Vision‐based approach for predicting the probability of vehicle–pedestrian collisions at intersections

Abstract

Road accidents impose serious problems on society. Possible collisions between vehicles and pedestrians must be detected before they occur so that a timely warning may be issued. By using the vision‐based approach, this study presents an effective and efficient algorithm to estimate the vehicle–pedestrian collision probability at intersections. The real‐time trajectories and movement parameters (position, speed, acceleration or direction) of vehicles and pedestrians are obtained based on state‐of‐the‐art detection and tracking algorithm which include background subtraction method, faster regions with convolutional neural networks and optical flow method. To find the appropriate time to identify the latent collision risk for calculating the collision probability, this study defines the critical time based on different collision patterns of perception‐reaction failure and evasive action failure. In addition, based on discrete acceleration and discrete angle, the authors get different extended trajectories which can include most situation when the conflict happened. Trajectories generation probability are given by the discrete choice probability model based on the Logit model to get the accurate collision probability. Real‐world video data is implemented to demonstrate the approach. This proposed collision prediction method can provide some important results for designing the intelligent pedestrian signal timing schemes at intersections.

A novel method of predictive collision risk area estimation for proactive pedestrian accident prevention system in urban surveillance infrastructure

“…The use of vision sensors is supposed to make it easier to recognize vehiclepedestrian collisions in advance and warn immediately them of such situations, and to evaluate vehiclepedestrian behavioral interactions that pose a threat to pedestrian at crosswalk [7]- [10]. To date, an extensive variety of studies have reported on deriving a surrogate safety measurement (SSM) [11], [12], estimating collision probability [13], measuring levels of potential risks, analyzing road users' behaviors [14], [15] in order to prevent vehicle-pedestrian collisions proactively.…”

“…They used vehicle trajectories as indicators of safety performance, and TTC was applied as conflict risk indicator with TTC* threshold to decide risk related with the perception, reaction time and driving conditions. The authors in [13] estimated vehiclepedestrian collision probability at intersections. They defined the critical time depending on collision patterns of perception-reaction failure and evasive action failure to identify the latent collision risk for computing collision probability.…”

Robust Target Detection and Tracking Algorithm Based on Roadside Radar and Camera

“…Equations (24)-(31) are the recursive equations of the detection part. Note that in the weight update equation (27), the detection probability P D,k is removed because P D,k = 1 in this case. The classification of Gaussian components remains unchanged, and the loss tag L loss = 0.…”

“…Shuai Hua et al built a multi-vehicle tracking framework based on the Yolo network that can be used for real-time traffic applications [ 25 ]. Some scholars have also tried to use advanced image-processing methods to estimate the vehicle–pedestrian collision probability [ 26 ] or detect abnormal events [ 27 ] at intersections. The new generation of lidar has a 360-degree scanning field of view and more scanning beams, such as 32 lines, 64 lines, and 128 lines, which can provide higher detection accuracy.…”

Effectiveness of the Autonomous Braking and Evasive Steering System OPREVU-AES in Real World Vehicle-to-pedestrian collisions

“…These systems are the subject of ongoing research aimed at improving both pedestrian behavior identification algorithms and safe avoidance maneuvers [1][2]. Some authors [3] point out possible areas of optimization of AEB systems.…”