A Warning System for Obstacle Detection at Vehicle Lateral Blind Spot Area

Abstract

For a real driver assistance system, the weather, driving speed, and background could affect the accuracy of obstacle detection. In the past, only a few studies covered all the different weather conditions and almost none of them had paid attention to the safety at vehicle lateral blind spot area. So, this paper proposes a hybrid scheme for pedestrian and vehicle detection, and develop a warning system dedicated for lateral blind spot area under different weather conditions and driving speeds. More specifically, the HOG and SVM methods are used for pedestrian detection. The image subtraction, edge detection and tire detection are applied for vehicle detection. Experimental results also show that the proposed system can efficiently detect pedestrian and vehicle under several scenarios.

Vehicle detection in UAV traffic videos using GAN online augmentation: a transfer learning approach

“…The properties of IOU (Intersection over Union) dictate that it cannot only determine the positive and negative, but can also be used to calculate the overlap between the predicted and true frames. IOU and IoULoss are defined in equations (5,6).…”

Improving Vehicle Detection Accuracy Based on Vehicle Shadow and Superposition Elimination

“…Automated video surveillance has emerged as an important research topic in the computer vision community. There are many reports on vehicle detection and tracking in traffic video, purpose is to build an intelligent transportation system for solving the problems of road traffic congestion and traffic accidents in urban highway main road [2,3]. Intelligent transportation system has an important role in command and control system of traffic.…”

A Driver-Assistance System for Large Vehicles Based on Pedestrian and Vehicle Detection

“…A. Broggi et al used four laser scanners and a camera to detect pedestrians and vehicles, in order to assess dangerous situations in urban street scene [5]. J.-D. Lee et al proposed a new method to detect and track pedestrians depending on weather condition, such as day, night and rain day [6]. The accuracy of these methods is acceptable, but the detection speed is normally relatively low, so it cannot timely alarm the driver to avoid accidents…”