A Cascade of Boosted Generative and Discriminative Classifiers for Vehicle Detection

Abstract

We present an algorithm for the on-board vision vehicle detection problem using a cascade of boosted classifiers. Three families of features are compared: the rectangular filters (Haar-like features), the histograms of oriented gradient (HoG), and their combination (a concatenation of the two preceding features). A comparative study of the results of the generative (HoG features), discriminative (Haar-like features) detectors, and of their fusion is presented. These results show that the fusion combines the advantages of the other two detectors: generative classifiers eliminate "easily" negative examples in the early layers of the cascade, while in the later layers, the discriminative classifiers generate a fine decision boundary removing the negative examples near the vehicle model. The best algorithm achieves good performances on a test set containing some 500 vehicle images: the detection rate is about 94% and the false-alarm rate per image is 0.0003.

Motion Component Supported Boosted Classifier for Car Detection in Aerial Imagery

“…Also (She et al, 2004) are detecting cars by the use of Haar wavelets features in the HSV color space. A combination of Haar and HoG features which are formed to a strong cascading classifier by Boosting presents (Negri et al, 2008). In (Kasturi et al, 2009) a simple background subtraction is done which is only working for video data.…”

An Application for Stereo Vision based Vehicle/Obstacle Detection for Driver Assistance

“…Search space reduction is done first on the stereo input image to eliminate sky region retaining road region in which segmentation of vehicles/obstacles needs to be carried out. Figure 7,8,9(b) shows the stereo disparity image. Figure 7 …”

“…However, their performance may decisively rely on the created templates. Appearance based techniques uses features like Haar-like [7,8] and Gabor [9,10] cooperated with classifiers like SVM and Adaboost proved to yield a decent performance in the recent literatures. But the execution time of methods using databases tends too slow to be applied to an embedded system.…”

A Novel Approach for Vehicle Detection for Driver Assistance

“…In order to get these strong horizontal lines, a horizontal Haar-like feature mask [6,7] is used. Here the mask size of 9 * 9 is used for our experiments.…”

“…However, their performance may decisively rely on the created templates. Appearance based techniques uses features like Haar-like [6,7] and Gabor [8,9] cooperated with classifiers like SVM and Adaboost proved to yield a decent performance in the recent literatures. But the execution time of methods using databases tends too slow to be applied to an embedded system.…”