%0 Journal Article
%@ 21903018
%A Watada, J.
%A Zhang, H.
%A Melo, H.
%A Sun, D.
%A Vasant, P.
%D 2018
%F scholars:10954
%I Springer Science and Business Media Deutschland GmbH
%J Smart Innovation, Systems and Technologies
%K Accidents; Highway accidents; Multimedia signal processing; Object detection; Pedestrian safety; Signal processing; Traffic signals; Vehicles, Advanced safety vehicles; Boosting algorithm; Histograms of oriented gradients (HoG); Hog feature; Important features; Pedestrian detection; Traffic situations; Vehicle detection, Feature extraction
%P 340-348
%R 10.1007/978-3-319-63856-0₄₂
%T Boosted HOG features and its application on object movement detection
%U https://khub.utp.edu.my/scholars/10954/
%V 81
%X Nowadays, traffic accidents is universally decreasing due to many advanced safety vehicle systems. To prevent the occurrence of a traffic accident, the first function that a safety vehicle system should accomplish is the detection of the objects in traffic situation. This paper presents a popular method called boosted HOG features to detect the pedestrians and vehicles in static images. We compared the differences and similarities of detecting pedestrians and vehicles, then we use boosted HOG features to get an satisfying result. In detecting pedestrians part, Histograms of Oriented Gradients (HOG) feature is applied as the basic feature due to its good performance in various kinds of background. On that basis, we create a new feature with boosting algorithm to obtain more accurate result. In detecting vehicles part, we use the shadow underneath vehicle as the feature, so we can utilize it to detect vehicles in daytime. The shadow is the important feature for vehicles in traffic scenes. The region under vehicle is usually darker than other objects or backgrounds and could be segmented by setting a threshold. © Springer International Publishing AG 2018.
%Z cited By 3; Conference of 13th International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2017 ; Conference Date: 12 August 2017 Through 15 August 2017; Conference Code:195379