Dissertations / Theses on the topic 'GEE (Google Earth Engine)'

Wetlands are essential for controlling the global climate, sustaining the global hydrological cycle, conserving ecological variety, and ensuring human wellbeing. As wetlands are one of the most endangered environments due to land conversion, infrastructure development, and overexploitation, they require constant monitoring. In Sweden, there are 68 sites recognized as wetlands with international importance. The inundated area and the connectivity of the wetlands are affected by climate change. For this reason, we need to better delineate water bodies in these valuable environments. Advances in remote sensing technologies helped us to improve the monitoring of wetlands; however, detecting the presence of water under vegetation is still a challenge for correctly delineating the water extent. To address this issue and better detect the presence of water below vegetation, we employ different polarization of SAR sentinel-1 data in combination with optical sentinel-2. After preprocessing the images, we use the K-means clustering algorithm provided in the cloud computing platform of Google Earth Engine, to detect the increased backscatter coming from flooded vegetation duo to the double-bounce of the radar signal. We also take advantage of the high-resolution national land cover of Sweden as an ancillary layer to extract only the relevant information in our study area. Finally, we compare our results with hydroclimatic and field data gathered from the study area. Our workflow improves water-extent delineation in Swedish wetlands by 20% on average by detecting hidden water below the vegetation, which is generally not recognized by optical methods. The proposed method can be extended to monitor and study wetlands’ water availability and changes, contributing to the increase of their resilience to anthropogenic pressures and climate change.

La crescente disponibilità di nuove tecnologie, il miglioramento e la miniaturizzazione di quelle attuali, la maggiore accessibilità ai dati con particolare riferimento all’ accesso gratuito a numerosi (NASA, ESA) archivi di immagini telerilevate, la crescita delle potenzialità offerte dalle tecnologie di Earth Observation (EO), possono offrire una grande quantità di informazioni (EO) destinate alla comunità scientifica e non. Tuttavia, per estrarre utili informazioni dalla immensa quantità di dati oggi disponibili occorre definire, progettare ed implementare nuovi approcci di analisi ed interpretazioni soprattutto per quanto concerne le lunghe serie temporali Sorgono nuove sfide come l’elaborazione automatica, robusta ed efficace di tale quantità di dati. La definizione di nuovi approcci di CD è necessaria per sostituire le tecniche che erano efficaci sulla precedente generazione di immagini ma non su quelle di ultima generazione. Altre sfide consistono nella definizione di tecniche robuste ai dati di input con qualità variabile nel tempo. Le immagini di queste lunghe e dense serie temporali hanno informazioni ridondanti. Quindi, è plausibile che le mappe di cambio rilevate tra coppie di immagini estratte da una serie temporale siano correlate tra loro. Quindi diventa necessario, ma non sufficiente, eliminare nozioni ridondanti ai fini di ottenere risultati “puri” e validi. Questa tesi esplora i meccanismi per sfruttare la correlazione temporale delle immagini per migliorare l’estrazione di informazioni da lunghe serie temporali. Vengono proposte nuove tecniche per migliorare i risultati CD ottenuti per una coppia di immagini o più dati telerilevati utilizzando classificazioni non supervisionate, regressioni logistiche per comprendere i trend, estrapolazioni di valori per pixel per analizzare i cambi di vegetazione all’interno delle serie storiche. In questo contesto, è necessario il mascheramento delle nuvole e il ripristino dei pixel coperti da nuvole in lunghe serie temporali acquisite dai sensori ottici. In letteratura, le soluzioni a questo problema spesso non sfruttano la disponibilità di un gran numero di immagini disponibili per l’area di studio ma si basano solo sulla discriminazione monotemporale delle nuvole. Questo richiede lo sviluppo di nuovi paradigmi e nuove tecniche in grado di sfruttare pienamente tutti i dati e di elaborare in modo efficiente lunghe serie temporali di immagini. Gli obiettivi principali di questo lavoro sono lo sviluppo di tecniche automatiche di CD e lo studio di meccanismi non supervisionati per la valutazione e il miglioramento dei risultati di CD che sfruttano lunghe serie temporali. La tesi studia approcci multitemporali in grado di sfruttare appieno le informazioni acquisite sulla scena. In particolare, gli obiettivi e i nuovi contributi della tesi sono: 1. la proposta di un nuovo paradigma per l’estrazione di informazioni da lunghe serie temporali di immagini. Vengono proposti approcci per sfruttare strumenti open-source come Google Earth Engine e dati satellitari a medio-alta risoluzione geometrica su differenti scale spaziali riducendo i tempi di elaborazione delle immagini e di restituzione di output finali. 2. la definizione di modello che permette di monitorare la vegetazione post-incendio attraverso tecniche satellitari non invasive, riuscendo ad estrapolare trend sui massimi annuali degli indici di vegetazione consentendo l’analisi dei valori di singoli pixel ai fini di monitorare eventuali anomalie come: cambi di destinazione d’uso in aree naturali protette, eventi calamitosi innescati dall’incendio come frane, o di una non adeguata salvaguardia del territorio come eventuali tagli di aree forestali non controllate per scopi agricoli o di allevamento. 3. lo sviluppo di approcci per la valutazione e la stima del danno ecosistemico dovuto agli incendi. La comparazione delle mappe delle percorse dal fuoco con i piani urbanistici vigenti italiani, ai fini di perimetrare aree vulnerabili per la salvaguardia del Capitale Naturale.
