数据堂—4,995张越南语OCR标注及转写数据

4,995张越南语OCR标注及转写数据包括自然场景图片258张，互联网图像2,553张，文本图像2,184张。在标注方面，行级内容：行级四边形框标注、行级内容转写；竖列内容：竖列四边形框标注、竖列内容转写。本套越南语OCR标注及转写数据可用于多种场景下的越南语识别、越南语拍照翻译等任务

This Vietnamese OCR annotated and transcribed dataset contains 4,995 image samples, including 258 natural scene images, 2,553 internet images, and 2,184 text images. For annotation, two types of content are covered: line-level content with quadrilateral bounding box annotation and line-level content transcription; vertical column content with quadrilateral bounding box annotation and vertical column content transcription. This dataset can be applied to Vietnamese language recognition, Vietnamese photo translation and other relevant tasks across various scenarios.

搜集汇总

背景与挑战

背景概述

该数据集包含4,995张越南语OCR图像，涵盖自然场景、互联网和文本三类图像，并提供了行级和列级的四边形框标注与内容转写。它主要用于越南语识别和照片翻译等测试任务，数据格式为.jpg图像和.json标注文件，标注准确率不低于97%。