基于优化Mask-RCNN算法的遥感飞机目标检测
Remote Sensing Aircraft Target Detection Based on Optimized Mask-RCNN Algorithm
葛海婷 1杨铁梅1
作者信息
- 1. 太原科技大学电子信息工程学院,太原 030024
- 折叠
摘要
针对传统的图像检测算法在遥感领域中存在的问题,将改进的Mask-RCNN检测算法应用于遥感领域.通过优化Resnet特征提取网络,提高算法的特征提取能力;通过改进NMS非极大值抑制网络,优化区域推荐网络.并在自建的遥感飞机数据集上验证算法的稳定性以及有效性.经检测,改进的算法能够提升遥感图像中飞机的检测精度,并且有效降低了飞机目标的误检和漏检问题.
Abstract
Aiming at the problems of traditional image detection algorithms in the field of remote sensing,the im-proved Mask RCNN detection algorithm is applied to the field of remote sensing.By optimizing the Resnet feature extraction network,the feature extraction ability of the algorithm is improved,By improving the NMS non maximum suppression network,the regional recommendation network is optimized.The stability and effectiveness of the algo-rithm are verified on the self-built remote sensing aircraft data set.After detection,the improved algorithm can improve the detection accuracy of aircraft in remote sensing images,and effectively reduce the problem of false de-tection and missed detection of aircraft targets.
关键词
Mask-RCNN/深度学习/遥感图像/目标检测Key words
Mask-RCNN/deep learning/remote sensing images/object detection引用本文复制引用
出版年
2024
NETL
NSTL
维普
万方数据