The increasing availability of new technologies, the improvement and miniaturization of current technologies, the greater accessibility of data with particular reference to free access to numerous (NASA, ESA) archives of remote sensing images, the growth in potential offered by Earth Observation (EO) technologies, can offer a great deal of information (EO) for the scientific and non-scientific community. However, in order to extract useful information from the immense amount of data available today, new analysis and interpretation approaches need to be defined, designed and implemented, especially with regard to long time series New challenges arise such as automatic, robust and efficient processing of this amount of data. The definition of new CD approaches is needed to replace techniques that were effective on the previous generation of images but not on the latest generation. Other challenges include defining techniques that are robust to input data with time-varying quality. Images from these long, dense time series have redundant information. Therefore, it is plausible that the change maps detected between pairs of images extracted from a time series are correlated with each other. Therefore, it becomes necessary, but not sufficient, to eliminate redundant notions in order to obtain 'pure' and valid results. This thesis explores mechanisms for exploiting the temporal correlation of images to improve the extraction of information from long time series. New techniques are proposed to improve CD results obtained for a pair of images or multiple remote sensing data using unsupervised classifications, logistic regressions to understand trends, and per-pixel value extrapolations to analyse vegetation changes within time series. In this context, cloud masking and restoration of cloud-covered pixels in long time series acquired by optical sensors is required. In the literature, solutions to this problem often do not take advantage of the availability of a large number of images available for the study area but rely only on monotemporal cloud discrimination. This requires the development of new paradigms and techniques capable of fully exploiting all data and efficiently processing long time series of images. The main objectives of this work are the development of automatic CD techniques and the study of unsupervised mechanisms for the evaluation and improvement of CD results exploiting long time series. The thesis studies multitemporal approaches capable of fully exploiting the information acquired on the scene. Specifically, the aims and new contributions of the thesis are: 1. the proposal of a new paradigm for extracting information from long time series of images. Approaches are proposed to exploit open-source tools such as Google Earth Engine and medium-high geometric resolution satellite data on different spatial scales, reducing image processing times and returning final outputs. 2. the definition of a model to monitor post-fire vegetation through non-invasive satellite techniques, being able to extrapolate trends on the annual maxima of vegetation indices, allowing the analysis of single pixel values in order to monitor possible anomalies such as: changes of use in protected natural areas, calamitous events triggered by the fire such as landslides, or inadequate land protection such as the cutting of uncontrolled forest areas for agricultural or breeding purposes. 3. the development of approaches for assessing and estimating ecosystem damage due to fires. The comparison of fire maps with current Italian urban plans, in order to delimit vulnerable areas for the safeguarding of Natural Capital.

Об’єкт дослідження: територія міста Києва. Предметом є її зміна за рахунок забудови, в тому числі неспланованої та незаконної. Мета дипломної роботи: визначення проблем та встановлення масштабів забудови території міста Києва за допомогою методів космічного моніторингу. Методи дослідження: обробка літературних джерел, застосування методів класифікації для візуалізації масштабів забудови території міста Києва, використання платформи Google Earth Engine для виконання класифікації та порівняння її результатів, аналіз отриманих даних, виконання певних розрахунків на основі отриманих даних. Результатом виконання роботи є створений алгоритм виконання класифікації знімків супутників Landsat 7 і Sentinel-2 за допомогою платформи Google Earth Engine, отримані зображення масштабів забудови території міста Києва за 2000 - 2019 роки та наочний аналіз ориманих результатів, обрахований відсоток зменшення площі зелених зон, і рослинності загалом, у місті Києві за період 2000-2019 років. Дані, отримані у результаті виконання роботи, можна використовувати для аналізу забудови міста Києва та для впровадження необхідних заходів щодо запобігання винекнення непоправних наслідків безконтрольного містобудування. Застосовані методи та технології рекомендується використовувати в наукових та організаційно-правових цілях, а також, для отримання більш точних результатів, визначений алгоритм класифікації слід виконати за допомогою супутникових знімків з більшою роздільною здатністю.

Dissertação (mestrado)—Universidade de Brasília, Instituto de Geociências, Programa de Pós-Graduação em Geociências Aplicadas, 2017.
