[{"@type":"PropertyValue","name":"数据规模","value":"39,993张，227,910个框"},{"@type":"PropertyValue","name":"采集环境","value":"包括影视字幕、广告、手机截屏、漫画、表情包、海报、杂志封面等"},{"@type":"PropertyValue","name":"采集多样性","value":"包括多种互联网图像"},{"@type":"PropertyValue","name":"语言分布","value":"中文、英文（少量）"},{"@type":"PropertyValue","name":"数据格式","value":"图像数据格式为.jpg，标注文档格式为.json"},{"@type":"PropertyValue","name":"标注内容","value":"行级矩形框标注、行级内容转写（少量数据为列级矩形框标注、列级内容转写）"},{"@type":"PropertyValue","name":"准确率","value":"矩形框顶点偏差不超过五个像素为正确检测，检测框精度不低于97%；文本转写精度不低于97%"}]
{"id":171,"datatype":"1","titleimg":"https://bj-oss-datatang-03.oss-cn-beijing.aliyuncs.com/asset/productNew/datatangDatas/%E6%91%84%E5%9B%BE%E7%BD%91_323475251_%E5%B9%BF%E5%91%8A%E6%A6%82%E5%BF%B5%20%E9%BB%91%E6%96%87%E5%AD%97%E5%9C%A8%E7%BA%BF%E9%94%80%E5%94%AE%20%E5%9C%A8%E6%92%95%E7%A0%B4%E7%BA%B8%E4%B8%8B%E7%9A%84%E4%B8%80%E9%A1%B5%E4%B8%8B%E5%BA%8A%E5%8D%95%E6%B4%BB%E5%8A%A8%E5%AE%A3%E4%BC%A0%E6%88%98%E7%95%A5%E6%B6%82%E9%B8%A6%E5%B8%82%E5%9C%BA%E7%AC%94%E8%AE%B0%E8%BE%B9%E7%BC%98%E8%A7%92%E8%90%BD3d%28%E4%BC%81%E4%B8%9A%E5%95%86%E7%94%A8%29.jpg?Expires=4102329599&OSSAccessKeyId=LTAI8NWs2pDolLNH&Signature=pW38Acfk%2FIySHHoE8p0awCyA9Ag%3D","type1":"147","type1str":null,"type2":"147","type2str":null,"dataname":"39,993张互联网图像OCR数据","datazy":[{"title":"数据规模","value":"39,993张，227,910个框"},{"title":"采集环境","value":"包括影视字幕、广告、手机截屏、漫画、表情包、海报、杂志封面等"},{"title":"采集多样性","value":"包括多种互联网图像"},{"title":"语言分布","value":"中文、英文（少量）"},{"title":"数据格式","value":"图像数据格式为.jpg，标注文档格式为.json"},{"title":"标注内容","value":"行级矩形框标注、行级内容转写（少量数据为列级矩形框标注、列级内容转写）"},{"title":"准确率","value":"矩形框顶点偏差不超过五个像素为正确检测，检测框精度不低于97%；文本转写精度不低于97%"}],"datatag":"OCR,多种互联网图像","technologydoc":null,"downurl":null,"datainfo":"40,007张OCR图像数据，文字载体为弹幕、广告、手机截屏和字幕，字体分布包括常规字体、特殊字体、标点符号250,573个框。数据标注方面，标注文字的检测框及对文字内容转写。数据可用于OCR相关任务","standard":null,"dataylurl":null,"flag":null,"publishtime":null,"createby":null,"createtime":null,"ext1":null,"samplestoreloc":null,"hosturl":null,"datasize":null,"industryPlan":null,"keyInformation":["39,993张","OCR图像数据","矩形框标注"],"samplePresentation":["jpg","https://bj-oss-datatang-03.oss-cn-beijing.aliyuncs.com/filesInfoUpload/data/apps/damp/temp/ziptemp/APY161101444_1_demo1691488800839/APY161101444_1_demo/1.jpg?Expires=4102329599&OSSAccessKeyId=LTAI8NWs2pDolLNH&Signature=vAd7NzJlIGbkmTkB7yHkQkWjElQ%3D","/data/apps/damp/temp/ziptemp/APY161101444_1_demo1691488800839/APY161101444_1_demo/1.jpg",""],"officialSummary":"39,993张互联网图像OCR数据采集场景包括影视字幕、广告、手机截屏、漫画、表情包、海报、杂志封面等。语言分布为中文和英文（少量）。在标注方面，对互联网图像进行行级矩形框标注、行级内容转写（少量数据为列级矩形框标注、列级内容转写）。本套互联网图像OCR数据可用于多种互联网图像OCR任务。","dataexampl":"","datakeyword":"互联网图像OCR数据,OCR图像数据,OCR数据,OCR数据集,OCR标注,OCR采集,OCR数据处理,OCR转写数据rn","isDelete":null,"ids":null,"idsList":null,"datasetCode":null,"productStatus":null,"tagTypeEn":null,"tagTypeZh":"类型,语种","website":null,"samplePresentationList":null,"datazyList":null,"keyInformationList":null,"dataexamplList":null,"bgimg":null,"datazyScriptList":null,"datakeywordListString":null,"sourceShowPage":"ocr","BGimg":"","voiceBg":["/shujutang/static/image/comm/audio_bg.png","/shujutang/static/image/comm/audio_bg2.png","/shujutang/static/image/comm/audio_bg3.png","/shujutang/static/image/comm/audio_bg4.png","/shujutang/static/image/comm/audio_bg5.png"],"single":"yes"}
[{"@type":"VideoObject","embedUrl":"p"},{"@type":"VideoObject","embedUrl":"t"},{"@type":"VideoObject","embedUrl":"d"},{"@type":"VideoObject"}]
13051623904
services@datatang.com
