사물인터넷융복합논문지 (Journal of Internet of Things and Convergence)

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

DOI QR Code

그래프 컨벌루션 네트워크 기반 주거지역 감시시스템의 얼굴인식 알고리즘 개선

Improvement of Face Recognition Algorithm for Residential Area Surveillance System Based on Graph Convolution Network

  • 담하의 (목원대학교 정보통신용합공학부) ;
  • 민병원 (목원대학교 정보통신용합공학부)
  • Tan Heyi (Division of Information and Communication Convergence Engineering, Mokwon University) ;
  • Byung-Won Min (Division of Information and Communication Convergence Engineering, Mokwon University)
  • 투고 : 2024.02.18
  • 심사 : 2024.03.15
  • 발행 : 2024.04.30
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

스마트 지역사회의 구축은 지역사회의 안전을 보장하는 새로운 방법이자 중요한 조치이다. 촬영 각도로 인한 얼굴 기형 및 기타 외부 요인의 영향으로 인한 신원 인식 정확도 문제를 해결하기 위해 이 논문에서는 네트워크 모델을 구축할 때 전체 그래프 컨벌루션 모델을 설계하고, 그래프 컨벌루션 모델에 협력하여 얼굴의 핵심을 추출한다. 또한 얼굴의 핵심을 특정 규칙에 따라 핵심 포인트를 구축하며 이미지 컨벌루션 구조를 구축한 후 이미지 컨벌루션 모델을 추가하여 이미지 특징의 핵심을 개선한다. 마지막으로 두 사람의 얼굴의 이미지 특징 텐서를 계산하고 전체 연결 레이어를 사용하여 집계된 특징을 추출하고 판별하여 인원의 신원이 동일한지 여부를 결정한다. 최종적으로 다양한 실험과 테스트를 거쳐 이 글에서 설계한 네트워크의 얼굴 핵심 포인트에 대한 위치 정확도 AUC 지표는 300W 오픈 소스 데이터 세트에서 85.65%에 도달했다. 자체 구축 데이터 세트에서 88.92% 증가했다. 얼굴 인식 정확도 측면에서 이 글에서 제안한 IBUG 오픈 소스 데이터 세트에서 네트워크의 인식 정확도는 83.41% 증가했으며 자체 구축 데이터 세트의 인식 정확도는 96.74% 증가했다. 실험 결과는 이 글에서 설계된 네트워크가 얼굴을 모니터링하는 데 더 높은 탐지 및 인식 정확도를 가지고 있음을 보여준다.

The construction of smart communities is a new method and important measure to ensure the security of residential areas. In order to solve the problem of low accuracy in face recognition caused by distorting facial features due to monitoring camera angles and other external factors, this paper proposes the following optimization strategies in designing a face recognition network: firstly, a global graph convolution module is designed to encode facial features as graph nodes, and a multi-scale feature enhancement residual module is designed to extract facial keypoint features in conjunction with the global graph convolution module. Secondly, after obtaining facial keypoints, they are constructed as a directed graph structure, and graph attention mechanisms are used to enhance the representation power of graph features. Finally, tensor computations are performed on the graph features of two faces, and the aggregated features are extracted and discriminated by a fully connected layer to determine whether the individuals' identities are the same. Through various experimental tests, the network designed in this paper achieves an AUC index of 85.65% for facial keypoint localization on the 300W public dataset and 88.92% on a self-built dataset. In terms of face recognition accuracy, the proposed network achieves an accuracy of 83.41% on the IBUG public dataset and 96.74% on a self-built dataset. Experimental results demonstrate that the network designed in this paper exhibits high detection and recognition accuracy for faces in surveillance videos.

