anaconda使用相关:https://www.zhihu.com/question/58033789
图像转换:(图像转数组,获得灰度图等)http://www.cnblogs.com/yinxiangnan-charles/p/5928689.html
图像处理基本步骤:https://www.jb51.net/article/128045.htm
图像处理(PIL中函数):https://blog.csdn.net/gzlaiyonghao/article/details/1852726
图像处理库介绍:https://www.cnblogs.com/qiaozhoulin/p/4509954.html
特征提取:HOG,LBP,HAAR:https://blog.csdn.net/jscese/article/details/52954208
特征提取:灰度共生提取:https://blog.csdn.net/qq_23926575/article/details/80599199
https://www.e-learn.cn/content/python/710079
libsvm安装:http://www.mamicode.com/info-detail-2162669.html
PyWavelets安装包:https://pypi.org/project/PyWavelets/#files
深度学习入门(AlexNet):https://blog.csdn.net/chaipp0607/article/details/72847422?utm_source=blogkpcl0
安装TensorFlow:
(1)TensorFlow不支持Python3.7,可以安装另一个环境python3.6:
(2)安装tensorflow: https://www.cnblogs.com/aloiswei/p/6510355.html
第一个例程(mnist): https://blog.csdn.net/ahnu120705097/article/details/78026237
程序有一点错误。根据:http://blog.csdn.net/md5xwl/article/details/71159486 修改
softmax跑程序详细说明:http://wiki.jikexueyuan.com/project/tensorflow-zh/tutorials/mnist_beginners.html
AlexNet网络搭建(没有归一化和响应层):https://blog.csdn.net/qq_28123095/article/details/79776329 具体代码参见https://github.com/yqtaowhu/MachineLearning提供的github链接,代码有一点错误。具体的,删除图中红框框出的那一句话:“with tf.Graph().ad_default():”即可