中文摘要：

在基于稀疏表示分类的模式识别中,字典学习（DL）可以为稀疏表示获得更为精简的数据表示。最近的基于Fisher判别的字典学习（FDDL）可以学习到更加判别的稀疏字典,使得稀疏表示分类具有很强的识别性能。核空间变换可以学习到非线性结构信息,这对判别分类非常有用。为了充分利用核空间特性以学习更加判别的稀疏字典来提升最终的识别性能,在FDDL的基础上,提出了两种核化的稀疏表示DL方法。首先原始训练数据被投影到高维核空间,进行基于Fisher判别的核稀疏表示DLFDKDL;其次在稀疏系数上附加核Fisher约束,进行基于核Fisher判别的核稀疏表示DL（KFDKDL）,使得所学习的字典具有更强的判别能力。在多个公开的图像数据库上的稀疏表示分类实验结果验证了所提出的FDKDL和KFDKDL方法的有效性。

英文摘要：

In pattern recognition based on sparse representation classification,concise representation of data can be obtained for sparse representation via dictionary learning.Recently,Fisher discrimination dictionary learning（FDDL）can obtain very discriminant sparse dictionary,which will bring high recognition performance for sparse representation classification.Transforming data into kernel spaces usually can learn non-linear structure information,which is very useful for discrimination and classification.To make full use of properties of kernel space transformation and to learn more discriminant dictionaries for higher recognition performance,two new dictionary learning methods,based on FDDL,are proposed for kernel sparse representation classification.First,the original training data are projected into high dimensional kernel space and then Fisher discrimination kernel dictionary learning（FDKDL）is proposed for kernel sparse representation classification.Second,kernelized Fisher discrimination criterion is imposed on the sparse coefficients,and then kernelized Fisher discrimination kernel dictionary learning（KFDKDL）is proposed for kernel sparse representation classification,which makes the obtained dictionary have higher discrimination ability.Experiments of sparse representation-based classification on several public image databases demonstrate the effectiveness of the proposed FDKDL and KFDKDL dictionary learning methods.