IMAGE PROCESSING, PATTERN RECOGNITION

Preprocessing of digital images in systems of location and recognition of road signs

P. Yu. Yakimov

Image Processing Systems Institute of the RAS

Abstract: The problem of localization and recognition of road signs is actual for today. Such a system can not only improve safety, compensating the probable human inattention, but it also helps to reduce tiredness, helping drivers keep an eye on the surrounding traffic conditions. This article proposes an efficient algorithm for preprocessing digital images for further detection of road signs in real time. The article considers the possibility of using HSV color space to extract the red. A denoising algorithm was developed to improve the accuracy and speed of detection. Parallel implementation on the GPU was used to remove the noise. The resulting images are best suited for further localization of road signs.

Keywords: HSV color space, image denoising, traffic signs detection, traffic signs recognition, CUDA.

Received: 17.06.2013