Remote sensing image element identification method based on deep learning

The invention discloses a remote sensing image element identification method based on deep learning. The method includes: training the remote sensing image training set by adopting a deep neural network, adding a classification layer to the trained deep neural network, adjusting the deep neural network to obtain an optimal deep neural network, and performing ground feature element automatic identification on the remote sensing image through the optimal deep neural network. According to the method, remote sensing image feature extraction can be completed quickly and accurately, the adaptabilityof features to complex changes such as illumination, angles, scales and backgrounds of the remote sensing images is improved, and the accuracy and efficiency of remote sensing image element recognition are further improved. 本发明公开了一种基于深度学习的遥感图像要素识别方法，采用深度神经网络对遥感影像训练集进行训练，对训练好的深度神经网络添加一个分类层，再对该深度神经网络进行调整，得到最优的深度神经网络,通过该最优的深度神经网络对遥感图像进行地物要素自动识别。本发明能快速和准确的完成遥感图像特征提取，提高特征对遥感图像光照、角度、尺度和背景等复杂变化的适应性，进而提高遥感图像要素识别的准确率和效率。