No abstract

Traffic sign recognition usually consists of two parts : detection and classification. In this paper we describe the classification stage using ring partitioned method. In this method, first the RGB image is converted into gray scale image using color thresholding and histogram specification technique. This gray scale image, called as specified gray scale image is invariant to the illumination changes. Then the image is classified using ring partitioned method. The image is divided by several concentric areas like rings. In every ring the histogram is used as an image descriptor. The matching process is done by computing the histogram distances for all rings of the images by introducing the weights for every ring. The method doesn't need a lot of samples of sign images for training process, alternatively only the standard sign images are used as the reference images. The experimental results show the effectiveness of the method in the matching of occluded, rotated, and illumination problems of traffic sign images.

Aryuanto SOETEDJO•õa), Nonmember and Koichi YAMADA•õ•õb), Member SUMMARY This paper describes a new color segmentation based on a normalized RGB chromaticity diagram for face detection. Face skin is extracted from color images using a coarse skin region with fixed boundaries followed by a fine skin region with variable boundaries. Two newly developed histograms that have prominent peaks of skin color and nonskin colors are employed to adjust the boundaries of the skin region. The proposed approach does not need a skin color model, which depends on a specific camera parameter and is usually limited to a particular environment condition, and no sample images are required. The experimental results using color face images of various races under varying lighting conditions and complex backgrounds, obtained from four different resources on the Internet, show a high detection rate of 87%. The results of the detection rate and computation time are comparable to the well known real-time face detection method proposed by Viola-Jones [11], [12].

No abstract

No abstract

Abstract

We propose an efficient algorithm for detecting traffic signs in images.Geometric fragmentationdetects circular red traffic signs in an image by finding and combining the left and right fragments of elliptical objects to increase the accuracy of detection and cope with occlusion. The search for fragments resembles a genetic algorithm (GA) in that it uses the termsindividual,population,crossover, andobjective functionused in the GA. It is different in that it conducts a concurrent random search in a small two-dimensional space devised heuristically. The objective function for evaluating individuals is devised to increase detection accuracy and reduce computation time. The algorithm was tested for detecting circular red traffic signs both from artificial sign images and real scene images. Experimental results demonstrated that the proposed algorithm has higher detection, fewer false alarms, and lower computation cost than GA-based ellipse detection. Compared to conventional template matching, the proposed algorithm performs better in detection and execution time and does not require a large number of carefully prepared templates.

No abstract