To improve speed and accuracy of vehicle detection

a vehicle detection algorithm based on two types of features

which are multi-scale histogram of oriented gradient（ HOG） feature and multi-scale multi-block local binary pattern（ MB-LBP） feature

and nested cascade Gentle Adaboost is proposed. The algorithm uses integral histogram and integral image to speed up the feature extraction process of multi-scale HOG and multi-scale MBLBP feature

respectively.Two types of weak classifiers are built for Gentle Adaboost based on two kinds of feature

and nested cascade Gentle Adaboost classifier is used to increase detection rate and detection speed. Simulation experimental results show that the proposed vehicle detection algorithm has faster detection speed and higher detection precision and recall-rate

compared with several existing vehicle detection algorithms.