The aim of this study is to determine whether topographic models of an area undergoing significant morphological changes, generated using the SfM (Structure-from-Motion) method from aerial photogrammetry images acquired in different epochs, can be compared without the need for georeferencing based on ground reference points acquired in each epoch. The main objective is to investigate the possibility of using a single acquisition of ground reference points made with GNSS instruments during a given period to compare the digital terrain model during different epochs. The technological evolution of the land surveying sector means that the instrumentation used to acquire information for the reconstruction of photogrammetric models is constantly improving, with ever higher levels of precision, quality, and accuracy. The method presented in this work, which allows the alignment of different point clouds, their correct spatial georeferencing and the possibility of assessing possible morphological changes in the terrain, has been applied to the monitoring of the tongue and terminal part of the Forni Glacier (Rhaetian Alps, Italy), located in the Stelvio National Park. Due to the nature of these sites, during the acquisition of aerial images at different times (from 2014 to 2021), it was not always possible to acquire points on the ground with GNSS instrumentation for georeferencing, in addition to the fact that the instrumentation used to acquire the images themselves has changed over time. It was therefore necessary to find a method that could make use of the entire available data set. Using Agisoft Metashape photogrammetry software, it was possible to align all the images in a single operation, georeferenced the resulting cloud of tie points, subdivide it into several clouds, each according to its own reference epoch, and generate the corresponding dense clouds. Using the Cloud Compare software, it was possible to visualise and calculate the morphological changes in the terrain, excluding significant variations outside the glacier body. The results showed that, thanks to the ground control points of a single epoch, it was possible to use the entire available dataset and obtain the volume of ice lost, the displacement and the thinning rate of the glacier on annual time scales with excellent accuracy. This method can be applied in various fields to monitor the morphological changes of a surface, especially in mountainous contexts and in areas that are difficult to access. After the initial survey using terrestrial reference points, subsequent monitoring can be efficiently carried out by a single UAV operator without the need to re-acquire georeferencing data. This approach results in significant savings in terms of time, human resources and computational complexity during the processing phase. The only basic requirement is the presence of stable blocks within the area to be aero photogrammetrically surveyed, recognisable as such by the SfM software, such as hydraulic works or man-made structures. Finally, a demonstration of the application of the method to the monitoring of sediment transport in a mountain stream is presented in the penultimate chapter of this work.
Il presente studio si propone di determinare se i modelli topografici di un'area soggetta a significative modifiche morfologiche, prodotti grazie al metodo SfM (Structure-from-Motion) da immagini acquisite con fotogrammetria aerea durante diverse epoche, possano essere confrontati senza la necessità di georeferenziazione basata su punti di riferimento terrestri acquisiti in ciascun periodo. L'obiettivo principale consiste nell'esaminare la possibilità di utilizzare un'unica acquisizione di punti di appoggio a terra, effettuata con strumentazione GNSS in un dato periodo, per confrontare il modello digitale del terreno in diverse epoche. L’evoluzione tecnologica del settore dei rilevamenti terrestri comporta che la strumentazione utilizzata per l’acquisizione delle informazioni per la ricostruzione di modelli fotogrammetrici sia in continuo miglioramento, con livelli di precisione, qualità e accuratezza sempre migliori. Il metodo presentato in questa tesi per riuscire ad allineare diverse nuvole di punti, georeferenziarle correttamente nello spazio e ottenere la possibilità di valutare possibili cambiamenti morfologici del terreno, è stato condotto riferendosi al monitoraggio della lingua e della parte terminale del Ghiacciaio dei Forni (Alpi Retiche, Italia), situato nel Parco Nazionale dello Stelvio. Proprio per la natura di questi luoghi, durante l’acquisizione delle immagini aeree nelle diverse epoche (dal 2014 al 2021), non sempre è stato possibile acquisire punti a terra con strumentazione GNSS per la georeferenziazione, oltre al fatto che la strumentazione utilizzata per l’acquisizione delle immagini stesse è cambiata nel tempo. Da qui l’esigenza di trovare una metodologia che potesse rendere utilizzabile tutto il dataset a disposizione. Utilizzando il software di fotogrammetria Agisoft Metashape, è stato possibile allineare tutte le immagini con un’unica elaborazione, georeferenziare la nuvola di punti di legame ottenuta, suddividerla