Moving Car Detection and Model Recognition based on Deep Learning

Abstract

Vehicle analysis is an import task in many intelligent applications, such as automatic toll collection and driver assistance systems. Among these applications, moving car detection and model recognition are a challenging task due to the close appearance between car models. In this paper, we propose a framework to detect moving cars and its model based on deep learning. We first detect the moving car using frame difference; the resultant binary image is used to detect the frontal view of a car by a symmetry filter. The detected frontal view is used to identify a car based on deep learning with three layers of restricted Boltzmann machines (RBMs). Experiment results show that our proposed framework achieves favorable recognition accuracy.

Original document

The different versions of the original document can be found in:

• http://dx.doi.org/10.14257/astl.2015.90.13

• http://pdfs.semanticscholar.org/66fb/b53bc1996db6e11358550530fec78b736008.pdf

• http://onlinepresent.org/proceedings/vol90_2015/13.pdf,

https://academic.microsoft.com/#/detail/2329814265