[UWP]使用OCR [英] [UWP]Using OCR

问题描述

我正在使用Media.Ocr，使用OcrEngine和OcrResult是可以的。但是，我试图在位图中识别单个字符（数字），并且引擎无法识别它们。我一直试图找到文档一段时间，我找不到任何东西。
知道我能做什么。谢谢 

I am using Media.Ocr the use of the OcrEngine and OcrResult is ok. However, I am trying to identify single chars (numerical) in bitmaps and the engine doesn't recognize them. I have been trying to find documentation for a while and I couldn't find nothing. Any idea what I can do. Thanks 

Yair

推荐答案

你好Yair22，



OCR引擎有一些限制，OCR引擎假设图像上的所有文本都写在同一个方向，否则可能无法正确识别部分文本。文本识别的准确性取决于图像的质量。以下内容可能导致
读数不准确：

1.模糊图像

2.手写或草书文字

3。艺术字体样式

4.小文字尺寸（西方语言小于15像素，东亚语言小于20像素）

5.复杂背景

6.文字上的阴影或眩光

7.透视扭曲

8.单词开头的大写字母超大或掉落

9.下标，上标或删除文字

Hi Yair22,

The OCR engine have some limits, The OCR engine assumes that all the text on the image is written in the same direction, otherwise part of text might not be recognized correctly. And the accuracy of text recognition depends on the quality of the image. An inaccurate reading may be caused by the following:
1. Blurry images
2. Handwritten or cursive text
3. Artistic font styles
4. Small text size (less than 15 pixels for Western languages, or less than 20 pixels for East Asian languages)
5. Complex backgrounds
6. Shadows or glare over text
7. Perspective distortion
8. Oversized or dropped capital letters at the beginnings of words
9. Subscript, superscript, or strikethrough text

同时，有30种支持的语言。语言是正确文本识别所必需的信息。每种语言都使用一些特定于语言的资源，因此必须事先指定。

Meanwhile, There are 30 supported languages. And the language is required information for correct text recognition. Every language uses some language-specific resources, so it must be specified in advance.

有关OCR的更多信息，您可以参考  https://msdn.microsoft.com/en-us/library/windows/apps/ windows.media.ocr.aspx？f = 255& MSPPError = -2147217396

For more information about OCR, you can refer to https://msdn.microsoft.com/en-us/library/windows/apps/windows.media.ocr.aspx?f=255&MSPPError=-2147217396

这篇关于[UWP]使用OCR的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！