Classification of Hyperspectral Remote Sensing Images Based on Three-Dimensional Convolutional Neural Network Model

In response to the problems of low accuracy and long time consumption in traditional hyperspectral remote sensing image classification methods, this paper proposes a hyperspectral remote sensing image classification method based on a three-dimensional convolutional neural network model. Firstly, the image data is preprocessed and normalized. Based on this, a three-dimensional convolutional neural network is introduced into the learning of image data. On this basis, by optimizing the overall connectivity parameters of the convolutional kernel function, hyperspectral remote sensing image classification based on the convolutional kernel function was achieved. Experiments have shown that the algorithm proposed in this article can accurately classify hyperspectral images and achieve good results.