키워드

참고문헌

  1. J.Zhang, S.F.Liu, , L.FENG and Y. Zhang. "Research on Face Recognition Algorithm with Fusion Attention Mechanism". Foreign Electronic Measurement Technology, vol.42,No.02, pp.107-113,2023.
  2. Y.H.Fan, Y.Z.Wang, X.F.Yan, L.L.Gong, Y.W.Guo and M.Q.Wei,. "Low-light Image Enhancement Algorithm Driven by Face Recognition Task". Journal of Graphics, Vol.43, No.06, pp.1170-1181, 2022
  3. J.Feng, M.H.Gu, X.M.Liu and L.Cui, "Improving MobileNetv3 Face Recognition Algorithm in Classroom Scenarios." Foreign Electronic Measurement Technology, Vol.41, No.10, pp.47-55, 2022
  4. Boutros F, Damer N and Kirchbuchner F ,et al.Self-restrained triplet loss for accurate masked face recognition[J]. Pattern Recognition, 2022, 124:108473-. DOI:10.1016/j.patcog.2021.108473.
  5. Chen Z, Chen J and Ding G ,et al.A lightweight CNN-based algorithm and implementation on embedded system for real-time face recognition. Multimedia systems, 2023.
  6. T.C.Zhu and X.B.Zhou, "A comprehensive review of deep learning-based methods for face recognition", Modern Computer, Vol.29, No.17, pp.36-40, 2023
  7. L.Di, J.h.Zhangi and J.z.Liang. "Face alignment combined with shape constraints and Gaussian heatmap". International Journal of Machine Learning and Cybernetics, Vol.14, NO.12, PP. 4311-4324, 2023 https://doi.org/10.1007/s13042-023-01895-6
  8. S.Y.Guan, L.Di,J and Z.Liang ,"Face feature point localization algorithm with adaptive weighting of occlusion", Journal of Miniaturization of Microsystem, Vol.44, No.12, pp.2773-2783, 2023
  9. Y.Q.li, J.Y.Cai, X.Z.main and L.H.Lu, "Face alignment and reconstruction algorithm based on encoder-decoder network". Journal of Computer System Applications, Vol.30,No.07,pp.184-189,2021
  10. L.M.Han, C.Yang, Q.Li, B.Yao, Z.X.Jiao and Q.Y.Xie. "Dynamic deformable transformer for end-to-end face alignment". IET Computer Vision, Vol.17, No.8, pp.948-961, 2023 https://doi.org/10.1049/cvi2.12208
  11. X.Q.Yang and J.W.Mo, "End-to-end face alignment algorithm based on deep convolutional neural network", Computer Engineering and Design, Vol.40, No.09, pp.2666-2671+2717, 2019
  12. Y.H.Li, X.D.Liu, J.L.Chen and X.P.Su, "Accurate face alignment algorithm based on K-means", Sensors and Microsystems, Vol.40, No.03, pp.120-122+126, 2021
  13. J.Q.Liu, W.X.Tu and E.Zhu, "A comprehensive review of graph convolutional neural networks", Journal of Computer Engineering and Science, Vol.45, No.08, pp.1472-1481, 2023
  14. W.J.Li, J.Bai, B.Peng and Z.Y.Yang,. "A comprehensive review of graph convolutional neural networks and their applications in image recognition",Journal of Computer Engineering and Applications, Vol.59, No.22, pp.15-35, 2023.
  15. X.X.Zhang, Q.C.Tian, L.Lian and R.Tan, "A comprehensive review of facial keypoint detection", Journal of Computer Engineering and Applications, pp.1-13. 2023
  16. Jin H, Liao S and Shao L ."Pixel-in-Pixel Net: Towards Efficient Facial Landmark Detection in the Wild". International Journal of Computer Vision, NO.12, pp.129,2021. DOI:10.1007/s11263-021-01521-4.
  17. Lan X , Hu Q and Cheng J , "HIH: Towards More Accurate Face Alignment via Heatmap in Heatmap". 2021. DOI:10.48550/arXiv.2104.03100.
  18. Andres Prados-Torreblanca, Jose Miguel Buenaposada and Luis Baumela."Shape Preserving Facial Landmarks with Graph Attention Networks." 2022. DOI:10.48550/arXiv.2210.07233
  19. Xu Z, Li B and Yuan Y "AnchorFace: An Anchor-based Facial Landmark Detector Across Large Poses", National Conference on Artificial Intelligence.2021.
  20. Schroff F , Kalenichenko D and Philbin J . "FaceNet: A Unified Embedding for Face Recognition and Clustering". IEEE, 2015. DOI:10.1109/CVPR.2015.7298682.
  21. Wen Y , Liu W and Weller A , "SphereFace2: Binary Classification is All You Need for Deep Face Recognition", 2021. DOI:10.48550/arXiv.2108.01513.
  22. Wang H, Wang Y and Zhou Z, "CosFace: Large Margin Cosine Loss for Deep Face Recognition". computer vision and pattern recognition, pp.5265-5274. 2018
  23. Deng J, Guo J and Zafeiriou S , "ArcFace: Additive Angular Margin Loss for Deep Face Recognition", 2018. DOI:10.48550/arXiv.1801.07698.