Vehicle detection using simplified fast R-CNN

This paper proposes a simplified fast region-based convolutional neural network (R-CNN) for vehicle detection. Fast R-CNN is a well-known method for object recognition using deep convolution networks. The original fast R-CNN consists of two separated parts: regional proposal and object recognition. The object recognition part in Fast R-CNN is redundant for our system which can be removed to speed up the training process. In the experiments, we test our method by using SHRP 2 NDS database [10] offered by Virginia Tech Transportation Institute (VTTI) to show the detection accuracy.