opencv 资料备份及机器学习

http://www.360doc.com/content/14/0409/10/10724725_367441247.shtml  中文版本及代码
http://www.itnose.net/detail/6162169.html  mark ar 过程
http://blog.csdn.net/jinshengtao/article/details/17954427  车牌识别
http://blog.csdn.net/guoming0000/article/details/8290146 一卡通画与皮肤变色之初探
http://blog.csdn.net/aptx704610875/article/details/49586811 http://download.csdn.net/download/chuhang_zhqr/9298975 http://blog.csdn.net/jinshengtao/article/details/48604435 http://blog.csdn.net/chuhang_zhqr/article/details/50034669 http://blog.csdn.net/yang_xian521/article/category/910716 http://blog.csdn.net/yang_xian521/article/category/910716  opencv 学习博客
http://docs.opencv.org/3.1.0/index.html  3.1.0版本函数查询首页  http://opencv.org/mastering-opencv-with-practical-computer-vision-projects.html  opencv 官方书籍pdf文档和代码

/////////////////
http://www.52vr.com/article-631-1.html 全景

1.完成基础神经网络nodejs 库
http://blog.csdn.net/google19890102/article/details/32723459  基础神经网络核心公式

防止神经网络的过拟合方法（L2正则化，dropout和输入噪音等）

The Loss Surfaces of Multilayer Networks这篇论文讨论损失函数

大网络拥有许多隐藏层和神经元的情况下，很少有几率掉入局部极值中，而小网络则有可能

2.完成卷积神经网络库
http://cs.stanford.edu/people/karpathy/convnetjs/demo/cifar10.html http://cs.stanford.edu/people/karpathy/convnetjs/demo/mnist.html http://cs.stanford.edu/people/karpathy/convnetjs/demo/classify2d.html http://blog.csdn.net/lk274857347/article/details/53514364 新架构

3.完成增强学习nodejs 库
http://blog.csdn.net/dark_scope/article/details/47056361 rnn公式
http://blog.csdn.net/a635661820/article/details/45390671 rnn
https://arxiv.org/abs/1312.5602  RL增强学习英文版本

Human-level Control through Deep Reinforcement Learning  RL增强学习改进版
http://www0.cs.ucl.ac.uk/staff/D.Silver/web/Teaching.html 介绍QNN

4.完成循环神经网络库
http://lib.csdn.net/article/61/43020?knId=1752 lstm
http://papers.nips.cc/author/ilya-sutskever-3959  论文
http://papers.nips.cc/author/geoffrey-e-hinton-121 http://papers.nips.cc/author/alex-krizhevsky-5653