首页> 外国专利> DEEP LEARNING SYSTEM USING IMAGE PATTERNING BASED ON CONVOLUTION NEURAL NETWORK, AND IMAGE LEARNING METHOD USING SAME

DEEP LEARNING SYSTEM USING IMAGE PATTERNING BASED ON CONVOLUTION NEURAL NETWORK, AND IMAGE LEARNING METHOD USING SAME

机译:基于卷积神经网络的图像刻画深度学习系统及相同方法的图像学习方法

摘要

The present invention relates to a deep learning system using image patterning based on a convolution neural network, and an image learning method using the same. The deep learning system of the present invention comprises: an image input unit for inputting an input image; a patterning module for allowing the input image received from the image input unit to be generated as a plurality of patterned images; a convolution neural network (CNN) learning unit based on a CNN for learning the input image received from the image input unit, and the patterned images received from the patterning module; a CNN execution unit for receiving learning information from the CNN learning unit and the input image received from the image input unit; and a final classification unit for receiving image information from the CNN execution unit to classify objects of the image information into categories. According to the present invention, an image learning device capable of improving the quality of image learning information vulnerable to various environmental problems (shaking, illuminance, noise, degradation of recognition rate, etc.), and a deep learning system using the same can be provided.;COPYRIGHT KIPO 2017
机译:本发明涉及使用基于卷积神经网络的图像图案化的深度学习系统,以及使用该深度学习系统的图像学习方法。本发明的深度学习系统包括:图像输入单元,用于输入输入图像;以及图案化模块,用于将从图像输入单元接收的输入图像生成为多个图案化图像;基于CNN的卷积神经网络(CNN)学习单元,用于学习从图像输入单元接收的输入图像和从构图模块接收的构图图像; CNN执行单元,用于从CNN学习单元接收学习信息以及从图像输入单元接收的输入图像;最终分类单元,用于从CNN执行单元接收图像信息,以将图像信息的对象分类。根据本发明,能够改善易受各种环境问题(抖动,照度,噪声,识别率下降等)影响的图像学习信息的质量的图像学习装置以及使用该图像学习装置的深度学习系统。提供; COPYRIGHT KIPO 2017

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