Journal of the Korean Society of Industry Convergence (한국산업융합학회 논문집)

The Korean Society of Industry Convergence (한국산업융합학회)
권호 표지
DOI QR코드

DOI QR Code

Optimization of Pose Estimation Model based on Genetic Algorithms for Anomaly Detection in Unmanned Stores

무인점포 이상행동 인식을 위한 유전 알고리즘 기반 자세 추정 모델 최적화

  • 이상협 (경성대학교 전자공학과) ;
  • 박장식 (경성대학교 전자공학과)
  • Received : 2022.11.10
  • Accepted : 2022.12.30
  • Published : 2023.02.28
Download PDF
⟨ Previous Next ⟩

Abstract

In this paper, we propose an optimization of a pose estimation deep learning model for recognition of abnormal behavior in unmanned stores using radio frequencies. The radio frequency use millimeter wave in the 30 GHz to 300 GHz band. Due to the short wavelength and strong straightness, it is a frequency with less grayness and less interference due to radio absorption on the object. A millimeter wave radar is used to solve the problem of personal information infringement that may occur in conventional CCTV image-based pose estimation. Deep learning-based pose estimation models generally use convolution neural networks. The convolution neural network is a combination of convolution layers and pooling layers of different types, and there are many cases of convolution filter size, number, and convolution operations, and more cases of combining components. Therefore, it is difficult to find the structure and components of the optimal posture estimation model for input data. Compared with conventional millimeter wave-based posture estimation studies, it is possible to explore the structure and components of the optimal posture estimation model for input data using genetic algorithms, and the performance of optimizing the proposed posture estimation model is excellent. Data are collected for actual unmanned stores, and point cloud data and three-dimensional keypoint information of Kinect Azure are collected using millimeter wave radar for collapse and property damage occurring in unmanned stores. As a result of the experiment, it was confirmed that the error was moored compared to the conventional posture estimation model.

Keywords

Acknowledgement

본 논문은 부산광역시 및 (재)부산인재평생교육진흥원의 BB21플러스 사업 지원과 2022년도 인공지능 학습용 데이터 구축 지원사업의 2-97. 실내(편의점, 매장) 사람행동 영상 데이터 과제의 연구 결과임.

References

  1. 이진천, "무인점포 시대의 도래," 설비저널, vol. 47, no. 12, pp. 144-145, 2018.
  2. 김건희, 장철영, "빅데이터를 이용한 무인점포 범죄 연구." Crisisonomy, vol. 18, no. 9, pp. 95-110, 2022. https://doi.org/10.14251/crisisonomy.2022.18.9.95
  3. 이지애, 문성철, "스마트 관제를 위한 딥러닝 기반 이상행동 기술 동향 분석." 한국방송미디어공학회 학술대회, vol. 29, no. 8, pp. 14-16, 2019.
  4. S.-J. Horng and P.-S. Huang, "Building Unmanned Store Identification Systems Using YOLOv4 And Siamese Network."Applied Sciences, vol. 12, no. 8, pp. 3826-3843, 2022.
  5. D. Greene, "Facial recognition technology 'inherently biased' says Google cloud boss." Google, 2018.
  6. 개인정보침해 조사팀, "얼굴인식 기술의 최근 개인정보보호 논란 관련 동향 분석." KISA 동향보고서, pp. 1-10, 2019.
  7. B. Raj N., A. Subramanian, K. Ravichandran, and N. Venkateswaran, "Exploring Techniques To Improve Activity Recognition Using Human Pose Skeletons." 2020 IEEE Winter Applications of Computer Vision Workshops (WACVW), pp. 165-172, 2020.
  8. S. An and U. Y. Ogras, "MARS: mmWave-Based Assistive Rehabilitation System For Smart Healthcare." ACM Transactions on Embedded Computing Systems, vol. 20, no. 5, pp. 1-22, 2021. https://doi.org/10.1145/3477003
  9. A. Sengupta, F. Jin, R. Zhang, and S. Cao, "mm-Pose: Real-Time Human Skeletal Posture Estimation Using mmWave Radars And CNNs." IEEE Sensors Journal, vol. 20, no. 17, pp. 10032-10044, 2020. https://doi.org/10.1109/jsen.2020.2991741
  10. A. Sengupta and S. Cao, "mmPose-NLP: A Natural Language Processing Approach To Precise Skeletal Pose Estimation Using mmWave Radars." IEEE Transactions on Neural Networks and Learning Systems, pp. 1-12, 2022.
  11. S. Katoch, S. S. Chauhan, and V. Kumar, "A Review On Genetic Algorithm: Past, Present, And Future." Multimedia Tools and Applications, vol. 80, no. 5, pp. 8091-8126, 2020. https://doi.org/10.1007/s11042-020-10139-6
  12. T. Elsken, J. H. Metzen, and F. Hutter, "Neural Architecture Search." Automated Machine Learning, vol. 20, no. 1, pp. 63-77, 2019. https://doi.org/10.1007/978-3-030-05318-5_3