dicbelevel.zip

  • PUDN用户
    了解作者
  • Visual C++
    开发工具
  • 15.2MB
    文件大小
  • zip
    文件格式
  • 1
    收藏次数
  • 1 积分
    下载积分
  • 6
    下载次数
  • 2014-09-13 09:53
    上传日期
人工智能的经典资料,PCA和RobustPCA等算法,可用于数据降维
dicbelevel.zip
内容介绍
% ========================================================================= % KSVD - Toolbox % ========================================================================= The K-SVD is a new algorithm for training dictionaries for linear representation of signals. Given a set of signals, the K-SVD tries to extract the best dictionary that can sparsely represent those signals. Thorough discussion concerning the K-SVD algorithm can be found in: "The K-SVD: An Algorithm for Designing of Overcomplete Dictionaries for Sparse Representation", written by M. Aharon, M. Elad, and A.M. Bruckstein, and appeared in the IEEE Trans. On Signal Processing, Vol. 54, no. 11, pp. 4311-4322, November 2006. In this toolbox you can find the following files: ================================================ 1. KSVD - the main file in this toolbox that implements the KSVD algorithm. Input and output parameters are described inside. 2. KSVD_NN - a variation of the KSVD algorithm for non-negative matrix factorization (non-negative dictionary and coefficients). The following 3 files implements denoising according to 3 different methods described in "Image Denoising Via Sparse and Redundant representations over Learned Dictionaries", appeared in the IEEE Trans. on Image Processing, Vol. 15, no. 12, pp. 3736-3745, December 2006. =================================================================================================================================== 3. denoiseImageDCT - denoising of an image using an overcomplete DCT dictionary. 4. denoiseImageGlobal - denoising of an image using a global trained dictionary. The global dictionary is stored in the file 'globalTrainedDictionary.mat', which must exist in the directory. Alternatively, this function can be used for denoising of images using some other dictionary, for example, a dictionary that was trained by the K-SVD algorithm, executed by the user. 5. denoiseImageKSVD - denoising of an image using a dictionary trained on noisy patches of the image. The following 3 files are demo files that can be executed without any parameters, ================================================================================ 6. demo1 - run file that executes synthetic test to validate the K-SVD algorithm (the same synthetic test that was presented in the paper). 7. demo2 - run file that executes denoising by 3 different methods, all described in "Image Denoising Via Sparse and Redundant representations over Learned Dictionaries", appeared in the IEEE Trans. on Image Processing, Vol. 15, no. 12, pp. 3736-3745, December 2006. 8. demo3 - run file that executes synthetic test to validate the non-negative variation of the KSVD algorithm (the same test is presented in "K-SVD and its non-negative variant for dictionary design", written by M. Aharon, M. Elad, and A.M. Bruckstein and appeared in the Proceedings of the SPIE conference wavelets, Vol. 5914, July 2005. The rest of the files assist the above files: ============================================ 9. gererateSyntheticDictionaryAndData - Generates a random dictionary according to the parameters, and then generates signals as sparse combinations of the atoms of this dictionary. Finally, it adds while Gaussian noise with a given s.d. 10. displayDictionaryElementsAsImage - displays the atoms of a dictionary as blocks for presentation purposes (see for example, figure 5 in the paper "The K-SVD: An Algorithm for Designing of Overcomplete Dictionaries for Sparse Representation". 11. my_im2col - similar to the function 'im2col', only allow defining the sliding distance between the blocks. The following 3 files implements the OMP (orthogonal matching pursuit) algorithm and the non-negative basis pursuit algorithm. This algorithm is used by the above KSVD and NN-KSVD functions. However, different sparse coding functions (or, implementations) may also be used by changing the relevant call in the KSVD file. ==================================================================================================================================== 12. OMP - OMP algorithm. Finds a representation with fixed number of coefficients for each signal. 13. OMPerr - OMP algorithm. Finds a representation to the signals, allowing a (given) maximal representation error for each. 14. NN_BP - non-negative variation of the basis pursuit. finds a non-negative sparse representatation with a fixed number of coefficients for each signal. For comments or questions please turn to Michal aharon (michal.aharon@hp.com) or Michael Elad (elad@cs.technion.ac.il).
评论
    相关推荐
    • pca.zip
      基于pca降维的人脸识别,运用了多种机器学习方法,绘制了特征脸。
    • PCA人脸识别.zip
      PCA人脸识别全部代码及其相关资料,可以运行
    • PCA.zip
      各种算法特征提取,可利用于对一些数据进行降维处理,提高效果。
    • pca20180401.rar
      PCA程序段总结,PCA初级学习资料,matlab初级学者可以参考。下载学习。
    • PCA.zip
      这是一个关于PCA介绍的内容,该内容详细介绍了PCA的基本原理
    • PCA.ZIP
      pca学习资料,内含:1、A tutorial on Principal Components Analysis.pdf;2、ExercisePCA.m 3、PCA主成分分析.ppt
    • PCA9555
      PCA9555 芯片的资料......................。
    • 有关于PCA算法的资料
      关于PCA算法的一些英文资料 属于signal analysis 的Notes
    • PCA and Whitening
      PCA和白化算法的完整资料,内保包含数据文件以及执行的代码。
    • PCA人脸表情识别.zip
      PCA人脸表情识别全部代码及其资料,可以运行