Developing a Technique for Automatic Lineament Identification Based on the Neural Network Approach

The purpose of this scientific work is to study the potential of neural network technologies in the field of extracting linear structures from Shuttle Radar Topography Mission (SRTM) digital terrain models (DTMs). Linear structures, also known as lineaments, play an important role in the verificatio...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Izvestiya. Atmospheric and oceanic physics 2023-12, Vol.59 (10), p.1271-1280
Hauptverfasser: Grishkov, G. A., Nafigin, I. O., Ustinov, S. A., Petrov, V. A., Minaev, V. A.
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The purpose of this scientific work is to study the potential of neural network technologies in the field of extracting linear structures from Shuttle Radar Topography Mission (SRTM) digital terrain models (DTMs). Linear structures, also known as lineaments, play an important role in the verification of known faults, the identification of fault-fracture structures, and detailing the framework of discontinuous faults, as well as in the exploration of mineral resources. Their accurate and effective extraction in solving the depicted problems is of fundamental importance. The use of neural network technologies provides a number of advantages over sequential algorithms, including the ability to search for universal criteria for identifying lineaments based on a training sample. This paper considers a comprehensive innovative methodology that includes several key stages. The first stage is the author’s method of data preparation, which helps ensure the quality of the training sample and minimize the impact of noise. The second stage is to develop an algorithm for vectorizing the results of the neural network operation, which allows one to easily export the results (lineaments) to a geographic information system (GIS). The third stage provides a method for minimizing the noise component of the training sample and optimizing the selection of synaptic weighting coefficients by retraining the neural network using simulated data reflecting various localization conditions of the lineaments. To verify the results, a spatial comparison of linear structures extracted by the neural network and lineaments isolated by the operator is carried out. The results of this comparison demonstrate the high potential of the proposed approach and show that the use of neural network technologies is a topical and promising approach for solving the problem of extracting linear structures from digital models of the terrain. Positive conclusions are made about the expediency of using the results obtained for their practical application in the field of earth sciences.
ISSN:0001-4338
1555-628X
DOI:10.1134/S0001433823120101