计算机工程与应用 ›› 2012, Vol. 48 ›› Issue (14): 180-184.

• 图形、图像、模式识别 • 上一篇    下一篇

基于学习超分辨率重建中的样本选择方法

江  静1,2,张雪松3   

  1. 1.华北科技学院 机电工程系,北京 101601
    2.中国矿业大学(北京) 煤炭资源与安全开采国家重点实验室,北京 100083
    3.中国电子科技集团 第53研究所 光电信息控制和安全技术重点实验室,河北 三河 065201
  • 出版日期:2012-05-11 发布日期:2012-05-14

Sample selection method for learning-based image super-resolution

JIANG Jing1,2, ZHANG Xuesong3   

  1. 1.Department of Mechanical-Electrical Engineering, North China Institute of Science and Technology, Beijing 101601, China
    2.State Key Lab of Coal Resources and Safe Mining, China University of Mining and Technology(Beijing), Beijing 100083, China
    3.Electro-optical Information Security Control Laboratory, the 53rd Instute, CETC, Sanhe, Hebei 065201, China
  • Online:2012-05-11 Published:2012-05-14

摘要: 提出一种人脸图像超分辨率重建(Super-Resolution Reconstruction,SRR)的自适应学习样本选择方法。利用局部保持投影(Locality Preserving Projections,LPP)算法的局部保持能力,在人脸图像局部流形上分析其非线性结构特征,并给出了LPP变换向量的数值解法。在LPP的特征空间中动态搜索学习样本,即选择出与输入图像块最为相似的像素块集合。利用选择出的样本通过基于像素块的特征变换法完成超分辨率重建。实验表明,自适应样本选择方法可以快速、有效地选择出少量学习样本,具有良好的图像高频信息复原能力。

关键词: 样本选择, 超分辨率, 人脸图像, 局部保持投影

Abstract: This paper presents an adaptive learning sample selection method for face hallucination. The nonlinear structural features of face images are explored on facial local manifolds using Locality Preserving Projections(LPP) algorithm, and the efficient computation method of the transform vectors of LPP is presented. Learning samples are dynamically picked out in the eigen-space of LPP, i.e., the patch set most similar to the input image patch. The selected samples are used for super-resolution reconstruction by the patch-based eigen-transformation method. Experimental results fully demonstrate that the proposed adaptive sample selection method can fast and efficiently select out a small amount of learning samples, with good reconstruction performance in terms of high-frequency information restoration.

Key words: sample selection, super-resolution, face image, locality preserving projections