사물인터넷융복합논문지 (Journal of Internet of Things and Convergence)

한국사물인터넷학회 (The Korea Internet of Things Society)
권호 표지
DOI QR코드

DOI QR Code

사물인터넷과 Amazon Web Services를 활용한 공간 혼잡도 측정 시스템

Spatial Crowdedness Measurement System using IoT and Amazon Web Services

  • 김태국 (국립부경대학교 컴퓨터.인공지능공학부)
  • Tae-Kook Kim (School of Computer and Artificial Intelligence Engineering, Pukyong National University)
  • 투고 : 2023.06.01
  • 심사 : 2023.08.10
  • 발행 : 2023.08.31
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

본 논문에서는 사물인터넷과 Amazon Web Services(아마존 웹 서비스, AWS)를 활용한 공간 혼잡도 측정 시스템에 관해 연구하였다. 기존의 공간 혼잡도 측정 시스템은 고가의 서버를 필요로 하고, 공간 인식 성능 개선을 위한 고비용의 알고리즘 개발 및 업데이트가 필요한 문제가 있다. 제안한 시스템은 소형/저가형 싱글 보드 컴퓨터인 라즈베리 파이(Raspberry Pi)에서 OpenCV를 통해 웹카메라(or CCTV) 화면을 캡쳐하여 AWS로 정보를 전송하고 처리하여 적은 비용으로 구현이 가능하다. AWS에서는 수신된 공간 화면 이미지 정보를 Amazon S3(Amazon Simple Storage Services)에 저장하고, Amazon Lambda에서 Amazon Rekognition으로 전송하여 이미지를 통해 혼잡도를 분석한다. Amazon Rekognition 서비스는 이미지 처리 건당 0.001달러로 적은 비용으로 인공지능(AI) 기술을 사용할 수 있고, 사람 객체 인식을 통해 혼잡도를 분석할 수 있다. 분석된 혼잡도는 DB(Database)에 저장하고, 결과를 화면에 출력한다. 제안한 공간 혼잡도 측정 시스템은 저비용으로 공공장소 등에서의 공간 혼잡도 확인 등에 활용될 수 있을 것으로 기대한다.

In this paper, we conducted research on a spatial crowdedness measurement system using the Internet of Things(IoT) and Amazon Web Services(AWS). Current spatial congestion measurement systems require expensive servers and entail significant investment in algorithm development and updates to enhance spatial recognition performance. The proposed system can be implemented at low cost by capturing the screen of a web camera(or CCTV) through OpenCV on a small/low-cost single board computer, Raspberry Pi, and transmitting and processing the information to AWS. Within AWS, the received spatial image information is stored in Amazon S3(Amazon Simple Storage Service). Subsequently, Amazon Lambda transfers the images to Amazon Rekognition for congestion analysis based on the images. The Amazon Rekognition service offers the capability to utilize artificial intelligence(AI) technology for image processing at a low cost of 0.001 dollars per image. Through human object recognition, it enables congestion analysis. The analyzed congestion level is stored in DB(Database), and the result is displayed on the screen. The proposed system is expected to be used for checking crowdedness at low cost in public places.

키워드

과제정보

이 성과는 정부(과학기술정보통신부)의 재원으로 한국연구재단의 지원을 받아 수행된 연구임(RS-2023-00242528).

참고문헌

  1. J.G.Kim, Y.J.Kim, K.H.Park, S.W.Jeon, T.K.Kim "A Study on the Congestion Measurement System of Public Transportation Using CCTV," 2022 Korea Multimedia Society Spring conference, Vol.25, No.1, pp.25-27, 2022. 
  2. S.Lee, H.Yoon, T.Kim, K.Hong, "Research on the Analysis of Changes in Train Congestion of Urban Railway according to COVID-19 Situation," Journal of Korean Society for Urban Railway, Vol.9, No.4, pp.1131-1141, 2021.  https://doi.org/10.24284/JKOSUR.2021.12.9.4.1131
  3. Y.G.Kim, T.D.Kim, Y.S.Kim, S.H.Lee, K.I.Jung, N.J.Kim, J.A.Oh, S.M.Chang, "Urban rail passenger congestion measurement practice tests for data validation," 2016 Korean Society for Urban Railway Spring conference, pp.27-32, 2016. 
  4. A.Bakht, H.Lee, "Deep Learning Framework for Spatial Crowdedness Estimation and Comparison Analysis with Machine Learning," Journal of Korean Institute of Intelligent Systems, Vol.32, No.1, pp.76-85, 2022.  https://doi.org/10.5391/JKIIS.2022.32.1.76
  5. C.J.Jeong, K.Y.Park, G.Park, "Real Time Crowd Estimation System Using Embedded Hardware," Journal of Satellite, Information and Communications (kosst), Vol.8, No.4, pp.26-29, 2013. 
  6. Amazon, Amazon Web Services[Internet], https://aws.amazon.com. 
  7. Amazon, Amazon Web Services Rekognition[Internet], https://aws.amazon.com/ko/rekognition/?nc2=h_ql_prod_ml_rek. 
  8. Amazon, Amazon Web Services S3[Internet], https://aws.amazon.com/ko/s3/?nc2=h_ql_prod_st_s3. 
  9. Amazon, Amazon Web Services Lambda[Internet], https://aws.amazon.com/ko/lambda/?nc2=h_ql_prod_fs_lbd. 
  10. Amazon, Amazon Web Services RDS[Internet], https://aws.amazon.com/ko/rds/?nc2=h_ql_prod_fs_rds. 
  11. J.H.Moon, B.Peng, J.H.Kwon, T.K.Kim, "Implementation of Smart Umbrella Stand Based on IoT," Journal of Internet of Things and Convergence, Vol.9, No.1, pp.57-64, 2023.  https://doi.org/10.20465/KIOTS.2023.9.1.057
  12. Raspberry Pi Foundation, Raspberry Pi[Internet], https://www.raspberrypi.com. 
  13. Intel, OpenCV[Internet], https://opencv.org. 
  14. D.H.Kim, S.Y.Kim, "A Study on Risk Situation Awareness Using OpenCV," Journal of The KIECS, Vol.16, No.2, pp.211-218, 2021. 
  15. D.J.Kim, W.S.Choi, S.P.Ju, S.M.Yoo, J.Y.Choi, H.K.Park, "Smart Streetlight based on Accident Recognition using Raspberry Pi Camera OpenCV," The Journal of The Korea Institute of Electronic Communication Sciences (Electronic Communication), Vol.17, No.6, pp.1229-1236, 2022.