Scale and skew‐invariant road sign recognition

Abstract

A fast and robust method to detect and recognize scaled and skewed road signs is proposed in this paper. In the detection stage, the input color image is first quantized in HSV color model. Border tracing those regions with the same colors as road signs is adopted to find the regions of interest (ROI). The ROIs are then automatically adjusted to fit road sign shape models so as to facilitate detection verification even for scaled and skewed road signs in complicated scenes. Moreover, the ROI adjustment and verification are both performed only on border pixels; thus, the proposed road sign detector is fast. In the recognition stage, the detected road sign is normalized first. Histogram matching based on polar mesh is then adopted to measure the similarity between the scene and model road signs to accomplish recognition. Since histogram matching is fast and has high tolerance to distortion and deformation while contextual information can still be incorporated into it in a natural and elegant way, our method has high recognition accuracy and fast execution speed. Experiment results show that the detection rate and recognition accuracy of our method can achieve 94.2% and 91.7%, respectively. On an average, it takes only 4-50 and 10 ms for detection and recognition, respectively. Thus, the proposed method is effective, yet efficient.

Using self-organising maps in the detection and recognition of road signs

“…Candidate objects were passed to a multi-layer perceptron (MLP) (operating in YCbCr) trained on RS and non-RS images. Colour quantisation in the HSV colour model was used by Liu et al [15] to find ROIs, followed by border tracing and ROI scaling.…”

Road Sign Recognition with Fuzzy Adaptive Pre-Processing Models

“…Finally, the density and the edge orientation of these 49 areas can be as templates, with a matching scheme used to recognize the road signs. Liu [ 10 ] employed color equalization in a HSV color space to separate the color information in road signs. The area and aspect ratio were used to detect the Regions of Interest (ROI), which were then normalized to obtain 32 distributed radial areas.…”

Fuzzy adaptive pre-processing models for road sign recognition

“…Gao [8] proposed a system in which the input images first had to be transformed into the CIECAM97 color space so that regions of interest could be extracted. Liu [9] employed color equa lization in a HSV color space to separate the color informa tion in road signs. The area and aspect ratio were used to detect the region of interested (ROI), which was then nor malized to obtain 32 distributed radial areas.…”