Convolutional neural networks based dimensionality reduction for hyperspectral images

This paper tackles the challenge of limited material recycling in wood recycling processes, focusing on the automated recognition and separation of diverse materials. Our primary goal is to boost recycling efficiency for cost-effectiveness and environmental sustainability. Leveraging artificial intelligence (AI), particularly deep learning (DL) with convolutional neural networks (CNNs) in hyperspectral imaging (HSI), we emphasize dimensionality reduction to enhance wood recycling processes. We propose a CNN-based dimensionality reduction method, complementing our previous studies, and prove its effectivness through a series of tests.