Advanced Machine Learning and Deep Learning Approaches for Remote Sensing II

doi:10.3390/books978-3-7258-0772-7

Reprint

Download Flyer

Advanced Machine Learning and Deep Learning Approaches for Remote Sensing II

Edited by

April 2024

336 pages

ISBN978-3-7258-0771-0 (Hardback)
ISBN978-3-7258-0772-7 (PDF)

https://doi.org/10.3390/books978-3-7258-0772-7

Free download (PDF)

This book is a reprint of the Special Issue Advanced Machine Learning and Deep Learning Approaches for Remote Sensing II that was published in

Engineering

Environmental & Earth Sciences

Summary

This publication elucidates the application of advanced technologies, including machine learning and deep learning, rooted in artificial intelligence, to the realm of remote sensing. It delineates the methodology employed to address prevailing challenges associated with the processing of images and image signals in remote sensing contexts. These methodologies are inherently computation-intensive, necessitating the utilization of high-performance computing apparatus, notably GPUs. With the evolution of such computational devices, alongside advancements in remote and aerial sensing technologies, it has become feasible to conduct Earth monitoring through high-definition imagery and to amass extensive datasets pertaining to Earth observations. The scholarly articles contained within this reprint detail the latest progress in the domains of big data processing and the employment of artificial intelligence-based techniques for enhancing remote sensing technologies.

Format

Hardback

License

Keywords

hyperspectral image (HSI) classification; transformer; convolutional neural network (CNN); Sequencer; long short-term memory network (LSTM); remote sensing object detection; point representation; sample quality assessment; aerial target recognition; center-ness quality; radar echo extrapolation; sequence-to-sequence (Seq2Seq) network; 3D-Unet; convective nowcasting; hyperspectral unmixing; spectral–spatial attention mechanism; deep learning; autoencoder; moving point target; low SNR; transient disturbance; temporal profile; skip connection; transformer; shifted window; spatial feature extraction (SFE); spatial position encoding (SPE); hyperspectral image (HSI) classification; geostatistical modeling; multiple-point statistics; uncertainty quantification; subglacial topographic model; hydrological model; wildfire detection; generative machine-learning; stochastic modeling; remote sensing; segmentation; uncertainty analysis; deep neural network; adversarial defense; deep ensemble model; unmanned aerial vehicle; remote sensing; image recognition; hyperspectral images classification; network pruning; multi-task optimization; knowledge transfer; multi-objective optimization; 2D DOA estimation; low-elevation-angle targets; L-shaped uniform array; L-shaped sparse array; dilated convolutional autoencoder; dilated convolutional neural network; 3D convolution; spatiotemporal fusion; machine learning; multi-source precipitation; ConvLSTM; F-SVD; ionosphere; peak height of F2 layer; hmF2; machine learning; prediction; data sensor fusion; extended Kalman filter; lidar; radar; super-resolution; remote sensing image; convolutional neural network; transformer; self-similarity; convolutional neural network; gated recurrent unit; ecological service value; ecological–economic harmony; driving mechanism