Bangladeshi road sign detection based on YCbCr color model and DtBs vector

Detection of road sign from a road image is very crucial for intelligent transportation system for awareness of drivers and blind pedestrians. In this regard, a framework has been proposed in this paper for detection of the road sign from road images. For detection of the road sign, two natural properties of a Bangladeshi road sign is utilized, they are - border color rim of the road sign and the shape of the road sign. Based on these ideas initially, YCbCr color model is used to eliminate the illumination sensitiveness for segmentation. In the second step, statistical threshold value is used for color segmentation. Next, labeling and filtering is used to extract the shapes. Finally, Distance to Borders (DtBs) vector is used to verify the region of interest (ROI) for detecting the Bangladeshi road sign (BRS). Various road images are used with a variety of conditions to test the proposed framework and results are presented to prove its efficiency.