Semantic Segmentation of High-Resolution Remote Sensing Images with Advanced Deep Learning Techniques

Dear Colleagues,

In recent years, semantic segmentation has emerged as a prominent research area within image processing and computer vision. The rapid advancement of deep learning (DL) has significantly fueled interest in this domain. A plethora of influential DL models, including convolutional neural networks (CNNs), generative adversarial networks (GANs), graph convolutional networks (GCNs), multimodal data fusion networks, and Transformer, have been developed for semantic segmentation tasks. These models exhibit remarkable performance across diverse applications, ranging from scene comprehension in autonomous driving to precise segmentation of skin lesions for medical diagnosis and hyperspectral/multispectral image segmentation for remote sensing applications.

By virtue of the advancements in spectral imaging and aerial photography technologies, there has been a notable facilitation in the acquisition of an extensive repository of aerial multispectral and hyperspectral images. These images serve as invaluable resources across various domains of remote sensing applications, encompassing tasks ranging from quantifying forest cover to conducting land-use assessments and projecting urban-planning scenarios. However, despite the commendable strides made in deep learning-based semantic segmentation within the domain of natural images, the transference of such methodologies to the intricate realm of pixel-level or superpixel-level classification/segmentation of remote sensing images (RSIs), including multispectral and hyperspectral imagery, poses a host of formidable challenges.

Diverging from natural images, high-resolution RSIs present a myriad of object categories alongside redundant details. Semantic segmentation methods for RSIs must accommodate their unique characteristics, handle interclass distinction, and maintain intraclass consistency. However, inputting full high-resolution images into DL models is computationally impractical, leading to excessive complexity. Some current approaches sacrifice segmentation accuracy for processing speed using spatial-based image decomposition. This Special Issue seeks original contributions from researchers pioneering high-performance semantic segmentation of high-resolution RSIs, leveraging deep learning to address these challenges.

Dr. Le Sun
Dr. Qian Sun
Dr. Kan Chen
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Remote Sensing is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2700 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• high-resolution/super-pixel remote sensing image segmentation
• semantic segmentation/change detection
• data augmentation
• pixel-wise classification
• zero-shot learning/ensemble learning
• attention mechanisms
• convolutional neural networks/generative adversarial networks/graph convolutional networks/transformer
• multi-scale feature fusion
• multimodal data fusion
• time series image analysis
• computational complexity
• domain adaptation
• explainable AI (XAI)

• Semantic Segmentation of High-Resolution Images with Deep Learning, Volume II in Remote Sensing (16 articles)