| 基于深度网络的遥感图像目标检测与识别 |
| |
| 作者姓名： | 杨童 朱炬波 王泽龙 刘吉英 |
| |
| 作者单位： | 国防科技大学文理学院,国防科技大学文理学院,国防科技大学文理学院,国防科技大学文理学院 |
| |
| 基金项目： | 国家自然科学基金项目(61671015),国家重点研发计划项目(2017YFB0502700) |
| |
| 摘 要： | 本文针对卫星遥感图像中的典型目标检测与识别问题,采用了基于深度神经网络的回归算法,利用深度网络所学习出来的特征结构针对多分类复杂场景具有很好的识别效果,同时将检测、分类与识别一同进行回归处理。可以有效缩短训练时间,并提升处理效率,保证目标检测的准确度,在实际应用中前景广泛。
|
| 关 键 词： | 深度学习 遥感图像 目标检测 回归算法 |
| 收稿时间： | 2019/3/7 0:00:00 |
| 修稿时间： | 2019/3/7 0:00:00 |
Remote sensing image target detection and recognition based on deep network |
| |
| Authors: | YANG Tong ZHU Jubo WANG Zelong and LIU Jiying |
| |
| Affiliation: | College of Liberal Arts and Sciences,National University of Defense Technology,College of Liberal Arts and Sciences,National University of Defense Technology, |
| |
| Abstract: | Research on typical target detection and recognition problems in satellite remote sensing images. We use a regression algorithm based on deep neural network. The feature structure learned by the deep network has a good recognition effect for multi-class complex scenes. And the structure of the regression processing together with detection, classification and recognition can improve the processing time. And improve the processing efficiency, to ensure the accuracy of target detection, in the practical application of a wide range of prospects. |
| |
| Keywords: | deep learning remote sensing image target detection regression algorithm |
|
| 点击此处可从《》浏览原始摘要信息 |
|
点击此处可从《》下载全文 |
|
