Application of Deep Learning in Geomatics and Satellite Image Processing

Dear Colleagues,

The field of geomatics, which includes geodesy, mapping, and remote sensing, as well as laser scanning and cartography, has seen tremendous development in recent years. With the advent of advanced technologies, such as deep learning, there has been a paradigm shift in the way data are processed and analyzed. Deep learning techniques have not only increased the accuracy and precision of geospatial data analysis but have also accelerated the decision-making process. The forthcoming SI intends to address the use of deep learning in geomatics and satellite image processing in various applications and emphasizes the importance and new capabilities and processing speed in solving complex spatial problems.

One of the most common applications of deep learning in geomatics is the classification of satellite and aerial imagery. Deep learning models, in particular convolutional neural networks (CNNs), show significant performance in automatically extracting features from satellite imagery and aerial images. By training CNN models on labeled datasets, researchers have successfully classified land cover types, vegetation density, urban and rural areas, and other spatial features present in satellite imagery. These accurate classifications help in land management, urban planning, and environmental monitoring, as well as archaeology and historical data processing, among other applications.

Another application of deep learning in geomatics is the detection and segmentation of objects in satellite and aerial images, as well as in laser scanning data. Deep learning methods have outperformed many of the traditional data processing techniques and have enabled the efficient and accurate extraction of individual objects such as buildings, vehicles, defined structures, and water bodies. Similarly, this is also true for point clouds from laser scanning. This information is very valuable in disaster management, infrastructure planning, natural resource monitoring, and the analysis of complex complexes.

Deep learning techniques have also been used to address the problem of change detection and time series analysis in satellite or aerial imagery. By training deep learning models on historical imagery and ground data, researchers can automatically identify and quantify the changes that occur over time. These changes can include deforestation, urban sprawl, glacier retreat, and other land use changes. This application helps in tracking patterns and trends, which ultimately supports conservation efforts and decision making.

The mapping and monitoring of land cover or the Earth's surface, in general, have historically relied on manual interpretations of satellite and aerial imagery or geomatics data in general, both for civil and military purposes. This laborious process has been time-consuming and prone to human error. Later, spectral signature-based classifications or object-oriented classifications based on multispectral or hyperspectral sensing were extensively developed, especially in the nineties and at the turn of the new millennium. Deep learning models have significantly improved this process by automating land cover mapping and monitoring. By training CNN models on multispectral satellite imagery, researchers can accurately delineate land cover categories such as forests, agricultural land, water bodies, and impervious surfaces. This application facilitates better land management, environmental planning, and habitat protection.

It should be noted that these new technologies are also applicable in other fields, such as geology, archaeology, and the analysis of old maps. Another application of deep learning is the exploration of other planets and asteroids, where, after all, we do not have data other than from remote sensing.

The application of deep learning in geomatics and satellite image processing has revolutionized the way geospatial information is acquired and analyzed. By using the power of deep learning models, researchers and industry professionals can automate the tasks that were previously time-consuming and manual in nature. Deep learning increases the accuracy, efficiency, and scalability of geospatial data analysis, leading to better decision making. With the further development of deep learning techniques, its application in geomatics will undoubtedly further contribute to our understanding of the Earth and its complex spatial dynamics.

Prof. Dr. Karel Pavelka
Guest Editor

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. Applied Sciences 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 2400 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.

• deep learning
• remote sensing
• photogrammetry
• laser scanning
• drones
• historical object documentation
• cartography
• geomatics