Car detection

A car detection system was developed using Haar like features, Edge Orientation Histograms (EOH) and Support Vector Machine (SVM) with Histogram of Oriented Gradients (HOG).   

An additional method that I call "Pipelets" was developed.   Using pipelets, car detection results can be improved

Thousands of weak classifiers are constructed around one image of 30 x 30 pixels and just the best of them are selected using Real Adaboost.  


Figure 1.  Results obtained after this method was applied.


This site was last updated 07/06/14