CNKI OCR 服务优势CNKI OCR QUALIFIED SERVICES

依托业界领先的深度学习技术,CNKI OCR智能图文在多语言识别、多类型加工等领域提供强有力的支持,可实现全自动(无人参与)的内容数字化。
CNKI has accumulated years of experience in advanced deep learning technology in the field of automate character recognition, providing strong support in multiple language processing and various type files processing, with automate content digitization technology.

精度识别
High Accuracy of Recognition
基于知网深度学习研究院,强大的深度学习算法和中文检测识别技术,整图中文识别准确率高达99%以上,英文99%以上。
99%+ of Chinese and English Recognition.
稳定高效
Stable Performance
经过上千万次人工对比打磨,产品成熟稳定,弹性灵活的高并发承载及99.95%的可用性保证。
Stable Performance: With millions of artificial test, CNKI OCR guarantees stable performance of 99.95% availability.
简单易用
Simple Operation
操作简单有效(极快速),并且可以同时生产单层和双层PDF。
Easy and efficient operation, to generate both single and double PDF files.

服务内容CNKI OCR SERVICE CONTENT

标准产品Standard Product

精度识别
High Accuracy of Recognition

中文 | 英语 | 俄语 | 法语 | 德语 土耳其语 | 阿拉伯语 | 日语 | 西班牙语等语种

Chinese | English | Russian | French German | Japanese etc.

多类型文件识别
Document Recognition of Various Type

表格 | 竖形文档 | 图纸 等各种类型文件

Tables | Vertical Text | Blueprint etc.

多格式文件识别
Multiple Formats Supported

.doc | .pdf | .jpg | .png | .txt | .gif 等等各种类型文件

.doc | .pdf | .jpg | .png | .txt | .gif etc.

解决方案Solutions

文字识别系统
Text Recognition System

CNKI OCR标准产品,系统模块包括多要素版面分析(要素包含表格、文本、图形、公式等)、版面分析、全文识别和检索、版面还原等。

It is the CNKI OCR standard product, whose modules include multi-element layout analysis (tables, text, figures, formula etc.), layout analysis, full-text recognition and retrieval, text recognition, layout restoring etc.

图纸识别系统
Blueprint Recognition System

系统可支持图纸元素识别、CAD格式图纸解析、数据库管理与检索等。

The system provides support for blueprint multi-element recognition, analysis of blueprint in CAD format, database management and retrieval etc.

图像相似检索与对比系统
Image Similarity Retrieval and Comparison System

系统模块包括图像特征提取、相似检索建库模块、检索查询引擎模块、相似对比模块等。

The system contains modules as image feature identification, image database establishment for similarity retrieval, searching engine for image similarity retrieval, image similarity comparison etc.

多语言识别
Multiple Language Recognition

识别语言包括简体中文、繁体中文、英语、阿拉伯语、日语、法语、德语、俄语、土耳其语等。

The recognized language includes simplified Chinese, traditional Chinese, English, Arabic, Japanese, French, German, Russian, Turkish etc.

软硬件一体化解决方案
Integration of Software & Hardware

解决方案包括标准文字识别系统+手动翻页免拆书籍扫描仪;标准文字识别系统+自动翻页免拆书籍扫描仪。

The solution contains standard text recognition system + automate page turnover scanner; standard text recognition system + manual page turnover scanner.

Trail