Abstract
The article proposes a solution for the problem of high-resolution remote sensing data classification by applying deep learning methods and algorithms in conditions of labeled data scarcity. The problem can be solved within the geosystem approach, through the analysis of the genetic uniformity of spatially adjacent entities of different scale and hierarchical level. Advantages of the proposed GeoSystemNet model rest on a large number of freedom degrees, admitting flexible configuration of the model contingent upon the task at hand. Testing GeoSystemNet for classification of EuroSAT dataset, algorithmically augmented after the geosystem approach, demonstrated the possibility to improve the classification precision in conditions of labeled data accuracy by 9% and to obtain the classification precision with a larger volume of training data (by 2%) which is slightly inferior in comparison with other deep models. The article also shows that synthesis of the geosystem approach with deep learning capabilities allows us to optimize the diagnostics of exogeodynamic processes, owing to the calculation of landscape differentiation regularities. Application of the presented approach enabled us to improve the accuracy in detecting landslides at the testing site “Mordovia” by 5% in comparison with the classical approach of using deep models for remote sensing data analysis. The authors advocate that application of the geosystem approach to improve the efficiency of remote sensing data classification through methods, proposed in the article, requires an individual project-based approach to source data augmentation.
Highlights
Development and experimental substantiation of new geoinformation methods and algorithms for automated analysis of spatial data instrumental in the analysis of the state of lands and prediction of natural and man-made emergencies is a pressing challenge of our times
We have developed an algorithm for training dataset augmentation that allows us to download different-scale images of the host area from MapBox application programming interfaces (API), using coordinates of the element from EuroSAT dataset
The findings, presented in the article, allow us to draw the following conclusions: 1) The main importance of the approach to geospatial data analysis by means of deep learning, presented in the paper, rests on in the use of the geosystem approach for profitable augmentation of the training dataset and development of GeoSystemNet deep model that is capable of efficient analysis of these data
Summary
Development and experimental substantiation of new geoinformation methods and algorithms for automated analysis of spatial data (satellite images, digital models and maps, attributive spatial-temporal information) instrumental in the analysis of the state of lands and prediction of natural and man-made emergencies is a pressing challenge of our times. Development of machine learning technologies, including those based on deep neural network models [1], enables us to perform a highly precise automated monitoring of natural resources management systems and to analyze regularities of occurrence of natural processes and phenomena. Automated analysis of spatial data can be made by both traditional hard computing and soft computing, based on a combined use of fuzzy logic, artificial neural networks and evolutionary modeling [2]. The first decade of the 21st century has seen the rise of deep learning [3], methods and principles relying on the use a variety of levels of the non-linear data processing for extraction and transformation of features, analysis and pattern classification
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
