基于RBPNN的退化交通标志图像的识别算法  

Identification of degraded traffic sign symbols using radial basis probabilistic neural networks

作　　者：李伦波[1] 马广富[1] 

机构地区：[1]哈尔滨工业大学控制科学与工程系,哈尔滨150001

出　　处：《吉林大学学报（工学版）》2008年第6期1429-1433,共5页Journal of Jilin University:Engineering and Technology Edition

基　　金：高等学校博士学科点专项科研基金项目(20050213010);国家自然科学基金项目(60674101)

摘　　要：为了识别退化的交通标志图像,提出了一种新的特征提取算法。该算法在处理图像退化问题时,采用模糊-仿射联合不变矩直接提取图像的特征,从而避免了需要较大计算量的图像复原处理过程。在利用模糊-仿射联合不变矩作为图像特征的基础上,采用递归正交最小二乘算法优化设计径向基概率神经网络分类器。仿真结果表明:模糊-仿射联合不变矩是一种有效的退化交通标志图像的特征提取算法,所设计的径向基概率神经网络分类器不仅具有精简的结构而且具有较好的推广性能。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.

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 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]