摘要
在地质勘探学领域,无人机遥感技术已成为获取地表数据的重要手段,而卷积神经网络(CNN)因其出色的图像处理能力被广泛应用于图像特征提取。文中介绍了一种基于卷积神经网络的无人机遥感测绘图像特征提取方法,分析了CNN的基本概念与结构、无人机遥感图像的特征类型、网络结构设计、数据预处理、特征提取及特征融合过程,旨在提高地质勘探中无人机遥感图像分析的自动化和准确性。
In the field of geological prospecting,drone remote sensing technology has become an important means to obtain surface data,and convolutional neural networks(CNN)are widely used in image feature extraction due to their excellent image processing capabilities.This paper introduces a feature extraction method of drone remote sensing mapping images based on convolutional neural networks,and analyzes the basic concept and structure of CNN,feature types of drone remote sensing images,network structure design,data preprocessing,feature extraction and feature fusion process,aiming to improve the automation and accuracy of drone remote sensing image analysis in geological exploration.
作者
马宏平
MA Hongping(Shandong Zhengwei Survey&Mapping Co.,Ltd.,Jinan 250000,China)
出处
《移动信息》
2024年第6期257-259,共3页
MOBILE INFORMATION
关键词
卷积神经网络
无人机
遥感测绘图像特征提取
Convolutional neural network
UAV
Feature extraction of remote sensing surveying and mapping images
