首页> 外文会议>International Geoscience and Remote Sensing Symposium >Natural language description of remote sensing images based on deep learning
【24h】

Natural language description of remote sensing images based on deep learning

机译:基于深度学习的遥感图像自然语言描述

获取原文

摘要

The semantic description of remote sensing image is a useful and meaningful task, which can help us to get a better understanding of the scene depicted in the remote sensing images and make better use of the remote sensing images. Nature language provides good solution for describing the semantic information of remote sensing images. Nature language description of a remote sensing image is to generate a meaningful sentence given a remote sensing image. This paper presents a novel method based on deep learning. First, a convolutional neural network is utilized to detect the main objects of the remote sensing images. Then a recurrent neural network language model is utilized to generate the natural language descriptions of the objects which are detected in the first step. Experimental results on a set of remote sensing images demonstrate that the proposed method is able to generate desirable description of the scene.
机译:遥感图像的语义描述是一项有益而有意义的任务,可以帮助我们更好地理解遥感图像中描绘的场景,更好地利用遥感图像。自然语言为描述遥感图像的语义信息提供了很好的解决方案。遥感图像的自然语言描述是在给定遥感图像的情况下生成有意义的句子。本文提出了一种基于深度学习的新方法。首先,利用卷积神经网络来检测遥感图像的主要对象。然后,使用递归神经网络语言模型来生成在第一步中检测到的对象的自然语言描述。在一组遥感图像上的实验结果表明,所提出的方法能够生成所需的场景描述。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

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