This study proposes a new vehicle type recognition method that combines global and local features via a two-stage classification.\nTo extract the continuous and complete global feature, an improved Canny edge detection algorithm with smooth filtering and\nnon-maxima suppression abilities is proposed. To extract the local feature fromfour partitioned key patches, a set of Gabor wavelet\nkernels with five scales and eight orientations is introduced. Different from the single-stage classification, where all features are\nincorporated into one classifier simultaneously, the proposed two-stage classification strategy leverages two types of features and\nclassifiers. In the first stage, the preliminary recognition of large vehicle or small vehicle is conducted based on the global feature\nvia a
Loading....