使用TEXT_DETECTION从Cloud Vision API结果区域设置 [英] Results locale from Cloud Vision API using TEXT_DETECTION

问题描述

我使用 npm node-cloud-vision-api

API正确地检测文档的语言但结果字符在西方字符子集中返回，而不与语言环境相对应。我假设它们应该以UTF-8字符返回，但所有特定于语言环境的字符都映射到基本的西方字符子集中。

例如：

Wartosc 是返回内置的Wartość

如何指示API返回正确的UTF-8字符？ b

解决方案

您是否尝试将语言提示传递给OCR检测调用。请按照下面的API参考。 https://cloud.google.com/vision/reference/rest / v1 / images / annotate＃AnnotateImageRequest

I'm using npm node-cloud-vision-api

API correctly detects language of the document but the results characters are returned in western character subset not corresponding to a locale. I assume they should be returned in UTF-8 characters but all the locale specific characters are mapped into basic western character subset.

For example:

Wartosc is return insted of Wartość

How to instruct the API to return correct UTF-8 characters?

解决方案

Have you tried passing in a language hint to the OCR Detection call. Please follow the below API Reference. https://cloud.google.com/vision/reference/rest/v1/images/annotate#AnnotateImageRequest

这篇关于使用TEXT_DETECTION从Cloud Vision API结果区域设置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！