稳健性(进化)
云计算
计算机科学
深度学习
遥感
条件随机场
人工智能
特征提取
计算机视觉
数据挖掘
地质学
生物化学
基因
操作系统
化学
作者
Zhixuan Zhang,Yong Liu,Mingguang Diao
标识
DOI:10.1109/iccasit58768.2023.10351495
摘要
This project intends to integrate SegNet and conditional random field to study SegNet-based cloud detection technology, so as to solve the problems such as noise sensitivity and inaccurate contour extraction of existing cloud detection algorithms. An image cloud recognition method based on multi-view images is proposed. This method is used to improve the image characteristics and algorithm robustness of the cloud cluster to be detected. This method has high accuracy on the premise that the cloud boundary is kept well. The experimental results show that the deep learning network is used for remote sensing target detection, which has good accuracy and robustness, thus providing diversified technical support for the application of domestic high-resolution remote sensing satellites.
科研通智能强力驱动
Strongly Powered by AbleSci AI