RETRACTED: Auto encoder based dimensionality reduction and classification using convolutional neural networks for hyperspectral images (Retracted article. See vol. 106, 2024)

Hyperspectral images (HSI) are adjacent band images commonly used in remote sensing environment; the deep learning methodologies have the important feature for classification process. Additionally, the highest dimensionality of HSI enhances the computational complexity which affects the overall performance. Hence, the dimensionality reduction plays a vital role to enhance the performance while processing the Hyperspectral images. The HSI is initially segmented into the pixels, it belongs to the similar correlation and it is optimized using the neural network framework. Auto Encoder based dimensionality reduction is proposed for performance enhancement that denoising removed. The reconstructed pixel using vectors and also identifying the reconstructing loss enhances the overall accuracy. The Convolutional Neural network framework implements the classification process for Hyperspectral images. The performance analysis results on the proposed technique have improved accuracy and performance compared to the related techniques.