor, there is a disadvantage of counting as one person when the number of people passes close with arms folded. In order to solve the existing method, which is expensive, affected by lighting, or inaccurate the number of people in certain cases, this paper proposes a new method of counting people using the principle of LiADAR. Accurate counting of the number of people entering the hospital will help manage hospital facilities, but it will also help to establish effective quarantine measures at the present time when Corona 19 is prevalent.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.7
• /
• pp.1441-1446
• /
• 2017

In recent, the studies on the systems have been progressing that count the number of people passing through passageway or count people who exist in specific space. The existing people counting systems count the number of people using ultrared sensors, ultrasonic sensors or camera sensors, which can be installed only on pillar or around wall. Though ultrared sensors and ultrasonic sensors is low cost, they are inadequate to detect incoming/outgoing when several pedestrians pass through passageway concurrently. In this paper, we designed a sensor mat using piezoelectric sensors to complement the above-mentioned disadvantage. Also, we implemented the system that detects direction of progress and counts the number of people. The sensor mat detects direction of progress using pressure given on sensors and timing information and counts the number of people when pedestrians pass through on a sensor mat.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.10
• /
• pp.1361-1368
• /
• 2020

The system for counting the number of people has traditionally been implemented in various ways. Existing systems include infrared sensors, lasers, cameras, etc. In the case of such an existing system, there are restrictions on space such as ceilings and sides of walls. In this paper, we propose a method of detecting the footsteps of pedestrians using a piezoelectric mat containing a number of piezoelectric sensors and counting the number of pedestrians passing simultaneously by using the data collected from the piezoelectric mat. When pedestrians pass over piezoelectric mats, the collected sensor data was aggregated using SPI communication and transmitted to PC server using TCP/IP communication. Performance analysis shows that approximately 600 step data can be recognized with 99% accuracy. This is to overcome the shortcomings of other counting systems.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.30 no.1
• /
• pp.28-37
• /
• 2019

In this paper, we propose a real-time system for counting people. The proposed system uses an impulse radio ultra-wideband(IR-UWB) radar to estimate the number of people in a given location. The proposed system uses learning-based classification methods to count people more accurately. In other words, a feature vector database is constructed by exploiting the pattern of reflected signals, which depends on the number of people. Subsequently, a classifier is trained using this database. When a newly received signal data is acquired, the system automatically counts people using the pre-trained classifier. We validated the effectiveness of the proposed algorithm by presenting the results of real-time estimation of the number of people changing from 0 to 10 in an indoor environment.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.43 no.5
• /
• pp.667-674
• /
• 2023

In the event of a disaster occurring within a building, the prompt and efficient evacuation and rescue of occupants within the building becomes the foremost priority to minimize casualties. For the purpose of such rescue operations, it is essential to ascertain the distribution of individuals within the building. Nevertheless, there is a primary dependence on accounts provided by pertinent individuals like building proprietors or security staff, alongside fundamental data encompassing floor dimensions and maximum capacity. Consequently, accurate determination of the number of occupants within the building holds paramount significance in reducing uncertainties at the site and facilitating effective rescue activities during the golden hour. This research introduces a methodology employing computer vision algorithms to count the number of occupants within distinct building locations based on images captured by installed multiple CCTV cameras. The counting methodology consists of three stages: (1) establishing virtual Lines of Interest (LOI) for each camera to construct a multi-camera network environment, (2) detecting and tracking people within the monitoring area using deep learning, and (3) aggregating counts across the multi-camera network. The proposed methodology was validated through experiments conducted in a five-story building with the average accurary of 89.9% and the average MAE of 0.178 and RMSE of 0.339, and the advantages of using multiple cameras for occupant counting were explained. This paper showed the potential of the proposed methodology for more effective and timely disaster management through common surveillance systems by providing prompt occupancy information.

