Journal of the Korea Institute of Information and Communication Engineering (한국정보통신학회논문지)

The Korea Institute of Information and Commucation Engineering (한국정보통신학회)
권호 표지
DOI QR코드

DOI QR Code

An OpenPose-based Child Abuse Decision System using Surveillance Video

감시 영상을 활용한 OpenPose 기반 아동 학대 판단시스템

  • Yoo, Hye-Rim (Department of Electronics, Information and Communications Engineering, Daejeon University) ;
  • Lee, Bong-Hwan (Department of Electronics, Information and Communications Engineering, Daejeon University)
  • Received : 2018.12.15
  • Accepted : 2019.02.22
  • Published : 2019.03.31
Download PDF
⟨ Previous Next ⟩

Abstract

Recently child abuse has occurred frequently in educational institutions such as daycare center and kindergarten. Therefore, government made it mandatory to install CCTVs, but it is not easy to inspect the CCTV images. In this paper, we propose a model for judging child abuse using CCTV images. First of all, child abuse is a physical abuse of children by adults, thus a model for classifying adults and children is needed. The existing Haar scheme uses the frontal image to classify adults and children. However, the OpenPose allows to classify adults and children regardless of frontal and side image. In this research, a child abuse judgment model was designed and implemented by applying characteristics of adult and child posture when a child was abused. Since the implemented system utilizes the currently installed CCTV image, it is possible to monitor the child abuse in real time without any additional installation, which enables us to cope with the abuse promptly.

최근 어린이집이나 유치원 등 교육기관에서 아동학대가 빈번히 발생하고 있다. 정부는 CCTV 설치를 의무화하였지만 CCTV 영상을 열람하는 것이 쉽지 않다. 본 논문에서는 CCTV 영상을 이용하여 아동학대를 판단하는 모델을 제안한다. 먼저 아동학대란 성인이 물리적으로 아동에게 가해를 하는 것이므로 성인과 아동을 분류하는 모델이 필요하다. 기존의 Haar기법을 사용하여 성인과 아동을 분류하려면 정면 영상이 필요하지만 OpenPose를 사용하면 정면과 측면에 구애받지 않고 성인과 아동을 분류할 수 있다. 본 연구에서는 아동이 학대를 당할 때 성인과 아동의 자세의 특성을 적용하여 아동 학대 판단 모델을 설계 및 구현하였다. 구현한 시스템은 현재 설치되어있는 CCTV를 활용하므로 추가적인 설치가 필요하지 않고 실시간으로 아동학대가 발생하고 있는지 모니터링 할 수 있으므로 이에 따른 빠른 대처가 가능할 것으로 사료된다.

Keywords

HOJBC0_2019_v23n3_282_f0001.png 이미지

Fig. 1 Configuration of the OpenPose System

HOJBC0_2019_v23n3_282_f0002.png 이미지

Fig. 2 Joint extraction procedure in Open Pose

HOJBC0_2019_v23n3_282_f0003.png 이미지

Fig. 3 k-NN Algorithm (k = 1)

HOJBC0_2019_v23n3_282_f0004.png 이미지

Fig. 4 Decision Tree Algorithm (depth = 9)

HOJBC0_2019_v23n3_282_f0005.png 이미지

Fig. 5 A Child Abuse Judgment System

HOJBC0_2019_v23n3_282_f0006.png 이미지

Fig. 6 Joint Point using OpenPose

HOJBC0_2019_v23n3_282_f0007.png 이미지

Fig. 7 A json file for pose information

HOJBC0_2019_v23n3_282_f0008.png 이미지

Fig. 8 Rear image of an adult and side image of a child

HOJBC0_2019_v23n3_282_f0009.png 이미지

Fig. 9 Child abuse image

HOJBC0_2019_v23n3_282_f0010.png 이미지

Fig. 11 Performance results of the three algorithms for classifying adult and child (AC-Model)

HOJBC0_2019_v23n3_282_f0011.png 이미지

Fig. 12 Performance results of the three algorithms for abuse decision when SVM AC-Model is used

HOJBC0_2019_v23n3_282_f0012.png 이미지

Fig. 10 An ordinary children’s image

Table. 1 Machine Learning Algorithms

HOJBC0_2019_v23n3_282_t0001.png 이미지

Table. 2 Number of Images for training set

HOJBC0_2019_v23n3_282_t0002.png 이미지

Table. 3 Experimental Environment

HOJBC0_2019_v23n3_282_t0003.png 이미지

Table. 4 Performance comparison result of the proposed AC-Model

HOJBC0_2019_v23n3_282_t0004.png 이미지

Table. 5 Performance comparison result of the proposed CA-Model when SVM AC-Model is used

HOJBC0_2019_v23n3_282_t0005.png 이미지

References

  1. Central Child Protection Agency, "Child Abuse & Near Korea," 2018.
  2. B. W. Yoon, "A study on video based child and adult classification with biometry," M.S. thesis, Department of Electrical and Electronic Engineering, University of Kyungsung, Feb. 2015.
  3. Z. Cao, T. Simon, S. E. Wei, and Y. Sheikh, "Realtime multi-Person 2D Pose Estimation using Part Affinity Fields," in Proceeding of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 7291-7299, 2016.
  4. S. E. Wei, V. Ramakrishna, T. Kanade, and Y. Sheikh, "Convolutional Pose Machines," in Proceeding of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 4724-4732, 2016.
  5. S. Qiao, Y. Wang, and J. Li, "Real-time human gesture grading based on OpenPose," in Proceeding of the 2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, pp. 1-6, 2017.
  6. K. S. Ahn, "Comparative Experiment and Evaluation on Machine Learning based k-Nearest Neighbor and Support Vector Machine," M.S. thesis, Department of Information and Communication Engineering, University of Dankook, Dec. 2016.
  7. T. M. Cover, and P. E. Hart, "Nearest neighbor pattern classification," IEEE Transactions on Information Theory, vol. it-13, no. 1, pp. 21-27, 1967.
  8. Y. Tang, "Deep learning using linear support vector machines," Workshop on Challenges in Representation Learning, ICML, 2013.
  9. C. C. Chang, and C. J. Lin, "LIBSVM: A library for support vector machines," Transaction on Intelligent Systems and Technology, vol. 2(3), no. 27, Nov. 2011.
  10. S. Tong, and D. Koller, "Support Vector Machine Active Learning with Applications to Text Classification," Journal of Machine Learning Research, vol.2(1), pp. 45-66, Nov. 2002.
  11. A. Papagelis, and D. Kalles, "Breeding Decision Trees Using Evolutionary Techniques," in Proceeding of the Eighteenth International Conference on Machine Learning, pp. 93-400, 2001.
  12. CMU-Perceptual-Computing-La. OpenPose [Internet]. Available: https://github.com/CMU-Perceptual-Computing-Lab/openpose.
  13. A. C. Muller and S. Guido, Machine Learning with Python, 1st ed. Germany, O'Reilly Media Inc, 2016.
  14. Image Set. [Internet]. Available: https://image-net.org/.

Cited by

  1. 수직 히스토그램 기반 그림자 제거 알고리즘을 이용한 영상 감지 시스템 설계 및 구현 vol.24, pp.1, 2019, https://doi.org/10.6109/jkiice.2020.24.1.91