바로가기메뉴

본문 바로가기 주메뉴 바로가기

ACOMS+ 및 학술지 리포지터리 설명회

  • 한국과학기술정보연구원(KISTI) 서울분원 대회의실(별관 3층)
  • 2024년 07월 03일(수) 13:30
 

다중 CCTV 사물인터넷 환경에서의 객체 추적 기법

A Scheme on Object Tracking Techniques in Multiple CCTV IoT Environments

한국사물인터넷학회논문지 / Journal of The Korea Internet of Things Society, (P)2466-0078;
2019, v.5 no.1, pp.7-11
https://doi.org/https://doi.org/10.20465/kiots.2019.5.1.007
홍지훈 (백석대학교)
이근호 (백석대학교)
  • 다운로드 수
  • 조회수

초록

본 연구는 최근 전국적으로 계속해서 사물인터넷 CCTV의 설치 대수가 증가함에 따라 CCTV의 활용범위를넓히고자 CCTV를 통하여 범죄 의심자 또는 이상 행동자를 추적하는 방법을 제안한다. 이상 행동 구분은 기존에 나와있던 연구들을 활용하여 범죄 의심자 또는 이상 행동자를 색출해내고 CNN을 활용하여 대상을 객체와 하여 추적을하고 주변 CCTV를 서로 네트워크로 연결하여 객체화된 대상의 이동 경로를 예측해 해당 경로 근방의 CCTV들에객체의 샘플 데이터를 공유하여 대상 판별 및 해당 대상을 추적하는 방식을 이용하였다. 해당 연구를 통하여 추적하기 힘든 범죄자의 위치를 추적하여 국가 치안에 기여하고 더욱 다양한 기술들이 CCTV에 접목될 수 있도록 지속적인연구가 필요하다.

keywords
Physical security, artificial intelligence, CNN algorithm, object detection, CCTV, 물리 보안, 인공지능, 씨앤앤알고리즘, 객체 검출, 씨씨티비

Abstract

This study suggests a methodology to track crime suspects or anomalies through CCTV in order to expand the scope of CCTV use as the number of CCTV installations continues to increase nationwide in recent years. For the abnormal behavior classification, we use the existing studies to find out suspected criminals or abnormal actors, use CNN to track objects, and connect the surrounding CCTVs to each other to predict the movement path of objectified objects CCTVs in the vicinity of the path were used to share objects' sample data to track objects and to track objects. Through this research, we will keep track of criminals who can not be traced, contribute to the national security, and continue to study them so that more diverse technologies can be applied to CCTV.

keywords
Physical security, artificial intelligence, CNN algorithm, object detection, CCTV, 물리 보안, 인공지능, 씨앤앤알고리즘, 객체 검출, 씨씨티비

참고문헌

1.

Y.K.Son and T.H.Kim, “Practical use plan of equipment security system for efficient crime prevention (CCTV system in priority,” KOREAN INSTITUTE OF FIRE SCIENCE & ENGINEERING, pp.393-399. 2003.

2.

H.J.Kyung, “A Study on Establishment and Management of the Crime Prevention CCTV”, The korean Association for Public Society, Vol.8, No.4, pp.109-137, 2018.

3.

H.S.Young and M.T.Heon, “An Analysis on the CCTV Location Appropriateness and Effectiveness for the Crime Prevention”. The Korean Association Of Regional Geographers, Vol.21, No.4, pp.739-750, 2015.

4.

D.W.Kim, B.J.Park and S.K.Oh, “The Implementation of the structure and algorithm of Fuzzy Self-organizing Neural Networks(FSONN) based on CNN” Korean Institute of Intelligent Systems, Vol.10, No.1, pp.114-117. 2000.

5.

B.J.Park, S.K.Oh and H.K.Kim. “The Hybrid Multi-layer Inference Architectures and Algorithms of FPNN Based on CNN and PNN.” The Korean Institute of Electrical Engineers, Vol.49, No.7, pp.378-388. 2000.

6.

J.H.Kim, W.C.Gyun, K.H.Park and Y.H.Kim, “Shadow Detection for the Accuracy of Object Detection in CCTV Image”, Korean Institute of Information Technology, pp.191-193, 2018.

7.

S.H.Lee and M.S.Kang, “Design of Efficient Object Detection System Using Object Recognition Technology”. The Institute of Electronics and Information Engineers, pp.829-831. 2018.

8.

S.K.Kim, E.Dinesh, M.S.Sung and Y.H.Joo. “Connectivity Based Object Tracking Model for Intelligent Video Surveillance”, The Korean Institute of Electrical Engineers, pp.1131-1132. 2018.

9.

T.Y.Nam and D.W.Jung, “Design of KLT Tracker for Real-time Object Detection and Tracking using System-on-Chip”, The Korean Society for Aeronautical & Space Sciences, pp.805-807, 2016.

10.

10] Jonathan. Hui. “Object detection: speed and accuracy comparison (Faster R-CNN, R-FCN, SSD, FPN, RetinaNet and YOLOv3)”, https://medium.com/@jonathan_hui/object-detection-spee d-and-accuracy-comparison-faster-r-cnn-r-fcn-ssd-and -yolo-5425656ae359. 2018.

11.

J.H.Elder, J.D.Prince, Y.Hou, M.Sizintsev and E.Olevskiy, “Pre-Attoentive and Attentive Detection of Humans in Wide-Filed Scence”, Intematinal Journal of Computer, Vol.72, No.1,pp.47-66. 2007.

12.

12] W.J.Lee and B.H.Lee, “Multiple Object Detection and Tracking System robust to various Environment”. The Institute of Electronics and Information Engineers, Vol..46 No.6, pp.88-94, 2009.

13.

J.W.Park and S.Y.Kwak. “Crowed abnormal behaviors detection for video surveillance systems”. Korea Institute Of Communication Sciences, pp.376-377, 2014.

14.

Y.J.Jung and Y.g.Yoon., “A Study on Abnormal Behavior Analysis and Pattern Prediction using Multi-object”. The Korean Institute of Information Scientists and Engineers, pp.440-441. 2014.

15.

S.W.Lee and T.Y.Kim, J.H.Yoo, J.K.Paik. “Abnormal Behavior Detection Based on Adaptive Background Generation for Intelligent Video Analysis”. The Institute of Electronics and Information Engineers, Vol.48. No.1, pp.111-121. 2011.

한국사물인터넷학회논문지