首页> 外国专利> A Method For Providing Information Of Classifying Metal Type Printing And Woodblock Printing Using Convolutional Neural NetworkCNN And Image Extraction

A Method For Providing Information Of Classifying Metal Type Printing And Woodblock Printing Using Convolutional Neural NetworkCNN And Image Extraction

机译:利用卷积神经网络和图像提取提供金属类型印刷和木版印刷分类信息的方法

摘要

The present invention provides a method for providing classification information of metal type printing and woodblock printing using a convolutional neural network (CNN) and image extraction which can provide objective and accurate information to classify whether a printing document such as an old document is printed by a woodblock or printed by a metal type. The method for providing classification information of metal type printing and woodblock printing using a CNN and image extraction is executed in a storage device which stores learning image information extracting learning data, a CNN module which is a program executed in a computation device, and a random function, the computation device which is connected to the storage device and receives and computes an output value output from the storage device, an input device which is connected to the computation device and serves as a means for inputting a user command, and a display device which is connected to the computation device and outputs a computed result of the computation device.
机译:本发明提供一种使用卷积神经网络(CNN)和图像提取来提供金属类型印刷和木刻印刷的分类信息的方法,该方法可以提供客观而准确的信息以对诸如旧文档之类的印刷文档是否由印刷者进行印刷进行分类。木刻版或金属版印刷。在使用存储有学习图像信息的学习数据的存储装置,作为在计算装置中执行的程序的CNN模块,以及随机数的存储装置中,执行使用CNN提供金属类印刷和木版印刷的分类信息并进行图像提取的方法。功能,连接到存储设备并接收并计算从存储设备输出的输出值的计算设备,连接到计算设备并用作输入用户命令的装置的输入设备以及显示设备连接到计算设备并输出计算设备的计算结果。

著录项

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号