我自己编写的一个数字图像处理程序，演示了多种数字图像处理算法，包括直方图统计、线性变换、锐化、平滑、中值滤波、梯度法及Roberts梯度轮廓提取、伪彩色变换...

一个用Ｃ语言编写的动画程序：神奇的太空中有一飞船围绕一个星球旋转。-A use of C language animation process: the magic of space has one planet revolving around a spacecraft.

一个用VB.NET编写的图片浏览器-VB.NET prepared with a photo browser

练习绘制各种图形的源码-practice to draw various graphics FOSS

opencv 下的，能打开图片，垂直，图片的缩放，旋转，都被我以注释的形式放在main函数中了。我也是初学者-opencv under, to be able to open pictures, vertical, pictures of zooming and rotation, I have been in the form of notes on the main function. I also beginners

OCR算法代码。 这不是一个完整的系统，没法生成可执行程序。 代码基于Linux/KDE开发，用到了C++标准模板库。 代码的目的是向读者展示一个OCR系统包括哪些部分，如何工作， 读者可以借鉴这些代码，改动后用到自己的系统中。 由于这是从一个完整的OCR系统中抽取的部分代码， 所以这些代码无法单独编译。 两个*.dat文件如下： char_data.dat：字符特征数据 words_de.dat：词库（后处理用） 如果需要完整实用的中文OCR代码，可以和PRTsinghua@hotmail.com联系。-OCR algorithm code. This is not a complete system, we can not generate executable. Based on the Linux code/KDE development, use the C standard template library. The code is intended to show readers an OCR system, including which parts work, readers can learn from these code changes after their use in the system. As this is from a complete OCR system taking part code, so code can not be compiled separately. 2*. dat following documents : char_data.dat : Character Feature Data words_de.dat : Thesaurus (after treatment) for a complete practical needs if the Chinese OCR code, and PRTsinghua@hotmail.com links.

小巧实用的绘图板，可以涂鸦，保存、打开文件，多种颜色进行选择-Compact and practical drawing board, you can graffiti, save, open the file, a variety of colors to choose

利用opencv编写的sift匹配程序，对于研究此算法很有帮助-sift code by opencv

完成对摄像机的自动调焦的功能-Completion of the camera auto-focus function

matlab下的提升小波程序，速度很快，我已经运行过了  22568592matlab97lift

这是我参考了n(n>=5,^_^)篇BP神经网优化方法的论文写出的BP神经网源代码，使用了批处理训练方法，带动量项、学习速率的自适应调整、仅对学习精度没有达到指定要求的模式训练，并且训练精度逐步加大，通过这种方法，极大地加快了训练速度；另外，对于0模式和1模式数目相差很大(5-6倍)都能达到很高的学习精度。-This is my reference to the n (NGT; Chapter 5 = ,^_^) BP neural network optimization method to write the thesis of BP neural network source code, use the batch method of training, the volume driven, adaptive learning rate adjustment, not only learning accuracy requirements specified the model training, and training gradually increased accuracy, this method has greatly accelerated the pace of training; In addition, the 0 mode and a model number vary greatly (5-6 times) to achieve high accuracy of the study.