关键词: 计算机应用, 模式识别, 径向基概率神经网络, 交通标志, 模糊仿射联合不变矩, 递归正交最小二乘法

Abstract:

A feature extraction algorithm was presented for the recognition of traffic sign symbols under degradations. In order to cope with the degradation, combined bluraffine invariants (CBAIs) were adopted to recognize the traffic sign symbols without any restorations based on many computations. A radial basis probabilistic neural network (RBPNN) classifier was optimized using recursive orthogonal least algorithm (ROLSA) and applied to the classification of the degraded traffic signs. The simulation results indicate that CBAIs are efficient for the degraded images feature extraction. And the designed network is not only parsimonious but also competitive in generalization.

Key words: computer application, pattern recognition, radial basis probabilistic neural networks (RBPNN), traffic sign, combined bluraffine invariants (CBAIs), recursive orthogonal least algorithm (ROLSA)