• Proceedings of the Korean Information Science Society Conference
• /
• 2006.10d
• /
• pp.95-100
• /
• 2006

최근 반도체 기술과 무선 통신기술의 발전으로 센서 네트워크를 이용한 응용 분야가 다양하게 나타나고 있다. 본 논문에서는 센서 네트워크를 이용하여 산악지역의 자동인원계수시스템과 환경모니터링시스템을 개발했다. 이 시스템은 저전력으로 동작하고 이중 RF 보드를 이용하여 최대한 신뢰성 있는 데이터 전송이 가능하도록 개발하였다. 또한 산악지역에서 데이터 전송율과 통신 신뢰성을 만족시킬 수 있는 433MHz 대역이 적합함을 실험을 통하여 확인하였다. 기존의 CCD 카메라를 이용한 자동인원계수기는 전원공급과 네트워크 부재로 인하여 산악지역에서는 사용하기에 부적합하다. 본 논문에서는 산악지역의 탐방로 폭이 협소함을 이용하여 포토 빔 센서를 이용한 양방향 계수센서를 개발하였다. 산악지역에서 실험한 결과 1.5m 이하의 도로 폭에서 95% 이상의 계수 신뢰도를 보임을 확인할 수 있다.

• The Journal of the Korea Contents Association
• /
• v.9 no.12
• /
• pp.31-38
• /
• 2009

The objective of this paper is to propose the bus information system that provides bus passenger number. Recently, we use a bus conveniently through introduction of bus information system offering location-based information. However, we need additional information for decision making in order to use a bus comfortably. To do this, first, we analyzed the needs of providing passenger number; second, we adopted GPS and RFID type's transportation card to get location and passenger number. Finally, we implemented the system displaying that information by three types, and tested it. Our proposed system has advantages to use a bus comfortably and to manage a bus efficiently by providing bus passenger number which is the important element for decision making as well as location based information provided by existing bus information systems.

• Journal of the Korea Organic Resources Recycling Association
• /
• v.10 no.3
• /
• pp.92-97
• /
• 2002

This study is to analyze collection factor of food wastes using GIS. For calculation of collection crews and equipment, It was measured velocity(43m/min), moving coefficient(1.9), moving distance(17,200/day), collection rate(0.313kg/day) on the basis of S2Dong. Moving coefficient was able to analyze straight distance of S2Dong using GIS and other areas were also applied to analyze straight distance of each area using same method. At N autonomous district, 10 persons of collection crews were necessary, but collection of food wastes by four forces resulted to be 11 persons of crews necessary. 7 vehicles were necessary, but 8 vehicles were needed by same reason.

• 한국신재생에너지학회:학술대회논문집
• /
• 2009.11a
• /
• pp.264-269
• /
• 2009

신재생에너지기술 개발인력 확보는 국가의 지속적인 성장을 가능하게 하는 주요 요인이다. 본 연구에서는 신재생에너지인력양성이 원활하게 공급되지 않았을 경우 발생할 수 있는 인력공급지장효과를 분석하기 위하여 산업연관분석의 공급유도모형을 유도하고 2006년 신재생에너지 인력양성사업 배출인원을 기준으로 실증분석을 실시하였다. 또한 공급유도형의 감응도 계수와 영향력계수 분석을 통해 신재생에너지 관련 사업과 타 산업간의 전후방연쇄효과를 비교분석하였다. 연구 수행 결과 신재생에너지 관련 산업의 평균 영향력 계수는 1.37, 평균 감응도 계수는 0.96으로 최종 수요적 제조업의 성격을 띠고 있는 것으로 나타났다. 또한, 2006년 신재생에너지 인력양성사업에 의해 배출된 336명의 인력이 공급되지 않았을 경우 공급지장비용은 총 230억이며, 8개 신재생에너지원 중 태양열과 연료전지에 인력 공급지장비용이 타 에너지원에 비해 상대적으로 높은 것으로 분석되었다. 산업분류 기준으로는 일반목적용 기계산업과 전기기계 장치 산업이 신재생 에너지 인력공급 장애에 따른 지장비용이 높은 것으로 분석되었다.