实现图像的预处理，包括灰度值转换，二值化，梯度锐化，去离散噪声，整体倾斜调整，字符分割，尺寸归一，紧缩重排等功能，经过神经网络训练后可以字符识别。-image preprocessing, including gray-scale conversion value, the value of the gradient sharpening to discrete noise, overall tilt adjustment, character segmentation, to a size a

通过对图片的一系列处理，得到给定图片中的数字。C++实现，本程序包含灰度化，二值化，去离散，缩放，细化，曲线平滑，去枝，最终通过神经网络对提取的数字信息进行分析比对-Through a series of pictures, and to obtain the figures given in the picture. C++ implementation of this program include grayscale, binary, to discrete, scaling, thinn