A novel traffic sign detection method via color segmentation and robust shape matching

The robust and accurate detection of traffic signs is a challenging problem due to the many issues that are often encountered in real traffic video capturing such as the various weather conditions, shadows and partial occlusion. To address such adverse factors, in this paper, we propose a new traffic sign detection method by integrating color invariants based image segmentation and pyramid histogram of oriented gradients (PHOG) features based shape matching. Given the target image, we first extract its color invariants in Gaussian color model, and then segment the image into different regions to get the candidate regions of interests (ROIs) by clustering on the color invariants. Next, PHOG is adopted to represent the shape features of ROIs and support vector machine is used to identify the traffic signs. The traditional PHOG is sensitive to the cluttered background of traffic sign when extracting the object contour. To boost the discriminative power of PHOG, we propose introducing Chromatic-edge to enhance object contour while suppress the noises. Extensive experiments demonstrate that our method can robustly detect traffic signs under varying weather, shadow, occlusion and complex background conditions.