Improved traffic sign recognition for in-car cameras

We propose an enhanced approach for recognizing traffic signs; images are captured using a video camera installed in a car. Support vector machines detect the edges and gradients of each captured image to identify possible traffic sign regions in the image. In our system, bilateral Chinese transforms recognize circular traffic signs; vertex and bisector transforms detect triangular traffic signs. Our system applies color information for accurate identification of traffic sign regions in images. The traffic signs are recognized using a neural network. Results indicate that our system can obtain the desired results effectively.