基于改进的快速稀疏编码的图像特征提取  

Image feature extraction based on modified fast sparse coding algorithm

作　　者：尚丽[1,2] 苏品刚[1] 周燕[1,3] 

机构地区：[1]苏州市职业大学电子信息工程学院,江苏苏州215104 [2]中国科学技术大学自动化系,合肥230026 [3]苏州大学电子信息学院,江苏苏州215006

出　　处：《计算机应用》2013年第3期656-659,共4页journal of Computer Applications

基　　金：国家自然科学基金资助项目(60970058);江苏省自然科学基金资助项目(BK2009131);江苏省"青蓝工程"项目;2010苏州市职业大学创新团队项目(3100125)

摘　　要：考虑图像特征系数的最大化稀疏分布和特征基的正交性,在快速稀疏编码(FSC)模型的基础上,提出一种改进的FSC模型。该模型利用迭代法解决了基于L1范数的归一化最小二乘法和基于L2范数的约束最小二乘法的凸优化问题,能够实现完备基和过完备基的学习,有效提取出图像的最佳特征,且比标准稀疏编码(BSC)模型的收敛速度快。分别利用自然场景图像和掌纹图像作为训练数据进行特征提取测试,并进一步利用提取的特征基进行图像重构实验,同时与BSC模型的图像重构结果进行对比,实验结果证实了所提出的改进FSC模型能够快速、有效地实现图像的特征提取。On the basis of the Fast Sparse Coding （FSC） model, considering the maximum sparse distribution of feature coefficients and the orthogonality of feature bases of an image, a Modified FSC （MFSC） model was proposed in this paper. This FSC algorithm was based on iteratively solving two convex optimization problems： L1 -norm based regularized least square problem and L2-norm based constrained least square problem, and it can realize the learning of complete bases and overcomplete bases, as well as efficiently extract the features of images. Moreover, the convergence speed of FSC is quicker than that of Basic Sparse Coding （BSC）. The images of natural scene and palmprint were used to test the property of FSC algorithm proposed by the authors in feature extraction, and then the extracted features were utilized to image reconstruction. Compared with reconstructed images obtained by BSC, the experimental results verify the validity of the modified FSC in quickly extracting image features.

关 键 词：快速稀疏编码 最小二乘法 L1范数 L2范数 特征提取 图像重构 

分 类 号：TN391.41[电子电信—物理电子学]