A method for automatic segmentation of tactile pavement in outdoor environments

Abstract

Aiming at the complex outdoor situation, the tactile pavement area in the captured images is prone to shadows and noise, which causes inaccurate tactile pavement segmentation. Based on the Y component of the YUV color space representing the luminance information, the shadow regions in the images are extracted and chromaticity correction is performed; a tactile pavement automatic segmentation algorithm in the outdoor environment is proposed by combining the maximum inter-class variance (OTSU) algorithm and the watershed segmentation algorithm. The OTSU algorithm is used to segment the pre-processed image, initially extract the tactile pavement region, combine the distance transform to label the image, and the watershed algorithm then uses the label for tactile pavement segmentation. The experiment proves that the method can avoid the influence of the shadow part in the image on the segmentation algorithm and improve the effectiveness of the segmentation of outdoor tactile pavement.