[논문]YOLOv5 based Anomaly Detection for Subway Safety Management Using Dilated Convolution

Nusrat Jahan Tahira; Ju-Ryong Park; Seung-Jin Lim; Jang-Sik Park

doi:10.21289/ksic.2023.26.2.217

YOLOv5 based Anomaly Detection for Subway Safety Management Using Dilated Convolution 원문보기

한국산업융합학회 논문집 = Journal of the Korean Society of Industry Convergence, v.26 no.2/1, 2023년, pp.217 - 223

Nusrat Jahan Tahira (Dept. of Electrical, Electronics and Communications Engineering, Kyungsung University) , Ju-Ryong Park (Dept. of Electrical, Electronics and Communications Engineering, Kyungsung University) , Seung-Jin Lim (Busan Transportaion Corporation) , Jang-Sik Park (Dept. of Electronic Engineering, Kyungsung University)

Abstract ▼ AI-Helper

With the rapid advancement of technologies, need for different research fields where this technology can be used is also increasing. One of the most researched topic in computer vision is object detection, which has widely been implemented in various fields which include healthcare, video surveillance and education. The main goal of object detection is to identify and categorize all the objects in a target environment. Specifically, methods of object detection consist of a variety of significant techniq ues, such as image processing and patterns recognition. Anomaly detection is a part of object detection, anomalies can be found various scenarios for example crowded places such as subway stations. An abnormal event can be assumed as a variation from the conventional scene. Since the abnormal event does not occur frequently, the distribution of normal and abnormal events is thoroughly imbalanced. In terms of public safety, abnormal events should be avoided and therefore immediate action need to be taken. When abnormal events occur in certain places, real time detection is required to prevent and protect the safety of the people. To solve the above problems, we propose a modified YOLOv5 object detection algorithm by implementing dilated convolutional layers which achieved 97% mAP50 compared to other five different models of YOLOv5. In addition to this, we also created a simple mobile application to avail the abnormal event detection on mobile phones.

주제어

표/그림 (7)

그림 Fig. 1 Dilated convolution
그림 Fig. 2 Object detection framework
그림 Fig. 4 Dataset distribution
그림 Fig. 5 Comparison of F1-confidence curver (a) F1-Confidence Curve of YOLOv5n without dilation (b) F1-Confidence Curve of YOLOv5s with dilation
표 Table 1. Model layers and parameters
그림 Fig. 6 Comparison of confusion matrix (a) Confusion matrix without dilation (b) Confusion matrix with dilation
그림 Fig. 7 Model performance

AI 본문요약
AI-Helper

대상 데이터

The data was split into 73:17:10 corresponding to training, validation and testing respectively. Table 1 represents the model layers and parameters of the proposed model used in this experiment which is same as the five different YOLOv5 models.

성능/효과

As abnormal events do not occur frequently, it is hard to detect abnormal events. In this work, we implemented the five different YOLOv5 models and proposed a YOLOv5 based model with dilated convolution to detect anomaly events in the metro which achieved 97% mAP50. This model is very fast and easy to use not only for study purposes but also in real time.

후속연구

In future, we plan on testing and implementing the system in real time scenarios using the CCTV surveillance cameras.
The following chapters describes the dilated convolution and anomaly detection model architecture. We further elaborate the implementation of the mobile application for the anomaly detection task and results from the experiment, and conclude with what we hope to accomplish in the future.

참고문헌 (17)

M Paul, M. E. Shah and S. Chakraborty,？"Human detection in surveillance videos and？its applications-a review", EURASIP Journal？on Advances in Signal Processing, pp. 176,？2013.
Lo, Win-Tsung, Yue Shan Chang, Ruey-Kai？Sheu, Chao-Tung Yang, Tong-Ying Juang, and？Yu-Sheng Wu. "Implementation and evaluation？of large-scale video surveillance system based？on P2P architecture and cloud computing."？International Journal of Distributed Sensor？Networks 10, no. 4, 2014
Litjens, G., et al.: Deep learning as a tool for？increased accuracy and efficiency of？histopathological diagnosis. Sci. Reports 6？(2016)

