Vehicles detection using the MLP and the correlation measurement

Abstract

More and more deaths and damage in our highways, the numbers are appalling and the finding is real. In 2018, in Algeria more than 4,000 people have been mowed by road accidents, the half of them are young people. With these figures, it is no longer a question of just a simple classic road safety, but an automatic and accurate road safety is widely desired. Our work is just part of this problematic and offers at the first time and in terms of computer vision, a combination of different image processing approaches and artificial intelligence. Our method is designed to exactly detect vehicles in particular locations, it has been made based on a combination of a learning process and a correlation measurement approach. In our method, a secession of phases have been used which, features extraction types Haar, reducing the size of the vectors that were used for the training phase using the technical subspace, a learning procedure using an artificial neural network, and finally to refine the results a correlation measurement approach was performed.