首页> 外国专利> IMAGE STEGO-DETECTION METHOD BASED ON DEEP LEARNING

IMAGE STEGO-DETECTION METHOD BASED ON DEEP LEARNING

机译:基于深度学习的图像隐身检测方法

摘要

Disclosed is an image stego-detection method based on deep learning, comprising: filtering an image correspondingly having a steganographic marker or a real marker in a training set using a high-pass filter, so as to obtain the training set comprising a steganographic residual image and a real residual image; learning the training set on a deep network model, so as to obtain a trained deep network detection model; filtering an image to be detected using the high-pass filter, so as to obtain a residual image to be detected; and detecting the residual image to be detected on the deep network detection model, so as to determine whether the residual image to be detected is a stego-image. The image stego-detection method can realize the creation of a blind-detection model through automatic learning and can relatively accurately distinguish a stego-image.
机译:公开了一种基于深度学习的图像隐身检测方法,包括:使用高通滤波器对训练集中的具有隐写标记或真实标记的图像进行过滤,得到包含隐写残留图像的训练集。和真实的残像;学习基于深度网络模型的训练集,以获得训练好的深度网络检测模型;利用高通滤波器对待检测图像进行滤波,得到待检测残留图像;在深度网络检测模型上检测待检测残留图像,以确定待检测残留图像是否为隐秘图像。该图像隐身检测方法可以通过自动学习来实现盲检测模型的创建,并且可以相对准确地区分出图像。

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