상세보기
Thomas Defard, Aleksandr Setkov, Angelique？Loesch, Romaric Audigier, "PaDiM: a Patch？Distribution Modeling Framework for Anomaly？Detection and Localization", Proceedings of the？IEEE Conference on Computer Vision and？Pattern Recognition Workshops. 2020.
F. Mercaldo, F. Martinelli, and A. Santone, "A？proposal to ensure social distancing with deep？learning-based object detection," in 2021？International Joint Conference on Neural？Networks (IJCNN). IEEE, 2021, pp. 1-5.
Krizhevsky A., Sutskever I., Hinton G.E.,？Imagenet classification with deep convolutional？neural networks, Advances in Neural Information？Processing Systems (2012), pp. 1097-1105.
Zhang Q., Yang L.T., Chen Z., Li P. , A？surv ey on deep learning for big data, Inf.？Fusion, 42 (2018), pp. 146-157.
Yuan J., Hou X., Xiao Y., Cao D., Guan W.,？Nie L., Multi-criteria active deep learning for？image classification, Knowl.-Based Syst., 172？(2019), pp. 86-94
Joseph Redmon, Santosh Divvala, Ross？Girshick, Ali Farhadi, "You Only Look Once: Unified, Real-Time Object Detection., published？at Computer Vision and Patter Recognition？(CVPR)., 2015.
Fisher Yu, Vladlen Koltun. "Multi-Scale Context？Aggregation by Dilated Convolutions" Computer？Vision and Pattern Recognition; 2016.
Yulia Gavrilova, 'Convolutional Neural Networks？for Beginners', Serokell, August 3rd, 2021,？https://serokell.io/blog/introduction-to-convolutional-neural-networks
'Dilated Convolution', Geeksforgeeks, 02 Mar, 2022,？https://www.geeksforgeeks.org/dilated-convolution/
Benyamin Ghojogh, Maria N. Samad, Sayema？Asif Mashhadi, Tania Kapoor, Wahab Ali,？Fakhri Karray, Mark Crowley, "Feature？Selection and Feature Extraction in Pattern？Analysis: A Literature Review", Proceedings of？the IEEE Conference on Computer Vision and？Pattern Recognition Workshops. 2019.
Alexy Bochkovskiy, Chien-Yao Wang, Hong-Yuan？Mark Liao; "YOLOv 4: Optimal Speed and？Accuracy of object Detection". Computer？Vision and Pattern Recognition(CVPR); 23？April 2020.
Liu S.,. Qi L., Qin H., Shi J., Jia J. Path？aggregation network for instance segmentation;？Proceedings of the IEEE Conference on？Computer Vision and Pattern Recognition？(CVPR); Salt Lake City, UT, USA. 18-23？June 2018; pp. 8759-8768.
Shiv Ram Dubey, Satish Kumar Singh, Bidyut？Baran Chaudhuri, "Activation Functions in？Deep Learning: A Comprehensive Survey and？Benchmark", Proceedings of the IEEE？Conference on Computer Vision and Pattern？Recognition Workshops. 2021.
Zhilu Zhang, Mert R. Sabuncu, "Generalized？Cross Entropy Loss for Training Deep Neural？Networks with Noisy Labels", Proceedings of？the IEEE Conference on Computer Vision and？Pattern Recognition Workshops. 2018.

내보내기 구분	파일저장 인쇄 메일전송
구성항목	기본정보 상세정보 관리번호, 논문명, 저널/프로시딩명, 저자 , 발행년, 권, 호, 시작페이지, 끝페이지, 발행기관 관리번호, 논문명, 대등논문명, 저자 , 저널/프로시딩명, 발행기관, 발행년, 발행언어, 권, 호, 시작페이지, 끝페이지, ISBN, ISSN, 주제분야, 키워드, 초록(한글), 초록(영문), 저자(소속기관)
저장형식	Text(ASCII format) Excel format RefWorks Direct Export RIS format (for Reference Manager, ProCite, EndNote), Scholar's Aids, Mendeley
메일정보	받는사람 (필수) @ 보내는사람 (선택) @ 제목 내용 KISTI 검색결과 이메일 서비스
안내	총 건의 자료가 검색되었습니다. 다운받으실 자료의 인덱스를 입력하세요. (1-10,000) 검색결과의 순서대로 최대 10,000건 까지 다운로드가 가능합니다. 데이타가 많을 경우 속도가 느려질 수 있습니다.(최대 2~3분 소요) 다운로드 파일은 UTF-8 형태로 저장됩니다. 파일의 내용이 제대로 보이지 않을실 때는 웹브라우저 상단의 보기 -> 인코딩 -> 자동선택 여부를 확인하십시오. ~ Text(ASCII format) Excel format

연합인증