中文摘要：

首先从Hopfield自联想记忆模型（HAM）出发，对其回忆规则运用机器学习中流行的核技巧，构建一个核自联想记忆模型框架（KAM）．并通过核函数的选取，使指数型相关联想记忆模型（ECAM）和改进的ECAM（IEC—AM）模型成为其中的两个特例．然后针对二维视觉图像的识别，在核函数中引入反映视觉特性的二维（2-D）距离因子，进一步提出一个距离加权的2-D核自联想记忆模型框架（DW2D—KAM）．由此较大改进KAM对图像的存储和纠错性能，并且使该模型更加符合神经生理学和解剖学的思想．最后，计算机模拟不仅证实DW2D—KAM比KAM在字符识别上具有更高的存储和纠错性能，而且其同样优于Seow和Asari提出的模块化HAM的识别效果．

英文摘要：

By using the kernel trick to modify Hopfield auto-associative memory model （HAM）, a framework of kernel auto-association memory model （KAM） is proposed. KAM makes exponential correlation associative memory （ ECAM ） and improved ECAM （ IECAM ） become two special cases. Then, the framework of distance weighted 2-D kernel auto-association memory model （DW2D-KAM） is constructed by introducing distance factors tO the kernels. DW2D-KAM improves the Storage capacity and error-correcting capability of KAM when recognizing binary visual images. Simulation results verify that DW2D-KAM has higher storage capacity and better error-correcting capability than those of KAM, and outperforms the recently proposed modular HAM by Seow and Asari.