In my diploma thesis Usage of GPU acceleration in traffic sing recognition, I started to focus on the problems of detection and identification of traffic signs and how this process can be accelerated using a GPU. I designed an algorithm that detects traffic signs in RGB data using a Viola-Jones algorithm. The detected signs are subsequently identified by a trained SVM algorithm using histogram of oriented gradients. The detection and identification process was resolved over the same data through the CPU and GPU versions. This design resulted in an experimental tool called RsId, which implements the process and allows visualization of the differences between CPU and GPU versions. The multi-platform implementation was done using OpenCV library and C++ language.