基于深度学习理论的无人机影像分类

针对无人机影像深度学习分类方法缺乏现状,本文利用深度学习理论卷积神经网络方法对无人机影像进行了分类。该法首先抽取无人机影像作为训练集和检验集,然后建立一个2个卷积层-池化层的卷积神经网络模型进行深度学习,通过设定参数并运行模型实现无人机影像分类。实验表明,本文提出的方法可完成较复杂地区无人机影像分类,其分类精度与支持向量机方法相当,为无人机遥感影像分类提供了一个崭新的技术视点。

Unmanned Aerial Vehicle Image Classification Based on Deep Learning Theory

Aiming at the lack of deep learning classification methods for UAV images,this paper classified UAV images using convolutional neural network based on deep learning theory.Firstly,the UAV image was extracted as training set and test set.Then,a convolution neural network model with two convolution layers and pooling layers was established for deep learning.The UAV image classification was realized by setting parameters and running models.Experiments show that the proposed method could classify UAV images in complex areas,and its classification accuracy is comparable to that of support vector machine.It provides a new technical perspective for UAV remote sensing image classification.