AFORGE 图像处理手势识别.docx

?/hellogv/article/details/5555689基于Aforge的手势识别之二~~~单点手势识别分类：?/hellogv/article/category/683095计算机视觉2010-05-04 13:46?9058人阅读?/hellogv/article/details/5555689评论(12)?javascript:void(0);收藏?/hellogv/article/details/5555689举报/tag/imageimage/tag/processingprocessing/tag/%e5%9b%be%e5%bd%a2图形/tag/objectobject/tag/cc本文来自/hellogv/ ，引用必须注明出处！ 本文把/hellogv/archive/2010/04/25/5526284.aspxAforge的运动识别与/hellogv/archive/2010/05/04/5554691.aspx前面介绍的手写识别融合在一起，实现单个手指的手势识别。下图演示了本文代码运行的结果，图片有点大，请稍候。。。 我预先让程序学习了B和C这两个字母，然后通过手指的手势识别向程序绘画图形，所以点击recorgize时，就自动把图形的特征对应的字母给识别出来了。 这个例子关键部分在于如何灵活运用Aforge的运动识别，如何判断是要画图，还是普通的移来移去呢？在这里，我判断移动对象的大小，当突然面积增大(即两个指套合并)则开始绘图(手势识别的开始)，当拆开再合并则为解除绘图(手势识别的结束)，说白了就是用一个当前状态=!当前状态去做。?本文的代码可以到这里下载：/source/2313846/source/2313846?下面贴出运动判断的核心代码：[c-sharp]?/hellogv/article/details/5555689view plain/hellogv/article/details/5555689copy/hellogv/article/details/5555689print/hellogv/article/details/5555689?private?void?videoSourcePlayer1_NewFrame(?object?sender,?ref?Bitmap?image?)??{???nowImg?=?(Bitmap)image.Clone();????Bitmap?objectImage?=?colorFilter.Apply(?image?);????//?lock?image?for?further?processing??BitmapData?objectData?=?objectImage.LockBits(?new?Rectangle(?0,?0,?image.Width,?image.Height?),???ImageLockMode.ReadOnly,?image.PixelFormat?);????//?grayscaling??UnmanagedImage?grayImage?=?grayFilter.Apply(?new?UnmanagedImage(?objectData?)?);????//?unlock?image??objectImage.UnlockBits(?objectData?);????//?locate?blobs???blobCounter1.ProcessImage(?grayImage?);???ListRectangle?rects?=?new?ListRectangle();???rects.AddRange(blobCounter1.GetObjectsRectangles());????if?(?rects.Count?0?)???{???#region?去掉内部和黏在一起的对象??for?(int?i?=?0;?i??rects.Count?-?1;?i++)???{???//true表示X轴上不能相交，false表示相交??Boolean?isNoTouchX?=?Math.Max(rects[i?+?1].Right?,?rects[i].Right)?-?Math.Min(rects[i?+?1].Left?,rects[i].Left)??(rects[i].Width?+?rects[i?+?1].Width);???//true表示Y轴上不能相交，false表示相交??Boolean?isNoTouchY?=?Math.Max(rects[i?+?1].Bottom,?rects[i].Bottom)?-?Math.Min(rects[i?+?1].Top,?rects[i].Top)??(re