in più nuvole, ciascuna secondo la propria epoca di riferimento, e generare le rispettive nuvole dense. Grazie all’utilizzo del software Cloud Compare, è stato possibile visualizzare e calcolare le modifiche morfologiche del terreno, escludendo variazioni significative al di fuori del corpo glaciale. I risultati hanno dimostrato che, grazie ai punti di controllo a terra di una sola epoca, è stato possibile utilizzare tutto il dataset a disposizione ottenendo con un’ottima precisione i volumi di ghiaccio persi, lo spostamento e la velocità di assottigliamento del ghiacciaio su scale di tempo annuali. La metodologia in esame può essere applicata in diversi ambiti per il monitoraggio delle modifiche morfologiche di una superficie, specialmente in contesti montani e in aree difficilmente accessibili. Dopo il primo rilevamento, effettuato utilizzando punti di riferimento terrestri, i successivi monitoraggi possono essere condotti in modo efficiente da un singolo operatore UAV, senza la necessità di acquisire nuovamente i dati di georeferenziazione. Questo approccio comporta risparmi significativi in termini di tempo, risorse umane e complessità computazionale durante la fase di elaborazione. L’unico prerequisito fondamentale è la presenza di blocchi stabili all'interno del dominio soggetto al rilevamento aerofotogrammetrico, riconoscibili come tali dal software SfM, come ad esempio opere di ingegneria idraulica o strutture artificiali. Infine, una dimostrazione dell’applicazione del metodo per il monitoraggio del trasporto solido in un torrente montano è presentata nel penultimo capitolo del presente elaborato.
Monitoring morphological changes in a glacier environment based on multitemporal structure-from-motion photogrammetry
CODEGA, NICOLA
2023/2024
Abstract
The aim of this study is to determine whether topographic models of an area undergoing significant morphological changes, generated using the SfM (Structure-from-Motion) method from aerial photogrammetry images acquired in different epochs, can be compared without the need for georeferencing based on ground reference points acquired in each epoch. The main objective is to investigate the possibility of using a single acquisition of ground reference points made with GNSS instruments during a given period to compare the digital terrain model during different epochs. The technological evolution of the land surveying sector means that the instrumentation used to acquire information for the reconstruction of photogrammetric models is constantly improving, with ever higher levels of precision, quality, and accuracy. The method presented in this work, which allows the alignment of different point clouds, their correct spatial georeferencing and the possibility of assessing possible morphological changes in the terrain, has been applied to the monitoring of the tongue and terminal part of the Forni Glacier (Rhaetian Alps, Italy), located in the Stelvio National Park. Due to the nature of these sites, during the acquisition of aerial images at different times (from 2014 to 2021), it was not always possible to acquire points on the ground with GNSS instrumentation for georeferencing, in addition to the fact that the instrumentation used to acquire the images themselves has changed over time. It was therefore necessary to find a method that could make use of the entire available data set. Using Agisoft Metashape photogrammetry software, it was possible to align all the images in a single operation, georeferenced the resulting cloud of tie points, subdivide it into several clouds, each according to its own reference epoch, and generate the corresponding dense clouds. Using the Cloud Compare software, it was possible to visualise and calculate the morphological changes in the terrain, excluding significant variations outside the glacier body. The results showed that, thanks to the ground control points of a single epoch, it was possible to use the entire available dataset and obtain the volume of ice lost, the displacement and the thinning rate of the glacier on annual time scales with excellent accuracy. This method can be applied in various fields to monitor the morphological changes of a surface, especially in mountainous contexts and in areas that are difficult to access. After the initial survey using terrestrial reference points, subsequent monitoring can be efficiently carried out by a single UAV operator without the need to re-acquire georeferencing data. This approach results in significant savings in terms of time, human resources and computational complexity during the processing phase. The only basic requirement is the presence of stable blocks within the area to be aero photogrammetrically surveyed, recognisable as such by the SfM software, such as hydraulic works or man-made structures. Finally, a demonstration of the application of the method to the monitoring of sediment transport in a mountain stream is presented in the penultimate chapter of this work.| File | Dimensione | Formato | |
|---|---|---|---|
|
2024_07_Codega.pdf
non accessibile
Dimensione
7.4 MB
Formato
Adobe PDF
|
7.4 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/222993