中文摘要：

提出了一种基于自适应Unit—Linking脉冲耦合神经网络（ULPCNN）赋时矩阵的图像特征识别算法。该方法在充分考虑图像局部信息的基础上．对ULPCNN闻值函数及链接强度做了修正改进，形成自适应连接的AULPCNN，利用AULPCNN模型对原始图像进行处理．生成一种从空间图像信息到时间信息的赋时矩阵映射图，并将其视为一幅图像进行处理。然后利用物理学相关概念定义赋时矩阵重心不变特征．且将这一特征运用在图像特征提取与目标识别中。理论分析和实验结果表明：AULPCNN赋时矩阵重心特征具有良好的抗几何畸变性（TRS）、抗亮度畸变性和抗噪声干扰不变性，具有提取特征参数少、提取方法简单、易于实现、识别正确率较高和稳健性强等特点。

英文摘要：

A novel image feature recognition algorithm using adaptive Unit-Linking Pulse Coupled Neural Networks （AULPCNN） is put forward. Firstly, unit-linking PCNN threshold function and linking strength are improved based on sufficient consideration of image local information, and then the adaptive linking AULPCNN is formed. Secondly, the time matrix is generated by PCNN, which is a mapping from the spatial image information to time information when original images is processed using AULPCNN, and can be regarded as an image. Finally, according to some physical concepts, a new invariable center feature of the time matrix is defined and used in image target recognition. Theoretical analysis and experimental simulations show that center feature of AULPCNN time matrix have the ability of anti-geometric distortions （translation, rotation and scaling, TRS）, anti-brightness distortions and anti-noise disturbance, the novel method have characteristics of simple extraction approach, little extraction parameter, easy implementation, higher accurate recognition ratio and strong robustness.