5.18

学习matlab的好资料第二部分，再此共享下啦。参考参考(Good information to learn the second part of matlab, and then under the shared friends. Reference Reference)

Exchange rate changes

exercise 4 for icarnegie ssd4

intel的ipp性能库sample code，提供了加速的静态图象处理示例，充分显示了ipp在多核平台下的性能优势(intel performance of ipp library sample code, provided the acceleration of static image processing example, it shows that in multi-core platform ipp performance advantages)

练习1：查找内存错误 练习2：阅读书籍高质量ç 编程指南.PDF林锐（林锐）P95，回答问题4。 练习3：阅读代码sizeof.rar。不运行它，它的分析和写入输出。(Practice 1: Finding Memory Bugs Practice 2: Read book 高质量C编程指南.pdf  by Lin, Rui(林锐) P95, answer question 4. Practice 3: Read the code in sizeof.rar. Without running it, analysis it and write the output.)

我也不是特别清楚，应该是手写数字识别的matlab程序(I do not particularly clear)

读取csv文件 可以实现与matlab交互用，或与其他文件交互，程序较小但可用(read vsc files Interacting with Matlab can be achieved using the Read CSV file, or interact with other documents, the program small but can be used)

估计一个点云数据的表面法线，表面法线是几何体表面的重要属性。(A point estimate the surface normal, surface normal geometrical surface cloud data is an important attribute volume.)

BP算法的实现，压缩感知l1算法，谢谢了(BP algorithm, the compressed sensing algorithm l1, thank you)

说明：  基于非负矩阵分解（NMF）的人脸特征提取算法，NMF基本思想是找到一个线性子空间W，使的构成子空间的基本图像的像素点都是正值，而且人脸图像在子空间上的投影系数也是正数(Non-negative Matrix Factorization (NMF) of facial feature extraction algorithm, NMF basic idea is to find a linear sub-space W, so that the composition of sub-space of the basic image pixels are positive, and face image in the sub-space projection coefficient is positive)