Study on Vehicle Accident Prediction Based Video
Peng Yinziang, Zhiyong Zhu · 2007
This article reports on a study that investigated the use of digital image processing (video) to predict traffic accidents. The authors describe the use of the hidden Markov model to predict traffic collisions in the setting of expressways. The model first obtains movement information by tracking the vehicle, then the hidden Markov model is applied to judge the probability of a collision, based on the mutual movement information abstracted. The authors' simulation demonstrates that this approach is valid and effective in predicting collisions.