数字识别代码 第一步：训练网络。使用训练样本进行训练。（此程序中也可以不训练，因为笔者已经将训练好的网络参数保存起来了，读者使用时可以直接识别）第二步：识别。 首先，打开图像（256色） 再次，进行归一化处理。点击“一次性处理” 最后，点击“R”或者使用菜单找到相应项来进行识别识别的结果显示在屏幕上，同时也输出到文件result.txt中 该系统的识别率一般为90％ -Digital identification code the first step: Training Network. The use of training samples for training. (This process may or may not be training, because I have trained network parameters saved, and readers can be directly used to identify) the second step: Identify. First, open the image (256 colors) once again, for normalized. Click on the " one-off deal with"　Finally, click " R"　or use the menu to find the corresponding items of identification to identify the results showed on the screen, but also output to a file Result.txt in the system' s recognition rate is generally 90％