测绘通报 ›› 2021, Vol. 0 ›› Issue (11): 59-64.doi: 10.13474/j.cnki.11-2246.2021.339

• 学术研究 • 上一篇    下一篇

基于Faster R-CNN的遥感影像船舶检测识别

赵振强, 何水原, 梁永志   

  1. 广州海洋地质调查局, 广东 广州 510075
  • 收稿日期:2020-10-22 出版日期:2021-11-25 发布日期:2021-12-02
  • 作者简介:赵振强(1991-),男,硕士,工程师,主要研究方向为遥感影像处理及海洋导航定位。E-mail:919070267@qq.com
  • 基金资助:
    国家重点研发计划(2016YFC0303004)

Research on remote sensing image ship detection and identification based on Faster R-CNN

ZHAO Zhenqiang, HE Shuiyuan, LIANG Yongzhi   

  1. Guangzhou Marine Geological Survey, Guangzhou 510075, China
  • Received:2020-10-22 Online:2021-11-25 Published:2021-12-02

摘要: 随着遥感技术的不断发展,遥感大数据在人们生活中扮演着越来越重要的角色,但由于遥感数据量大,数据处理较为困难。机器学习技术凭借如今硬件科技的发展,使得其自身计算处理能力得到了巨大的提升,因而被广泛地应用于各个领域。由于数据处理和计算机能力的提升,深度学习被广泛应用于遥感领域中。同时,深度学习在其他领域中也被证明是一个极其强大的工具。结合遥感数据量大的特点,本文通用Faster R-CNN的方法实现了对船舶遥感影像的快速检测,并且取得了较高的检测率。

关键词: 遥感影像, 大数据, 深度学习, Faster R-CNN

Abstract: With the development of remote sensing, remote sensing big data is playing an increasingly important role in people's lives, but due to the large amount of remote sensing data, data processing is difficult. Machine learning technology with the development of today's hardware technology, makes its own computational processing capacity has been greatly improved, and thus is widely used in various fields. Due to the increase in data and computer capabilities, deep learning is now widely used in remote sensing, while it has also been proved to be an extremely powerful tool in other fields. Combining the characteristics of large remote sensing data, this paper uses the Faster R-CNN method to achieve fast detection of ships and achieves a high detection rate.

Key words: remote sensing, big data, deep learning, faster R-CNN

中图分类号: