是否可以使用Windows Ink（手写识别）来分析图像？
是否可以使用 MS Ink API 来分析手写的扫描图像，还是仅适用于平板笔输入？
如果可以的话，相关文档在哪里？
Is it possible to use the MS Ink API to analyze a scanned image of handwriting, or does it only work with tablet pen input?
If it is possible, where is the relevant documentation?
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论
评论（2）
http://msdn.microsoft。 com/en-us/library/windows/desktop/ms698159(v=VS.85).aspx
此应用程序演示如何构建手写识别应用程序。Windows Vista SDK还提供了此示例的 C# 和 Visual Basic .NET 版本。本主题引用 Visual Basic .NET 示例，但不同版本之间的概念是相同的。
您还必须研究主类 InkAnalyzer。
http://msdn.microsoft。 com/en-us/library/windows/desktop/bb969147(v=VS.85).aspx
此示例演示如何使用墨迹分析创建表单填写申请，其中表格基于扫描的纸质表格。
http://msdn.microsoft.com/en-us/library/windows/desktop/ms698159(v=VS.85).aspx
This application demonstrates how you can build a handwriting recognition application.The Windows Vista SDK provides versions of this sample in C# and Visual Basic .NET, as well. This topic refers to the Visual Basic .NET sample, but the concepts are the same between versions.
And you must also investigate the main class InkAnalyzer.
http://msdn.microsoft.com/en-us/library/windows/desktop/bb969147(v=VS.85).aspx
This sample shows how to use Ink Analysis to create a form-filling application, where the form is based on a scanned paper form.
不，这是不可能的。
Windows Ink 是一种使用书写过程中的信息（例如笔的上下移动以及移动方向）的技术。这使得手写识别相对容易，但这只是因为它可以获得实时书写数据。
分析以前写的笔迹是非常不同的，而且更加困难，并且需要机器学习。在这种情况下，它使用与人类相同的识别方法。查看智能字符识别和智能单词识别的进展。它需要大量的处理能力，这就是为什么谷歌护目镜等服务必须将图像发送到“经过训练的”机器，即使这样它也无法真正很好地读取手写内容。其尖端技术尚未准备好大规模部署。
No, its not possible.
Windows Ink is technology that uses information from the writing process, such as pen up and down, as well as direction of movement. this makes handwriting recognition relatively easy, but only because it can get the live writing data.
Analyzing previously written handwriting is very different and much more difficult, and requires machine learning. In this case it is kind of using the same method of recognition as humans use. check out the developments in Intelligent Character Recognition and Intelligent Word Recognition. It requires a lot of processing power and thats why services such as google goggles have to send the image to a 'trained' machine, and even then it cant really read handwriting well. its cutting edge technology, not ready at all for mass deployment.