Dynamic obstacle identification based on global and local features for a driver assistance system

Abstract

This paper proposes a novel dynamic obstacle recognition system combining global feature with local feature to identify vehicles, pedestrians and unknown backgrounds for a driver assistance system. The proposed system consists of two main procedures: a dynamic obstacle detection model to localize an area containing a moving obstacle, and an obstacle identification model, which is a hybrid of global and local information, for recognizing an obstacle with and without occlusion. A dynamic saliency map is used for localizing an area containing a moving obstacle. For the global feature analysis, we propose a modified GIST using orientation features with MAX pooling, which is robust to translation and size variations of an object. Although the global features are a compact way to represent an object and provide a good accuracy for non-occluded objects, they are sensitive to image translation and occlusion. Thus, a local featurebased identification model is also proposed and combined with the global feature. As such, for the obstacle identification problem, the proposed system mainly follows the global feature-based object identification. If the global feature-based model identifies a candidate area as background, the system verifies the area again using the local feature-based model. As a result, the proposed system is able to provide information on both the appearance of obstacles and the class of an obstacle. Experimental results show that the proposed model can successfully detect obstacle candidates and robustly identify obstacles with and without occlusion.

Computer Vision Applications in Intelligent Transportation Systems: A Survey

“…In [ 227 ], the IS and SV methods were used together; in [ 228 , 229 ], the ANN and SV methods were used together; in [ 230 ], Haar-like features, IS, and principal component analysis with histograms of oriented gradients (PCA-HOG) were used together, while objects were classified with SVM. In [ 106 ], the authors used the SV and HOG methods together with the histograms of flow (HoF) technique; [ 107 ] employed the OF method of the forward–backward error algorithm; in [ 231 ], HOG was used together with cascade classifiers and Haar-like properties; [ 232 ] employed global and local features; finally, in [ 233 ], the HOG, hypothesis generation, and SVM methods were used together by the authors.…”