음향기반 교통사고 검지를 위한 딥러닝 모델 연구
A Study of Deep Learning Model for Traffic Accident Detection based on Acoustic Signals
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
C-ITS(Cooperative-Intelligent Transport Systems) 및 자율주행 환경에서 교통사고와 같은 돌발상황의 신속한 검지가 중요해짐에 따라, 최근에는 이를 위한 다양한 시스템 개발과 관련된 연구가 활발히 진행되고 있다. 본 연구는 기존의 영상 및 레이더 기반 검지 방식의 한계를 극복하기 위해 음향 정보를 이용한 교통사고 검지 기술을 개발하는 것을 목표로 한다. 구체적으로, 교통사고 시 발생하는 음향 데이터를 수집하고, 이를 통해 충돌 및 급제동과 같은 특징을 추출하였다. 추출된 음향 데이터를 바탕으로DNN, CNN, RNN의 딥러닝 모델을 학습시켜 각 모델을 통해 사고 패턴을 검지하는 방법을 연구하였다. 이후, 학습된 모델들의 검지 결과를 비교 분석하여 음향 패턴을 검지하는 데 가장 적합한 알고리즘 모델을 선정하였다. 최적의 알고리즘을 적용하여 음향기반 자동 사고 검지 시스템을 개발하였으며, 이 시스템은 교통사고를 신속하고 정확하게 판별할 수 있다. 본 연구는 딥러닝 알고리즘과 음향 특징 추출 기술을 통해 정확한 교통사고 검지를 가능하게 하여 교통 안전 향상에 기여할 수 있을 것으로 기대된다.
Abstract
In the context of C-ITS and autonomous driving, rapid detection of traffic incidents is crucial. This research addresses the limitations of traditional methods such as imaging and radar by developing a technology that utilizes acoustic information for incident detection. Specifically, we analyzed acoustic data to extract features and trained deep learning models—DNN, CNN, and RNN—to identify patterns associated with crash and skid. Through comparative analysis of the models' detection results, we identified the optimal algorithm for acoustic pattern recognition. This study demonstrates the potential of advanced deep learning algorithms and acoustic feature extraction in enhancing automated accident detection systems.
참고문헌 (26)
- Y-S. Hong, "Prevention system for real-time traffic accident", Journal of the Korea Society of Computer and Information, 11(4), pp.47~54 (2006)
- Mahmoud N. E., "A machine learning framework for automated accident detection based on multimodal sensors in cars", Sensors, 22(10), pp.3634~ (2022)
- Development of an automatic traffic accident detection system in tunnels based on acoustic signal processing, "Y-D. Shin", Ministry of Land, Infrastructure and Transport, ANJ Solutions, ~ (2019)
- F. Rosenblatt, "The perceptron: A probabilistic model for information storage and organization in the brain", Psychological Review, 65(6), pp.386~408 (1958), Cornell Aeronautical Laboratory
- LeCun Y., "Gradient-based learning applied to document recognition", Proceedings of the IEEE, 86(11), pp.2278~2324 (1998)
- Feng, W., "Audio visual speech recognition with multimodal recurrent neural networks", 2017 International Joint Conference on neural networks (IJCNN), pp.681~688 (2017)
- Ghahremannezhad, H., "Realtime accident detection in traffic surveillance using deep learning", 2022 IEEE international conference on imaging systems and techniques (IST), pp.1~6 (2022)
- Robles-Serrano, S., "Automatic detection of traffic accidents from video using deep learning techniques", International Journal of Advanced Computer Science, 23(2), pp.230~240 (2021)
- Garcia, K., "Robust traffic and intersection monitoring using millimeter wave sensors", Texas Instruments, pp.1~13 (2018)
- Choi, K., "A comparison of audio signal preprocessing methods for deep neural networks on music tagging", 201826th European Signal Processing Conference (EUSIPCO), pp.1870~1874 (2018)
- Gimeno, P., "Multiclass audio segmentation based on recurrent neural networks for broadcast domain data", EURASIP Journal on Audio, Speech, and Music Processing, pp.1~19 (2020)
- Zaman, K., "A survey of audio classification using deep learning", IEEE Access, 11, pp.106620~106649 (2023)
- Allen, J. B., "Efficient spectrum sensing method using the short time fourier transform algorithm", IEEE Signal Processing Magazine, 35(2), pp.77~85 (2018)
- Krizhevsky, A., "ImageNet classification with deep convolutional neural networks", Advances in Neural Information Processing Systems, 25, pp.1097~1105 (2012)
- Rumelhart, D. E., "Learning representations by back-propagating errors", Nature, 323(6088), pp.533~536 (1986)
- Srivastava, N., "Dropout : a simple way to prevent neural networks from overfitting", The Journal of Machine Learning Research, 15(1), pp.1929~1958 (2014)
- D. E. Rumelhart, "Learning representations by back-propagating errors", Nature, 323(6088), pp.533~536 (1986)
- Paszke, A., "Pytorch: An imperative style, high-performance deep learning library", Advances in Neural Information Processing Systems, 32 (2019)
- Jang, J., "Instantaneous incident detection system based on analysis of acoustic signal from crash and skid in tunnel", The Open Transportation Journal, 12(1) (2018)
- Sammarco, M., "Crashzam:Sound-based car crash detection", Proceedings of the 4th International Conference on Vehicle Technology and Intelligent Transport Systems (VEHITS 2018), pp.27~35 (2018)
- Foggia, P., "Audio surveillance of roads : A system for detecting anomalous sounds", IEEE Transactions on Intelligent Transportation Systems, 16(6), pp.1~12 (2015)