오토인코더 기반 이상 행동 탐지 시스템에서 3D CNN과 초해상화 기법의 적용 및 성능 분석
Application and performance analysis of 3D CNN and super-resolution in autoencoder-based anomaly detection system
2
조회
0
다운로드
2
인용
0
참고문헌
키워드
초록
비디오에서 이상 탐지(Anomaly detection)란 영상 프레임에서 보이는 비정상(Abnormal)적인 행위 또는 상황을 인식하는 것을 의미하며, 사고를 예방하거나 대응하는 기술로 발전이 가능하다. 본 논문에서는 연속되는 프레임의 시공간적인 특징을 동시에 학습하는 3D CNN과 자동으로 특징을 압축하고 복원하는 오토인코더(Autoencoder)를 활용하여 비디오에서 이상 상황을 탐지하는 시스템을 제안한다. 또한 저해상도(Low Resolution) 이미지를 고해상도 이미지(High Resolution) 이미지로 변환시키는 영상 초해상화(Image Super Resolution, ISR) 기법을 적용하여 모델의 성능이 데이터 전처리의 과정에 따라 어떻게 변화하는지를 살펴보았다. 본 논문은 UCSD 데이터셋을 사용하여 시각적 이상 탐지 성능 비교 실험을 수행하였으며 실험 결과는 고해
Abstract
Anomaly detection in video is recognizing an abnormal behavior or situation in frames, and it can be developed as a technology to prevent or respond to accidents. In this paper, we propose a system for detecting anomalies in video using a 3D CNN that learns spatio-temporal features of normal frames and an autoencoder that automatically compresses and restores features. Also, by applying the Image Super Resolution (ISR) technique, which converts a low-resolution image into a high-resolution image, we examined how the performance of the model changes according to the data preprocessing process. The performance of the proposed model was measured using the UCSD dataset and the experimental results show that the 3D CNN-based autoencoder using high-resolution images is superior to other techniques.
참고문헌 (25)
- Y. Sun, "Automatically Designing CNN Architectures Using the Genetic Algorithm for Image Classification", EEE Transactions on Cybernetics, 50(9), pp.3840~3854 (2020)
- Y. Sun, "Automatically Designing CNN Architectures Using the Genetic Algorithm for Image Classification", EEE Transactions on Cybernetics, 50(9), pp.3840~3854 (2020)
- 파우델 사하데브, "Medical Image Segmentation via Multi-scale Attention Guided Network", 한국차세대컴퓨팅학회 논문지, 16(3), pp.51~62 (2020), 한국차세대컴퓨팅학회[DOI]
- K. Rao, "Exploring architectures, data and units for streaming endto-end speech recognition with RNN-transducer", 2017 IEEE Automatic Speech Recognition and Understanding Workshop(ASRU), pp.193~199 (2017)
- D. Khattar, "Mvae: Multimodal variational autoencoder for fake news detection", The world wide web conference (2019)
- Yu, Jianbo, "A deep autoencoder feature learning method for process pattern recognition", Journal of Process Control, pp.1~15 (2019)
- 강재용, "객체 중심 증강 기법을 통한 군중 장면에서의 임베딩 유사도 기반 이상 탐지", 한국차세대컴퓨팅학회 논문지, 17(3), pp.7~17 (2021), 한국차세대컴퓨팅학회[DOI]
- Ranjai Baidya, "Tracking The Trajectory of a Golf Ball Based on Hough Transform and YOLO", 한국차세대컴퓨팅학회 논문지, 17(2), pp.42~52 (2021), 한국차세대컴퓨팅학회[DOI]
- Krizhevsky, Alex, "Imagenet classification with deep convolutional neural networks", Advances in neural information processing systems, 25 (2012)
- Varol, Gül, "Long-term temporal convolutions for action recognition", IEEE transactions on pattern analysis and machine intelligence, 40(6), pp.1510~1517 (2017)
- C. Dong, "IEEE Transactions on Pattern Analysis and Machine Intelligence", IEEE Transactions on Pattern Analysis and Machine Intelligence, 38, pp.295~307 (2016)
- Kim, Jiwon, "Accurate Image Super-Resolution Using Very Deep Convolutional Networks", 2016 IEEE Conference on Computer Vision and Pattern Recognition(CVPR), pp.1646~1654 (2016)
- G. Huang, "Densely Connected Convolutional Networks", 2017 IEEE Conference on Computer Vision and Pattern Recognition(CVPR) : 2261-2269, pp.2261~2269 (2017)
- Y. Zhang, "Residual Dense Network for Image Super Resolution", Proc. of IEEE Conf., CVPR (Conference on Computer Vision and Pattern Recognition), pp.2472~2481 (2018)
- Li, Qilei, "Gated Multiple Feedback Network for Image Super-Resolution", BMVC (2019)
- Gondara, Lovedeep, "Medical image denoising using convolutional denoising autoencoders", 2016 IEEE 16th international conference on data mining workshops(ICDMW) (2016)
- S. Dubey, "Anomalous Event Recognition in Videos Based on Joint Learning of Motion and Appearance with Multiple Ranking Measures", Applied Sciences, 11(3) (2021)
- Jaeyong Kang, "Anomaly Detection of the Brake Operating Unit on Metro Vehicles Using a One-Class LSTM Autoencoder", Applied Sciences, 11(19) (2021)
- Lim, Hyun-seok, "Anomaly detection performance improvement technique through weight matrix-based optical flow equalization", Proceedings of the Korean Society of Computer Information Conference (2021), Korean Society of Computer Information
- Jongmin Yu, "Abnormal event detection using adversarial predictive coding for motion and appearance", Information Sciences, 586, pp.59~73 (2022)
- P. Nagrath, "Real-Time Anomaly Detection Surveillance System", Proceedings of Data Analytics and Management, pp.665~678 (2022), Springer
- Sabih, M., "Crowd anomaly detection with LSTMs using optical features and domain knowledge for improved inferring", Vis Comput, 38, pp.1719~1730 (2022)[DOI]
- Mahadevan, V., "Anomaly detection in crowded scenes", Computer Vision and Pattern Recognition (CVPR), pp.1975~1981 (2010)