摘要：

针对冗余边缘对基于边缘统计特征的车牌定位算法存在较严重干扰的问题，提出一种基于脉冲耦合神经网络(PCNN)的车牌定位方法。在借鉴传统算法的基础上，为抑制干扰性边缘，引入简化的PCNN模型，仅对候选区进行数次PCNN迭代运算，可大幅降低运算复杂度并提高车牌定位率。对300幅车辆图像进行仿真实验，取得了98.3%的定位率。

关键词: 边缘, 车牌定位, 脉冲耦合神经网络

Abstract:

To solve serious interference problems of redundant edges for edge-based statistical features of license plate location, a new vehicle license plate location method based on Pulsed Coupled Neural Network(PCNN) is proposed, which introduces the simplified PCNN model based on traditional license plate location algorithms. In the algorithm only the candidate area is produced by PCNN iteration in order to reduce the computational complexity and improve the rate of license plate location. Simulation experiments with 300 frames of vehicle license plate image that are taken under various kinds of conditions can get a 98.3% extraction rate.

Key words: edge, vehicle license plate location, Pulsed Coupled Neural Network(PCNN)

中图分类号: 

• TP391