Classification of Hyperspectral Remote Sensing Images Using Deep Learning

Hyperspectral image (HSI) classification is a popular technique for analysing remotely detected images. One of the main issues in HSI classification is the large dimensionality of the data. Principal component analysis is a popular method used to simplify HSI. Convolutional neural networks are also utilised for processing visual input and have shown promising results in classification of data. In this project, an integrated method is recommended for HSI classification that combines PCA with two-dimensional convolutional neural networks, three-dimensional convolution neural networks, and support vector machines. We evaluate this approach based on performance on two widely used datasets, namely Indian Pines and KSC. Our test outcomes indicate that the suggested technique works effectively in terms of classification accuracy than other cutting-edge methods. The suggested method could be used in a variety of domains, including environment monitoring and vegetation mapping.