激光与光电子学进展, 2019, 56 (16): 162803, 网络出版: 2019-08-05
基于改进R-FCN的遥感图像舰船检测 下载: 1243次
Remote Sensing Image Ship Detection Based on Improved R-FCN
遥感 舰船检测 区域全卷积网络 ResNet 合成孔径雷达图像 remote sensing ship detection region-based fully convolutional network ResNet synthetic aperture radar images
摘要
针对传统舰船检测算法难以适应复杂多变的海洋杂波环境,无法实现智能舰船检测的问题,提出了一种改进的基于区域全卷积网络(R-FCN)的检测方法。针对合成孔径雷达(SAR)图像的特点,对R-FCN中的特征提取网络ResNet进行混合尺度卷积核处理,使特征提取网络能够抑制相干斑噪声的影响,有效提取舰船特征。选取高分辨率GF-3与低分辨Sentinel-1卫星SAR图像进行测试,均取得了良好的检测效果,证明了本文算法的有效性。
Abstract
The traditional ship detection algorithm is difficult to adapt in the complex and varied sea clutter environment, and intelligent ship detection is impossible to realize. This study proposes an improved region-based fully convolutional network (R-FCN) detection method. Aiming at the characteristics of synthetic aperture radar (SAR), the feature extraction network ResNet in R-FCN uses a mixed-scale convolution kernel. The feature extraction network can suppress the influence of the speckle noise and effectively extract the ship features. High-resolution GF-3 and low-resolution Sentinel-1 satellite SAR images are selected for the test. Consequently, good results are obtained, proving the effectiveness of the proposed algorithm.
王健林, 吕晓琪, 张明, 李菁. 基于改进R-FCN的遥感图像舰船检测[J]. 激光与光电子学进展, 2019, 56(16): 162803. Jianlin Wang, Xiaoqi Lü, Ming Zhang, Jing Li. Remote Sensing Image Ship Detection Based on Improved R-FCN[J]. Laser & Optoelectronics Progress, 2019, 56(16): 162803.