On road accidents is a key issue for road safety and is a major source of loss of lives both of the drivers and commuters. Even with all modern developments in the field of vehicle design, road lane design and management, accidents are still more common. Timely accident detection and the concurrent action in providing emergency health care to the victims are necessary to ensure road safety. This may be ensured by informing an emergency healthcare and administration units like hospitals and police stations about the accident on priority. Road traffic accidents contribute to large number of deaths world-wide every year. In these cases it is necessary to inform the family members in time and seek healthcare assistance at the earliest. Unfortunately, in many cases either the family members are communicated very late or the emergency services reach the spot at a delayed time. The delay in attending the accident victims, informing the concerned authorities and family and in mobilization of ambulance services for relocating the victims for immediate healthcare lead to casualty and many a times to fatal injuries. This paper proposes real-time accident detection and alerting system that uses smartphones. Every smartphone has number of sensors embedded in its design. Our system makes use of few of these commonly available sensors across all smartphones to build a web application for remote monitoring. The system will provide faster response time in locating and mobilizing the emergency services for the victims. When the system detects an accident it alerts the nearest emergency station like police administration, healthcare service and ambulance operators of the same. It also provides real time tracking for these emergency service providers.
Localizing moments in the videos has been a new challenging task in the field of Computer Science to provide faster search time for video retrieval, query processing and also behavioral analysis. The process involves stages such as video understanding, video segmentation, query processing using NLP and generation of localization of the concepts in the video. Though there have been many attempts to Video understanding in the field of NLP and Computer Vision in past years, they lack to cover the large untrimmed videos in current real-life scenarios. We propose the deep learning-based solution with the use of Random Forest and Bi-LSTM approach to localize the labels in segments and also the time at which they pertain to the particular segments. We used the YouTube 8M dataset provided by YouTube in Kaggle's challenge to train our frame-based model and use it to classify the segments using sliding windows of size 5. Our approach tries to provide a naïve and robust approach to model this concept and provide a way to tackle this large problem. Further improvements in the Bi-LSTM based models and Random Forest models with VLAD would lead to better results.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.