Real-Time Classification of Traffic Signs

✍ Scribed by Phil Douville

• Book ID: 102977141
• Publisher: Elsevier Science
• Year: 2000
• Tongue: English
• Weight: 568 KB
• Volume: 6
• Category: Article
• ISSN: 1077-2014
• DOI: 10.1006/rtim.1998.0142

✦ Synopsis

Real-Time Classi®cation of Trac Signs

A challenging real-time imaging problem is classifying video trac signs in background clutter under rotation, scale, and translation invariant conditions. Normalized Gabor Wavelet Transform features from multi-resolution ®lters were originally biologicallybased; however, optimized features proved more eective. Two whole image template matching techniques were unsuccessful. A statistical pattern recognition system recognized approximately 30% of the images for the original features and 50% for the optimized features; however, a multilayer perceptron (mlp) detected over 70% of the images with the optimized features. The research demonstrated the possibility of a future automotive navigation aid which robustly collects sign images and classi®es these images in real-time with a single Fast Fourier Transform (FFT), a bank of ®lters and a trained neural net.