[英]How can we make use of CoreML framework for handwriting detection in iOS 11
如何使用CoreML
将手写图像转换为文本?
One way to achieve this is to use the Vision framework to detect text in your image, which gives you a list of rectangles. 实现此目的的一种方法是使用Vision框架检测图像中的文本,从而为您提供矩形列表。 Then use a neural network that's trained to recognize text on each of these rectangles.
然后使用经过训练的神经网络识别每个矩形上的文本。 You can use Vision to drive Core ML but you still need to provide it with a suitable neural network yourself.
您可以使用Vision来驱动Core ML,但您仍需要自己提供合适的神经网络。 You can find pretrained networks for this on the web, but you'll need to convert them to Core ML using the tools supplied by Apple.
您可以在网上找到预训练网络,但您需要使用Apple提供的工具将它们转换为Core ML。
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.