Traffic Sign Recognition in Disturbing Environments

Abstract

Abstract. Traffic sign recognition is a difficult task if we aim at detecting and recognizing signs in images captured from unfavorable environments. Complex background, weather, shadow, and other lighting-related problems may make it difficult to detect and recognize signs in the rural as well as the urban areas. We employ discrete cosine transform and singular value decomposition for extracting features that defy external disturbances, and compare different designs of detection and classification systems for the task. Experimental results show that our pilot systems offer satisfactory performance when tested with very challenging data.

Detection and recognition of traffic signs inside the attentional visual field of drivers

“…They used the RGB color space for the color segmentation stage and template matching for locating speed limit signs. A converted version of the RGB color space and a Laplacian of Gaussian (LoG) edge detector have been used by [59] for detection of triangular traffic signs.…”

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

“…In [6], Yang et al converted the original image to a new image using a pre-selected formula and defined the range for the red color, and then identified triangular borders by using a corner detection algorithm. In [7], the red, yellow and blue color were first distinguished by thresholding on H and S components in HSV color space, then the distance to border (DtB) feature was extracted in the ROIs (regions of interests) to classify the shape using SVM classifiers.…”

“…We also build a testing dataset (see Fig.11 (b) ) to evaluate the proposed method in detecting traffic signs by sampling images from [6][7] [9]. The resulted dataset consists of 500 images for the four traffic signs, and some of them include partial occlusions, different lightings, or shadows.…”

Driving Supportive System for Warning Traffic Sign Classification

“…Hsiu, Chao, Kun and Shang [4] proposed a system that used discrete cosine transform (DCT) and singular value decomposition (SVD) for extracting features. For the preprocessing of data, they used ROI detection, RGB color segmentation and LoG edge detector.…”