범죄예방을 위한 CCTV 영상 기반의 실시간 안면인식 시스템

다국어 초록 (Multilingual Abstract)

In modern society, the rate of sexual offenses continues to increase. In particular, since the recidivism rate of sex offenders is high, the institutional or social need to prevent recurrence is gradually increasing. In this paper, we propose a real-time criminal face recognition system in CCTV video for crime prevention. This is a series of processes to prevent sexual offenses and re-offenses of sex offenders through artificial intelligence-based facial recognition. However, there is a problem in practically utilizing facial recognition technology. The facial recognition system consists of a detection step for finding a face in an image and a matching step for calculating the degree of matching with the detected face. It requires a high computational cost compared to a single operation. Therefore, this study shows higher recognition accuracy by using the AB-box method, which reduces the processing time through vectorization when detecting a face. Also, we reduce computational cost during real-time operation and increase the size of the detected facial image afterward.

참고문헌 (Reference)

1 정신교, "성범죄자 신상공개의 예방적 효과" 한국법학회 (39) : 269-290, 2010

2 J. Kim, "Trends of Sex Offenses against Minors: Focus on Sex Offenders Subject to Personal Information Registration Due to Offenses Committed in 2017" 2018

3 S. Zhang, "Survey of Image Recognition Algorithms" IEEE 542-548, 2020

4 J. Deng, "RetinaFace: Single-Shot Multi-Level Face Localisation in the Wild" 5203-5212, 2020

5 B. Amos, "OpenFace: A general-purpose face recognition library with mobile applications" 6 (6): 2016

6 S. Gidaris, "Object Detection via a Multi-region and Semantic Segmentation-Aware CNN Model" 1134-1142, 2015

7 "Korea Ministry of Government Legislation"

8 "Kindicator"

9 KOSIS, "KOrean Statistical Information Service"

10 V. Le, "Interactive Facial Feature Localization" 7574 : 679-692, 2012

11 N. Dalal, "Histograms of oriented gradients for human detection" 1 : 886-893, 2005

12 F. Schroff, "FaceNet: A unified embedding for face recognition and clustering" IEEE 815-823, 2015

13 Y. Kortli, "Face Recognition Systems: A Survey" 20 (20): 342-, 2020

14 I. Masi, "Deep Face Recognition: A Survey" IEEE 471-478, 2018