The Journal of the Korea institute of electronic communication sciences (한국전자통신학회논문지)

Korea Institute of Electronic Communication Science (한국전자통신학회)
권호 표지
DOI QR코드

DOI QR Code

An Automatic Data Collection System for Human Pose using Edge Devices and Camera-Based Sensor Fusion

엣지 디바이스와 카메라 센서 퓨전을 활용한 사람 자세 데이터 자동 수집 시스템

  • Received : 2023.11.02
  • Accepted : 2024.02.17
  • Published : 2024.02.29
Download PDF
⟨ Previous Next ⟩

Abstract

Frequent false positives alarm from the Intelligent Selective Control System have raised significant concerns. These persistent issues have led to declines in operational efficiency and market credibility among agents. Developing a new model or replacing the existing one to mitigate false positives alarm entails substantial opportunity costs; hence, improving the quality of the training dataset is pragmatic. However, smaller organizations face challenges with inadequate capabilities in dataset collection and refinement. This paper proposes an automatic human pose data collection system centered around a human pose estimation model, utilizing camera-based sensor fusion techniques and edge devices. The system facilitates the direct collection and real-time processing of field data at the network periphery, distributing the computational load that typically centralizes. Additionally, by directly labeling field data, it aids in constructing new training datasets.

지능형 선별 관제 시스템의 잦은 오탐지로 인해 관제 요원들의 업무 능률 및 시장 신뢰도 저하 문제가 꾸준히 보고되고 있다. 오탐지 문제 개선을 위해 새 AI 모델을 개발하거나 교체하는 것은 기회비용이 크므로, 훈련 데이터 세트 품질을 향상하여 문제를 개선하는 것이 현실적이다. 그러나 소규모 조직은 데이터 세트 수집 및 정제 역량이 부족한 실정이다. 이에 본 논문에서는 사람 자세 추정 모델을 중심으로 엣지 디바이스와 카메라 센서 퓨전을 활용한 사람 자세 데이터 자동 수집 시스템을 제안한다. 이 시스템은 네트워크 말단에서 현장 데이터를 직접 수집하고 레이블링하는 과정을 실시간으로 처리하도록 만들어, 중앙으로 집중되는 연산 부하를 분산시킨다. 또한 현장 데이터를 직접 레이블링하므로 새로운 훈련 데이터 구축에 도움을 준다.

Keywords

Acknowledgement

이 논문은 2023년도 교육부 재원으로 한국산업기술진흥원의 지원을 받아 수행된 연구임.

References

  1. S.-O. Yoon, M.-S. Kim, and G.-H. Seok,"Inter-Module Interworking Evaluation of TDMA-Based Wireless IP Video Transmission System," J. of The Korea Institute of Electronic Communication Sciences, vol. 18, no. 1, 2023, pp. 1-10.
  2. J.-S. Park, M. Wiranegara, and G.-Y. Son, "Multi-channel Video Analysis Based on Deep Learning for Video Surveillance," J. of The Korea Institute of Electronic Communication Sciences, vol. 13, no. 6, 2018, pp. 1263-1268.
  3. C.-S. Chung, "A Case Study on the Operation Enhancement of Integrated CCTV Control Center at Busan Metropolitan City," J. of Korean Associastion for Regional Information Society, vol. 18, no. 3, 2015, pp. 123-154.
  4. K.-S. Lim, and G.-W. Kim, "Cloud-based intelligent video surveillance platform for providing empirical-based deep learning video analysis technology," Review of Korea Institute of Information Security & Cryptology, vol. 29, no. 3, 2019, pp. 37-43.
  5. Y.-H. Kim, "Demonstration case study of deep learning-based smart city video surveillance solution in Osan," Information and Communications Mag., vol. 37, no. 5, 2020, pp. 42-48.
  6. H.-K. Lyu, and S. Yun, "Agricultural environment image data collection system development for artificial intelligence learning," Summer Annual Conf. of The Institute of Electronics and Information Engineers, Gwangju, Korea, 2020, pp. 887-888.
  7. D.-W. Jeong, S.-J. Yoo, Y.-H. Gu, K.-H. Lee, and J.-H. Park, "Pests Image Collect and Inspect System Design and Implementation for the Image-Based Pests Automatic Diagnosis," Summer Annual Conf. of The Institute of Electronics and Information Engineers, Jeju, Korea, 2016, pp. 1694-1697.
  8. E.-O. Joo, D.-Y. Kim, B.-S. Cho, and M.-S. Kim, "Implementation of Autonomous Driving Simulation System using 3D Spatial Data for Various Imagery Data Collection," J. of Korean Society for Geospatial Information Science, vol. 29, no. 3, 2021, pp. 3-12.
  9. T.-H. An, J. Kang, and K.-W. Min, "Analysis of the Effects of Data Augmentation Techniques with CycleGAN on Semantic Segmentation in Night and Rain Environments," Summer Annual Conf. of The Institute of Electronics and Information Engineers, Jeju, Korea, 2021, pp. 2237-2239.
  10. N.M. Fasial, and T. Helmy, "A Real-Time Deep Learning-based Smart Surveillance Using Fog Computing: A Complete Architecture," Procedia Computer Science, vol. 218, no. 107, 2023, pp. 1102-1111. https://doi.org/10.1016/j.procs.2023.01.089
  11. D.-H. Kim, and S.-Y. Kim, "A Study on Risk Situation Recognition Using OpenCV," J. of The Korea Institute of Electronic Communication Sciences, vol. 16, no. 2, 2021, pp. 211-217.