Submitted by Raquel Almeida (raquel.df13@gmail.com) on 2017-05-12T19:49:30Z No. of bitstreams: 1 2017_KhalilAliGanem.pdf: 7800118 bytes, checksum: 7922fe6795a865561d79dc0353d14fe4 (MD5)
Approved for entry into archive by Raquel Viana (raquelviana@bce.unb.br) on 2017-05-16T17:02:32Z (GMT) No. of bitstreams: 1 2017_KhalilAliGanem.pdf: 7800118 bytes, checksum: 7922fe6795a865561d79dc0353d14fe4 (MD5)
Made available in DSpace on 2017-05-16T17:02:32Z (GMT). No. of bitstreams: 1 2017_KhalilAliGanem.pdf: 7800118 bytes, checksum: 7922fe6795a865561d79dc0353d14fe4 (MD5) Previous issue date: 2017-05-16
O projeto intitulado Mapeamento Anual da Cobertura e Uso do Solo no Brasil -MapBiomas - é um projeto atual voltado para a classificação da cobertura do solo a nível nacional. As diversas classes abordadas no âmbito do projeto estão sendo identificadas e mapeadas com base em imagens das séries Landsat. Após a seleção de duas cartas, a SC-24-V-C e a SD-23-X-D, com base em critérios específicos, foi feito um comparativo entre o dado com correção atmosférica obtido por meio dos algoritmos FLAASH e QUAC, e o produto em reflectância no topo da atmosfera (LC8_L1T_TOA), sem correção atmosférica. Este último é o dado atualmente utilizado pelo MapBiomas para o Landsat-8 em classificações resultantes de uma árvore de decisão definida empiricamente pela equipe do projeto. Tal investigação se torna necessária pois o MapBiomas necessita obter um panorama acerca de qual dado apresenta o melhor ajuste às classificações, demonstrando-se mais realista ao contexto da Caatinga, o qual se trata do bioma proporcionalmente menos estudado dentre as regiões naturais brasileiras. As classificações foram geradas pelo Code Editor do Google Earth Engine, uma plataforma capaz de processar imagens de satélite na nuvem de forma distribuída e rápida, permitindo que ferramentas de alto desempenho interpretem e analisem uma gama de informações, as quais são visualizadas em mapas. A partir de análises visuais e da aplicação de testes estatísticos de exatidão global e por classe identificou-se o dado que melhor se ajustou ao contexto da Caatinga, nas cartas selecionadas, mostrando-se mais adequado para proceder com o mapeamento da cobertura do solo no bioma. A carta SC-24-V-C apresentou valores do coeficiente Tau para as classificações oriundas do dado sem correção atmosférica e com dados corrigidos pelo FLAASH e QUAC, de, respectivamente, 54,13%, 39,13% e 58,10%. Já a carta SD-23-X-D apresentou resultados para o mesmo índice de, respectivamente, 55,45%, 68,90% e 64,90%. Isso mostrou que o dado com correção atmosférica, de modo geral, mostrou melhor ajuste ao contexto da Caatinga em comparação com o dado em reflectância no topo da atmosfera. Além disso, dentre os dados utilizados, o FLAASH apresentou maior inconsistência, mostrando-se ora o melhor, ora o pior para cada carta, sendo ainda bem complexo para executar em comparação ao QUAC, que por sua vez é mais rápido no tempo de processamento e apresentou melhor desempenho. O dado sem correção atmosférica não demonstrou diferenças significativas em comparação com os dados corrigidos. E, apesar de ter sido mais baixo, mostrou resultados praticamente idênticos em ambas as análises, o que faz com que o dado não deva ser dispensado, devendo apenas ser feitos ajustes nos parâmetros da árvore de decisão para que seu uso seja mais eficiente. Apesar da variabilidade dos resultados estatísticos em função dos tipos de dados utilizados, constatou-se que a plataforma Google Earth Engine se demonstrou prática, rápida e satisfatória para proceder com a classificação da cobertura do solo na Caatinga.
The project entitled Mapping Annual Coverage and Land Use in Brazil - MapBiomas - is a current project focused on the classification of soil cover at a national level. The various classes addressed in the scope of the project are being identified and mapped based on Landsat images. After the selection of the SC-24-VC and SD-23-XD letters based on specific criteria, a comparison was made between the atmospheric correction data obtained using the FLAASH and QUAC algorithms, and the product of reflectance on the top of atmosphere (LC8_L1T_TOA), without atmospheric correction. The latter data is currently used by MapBiomas for Landsat-8 in the soil cover classifications based on a decision tree empirically defined by the project team. This is a necessary research because the MapBiomas project needs to obtain a panorama about which data represents the best adjustment to the classifications, proving to be more realistic to the context of the Caatinga, which is the biome proportionally less studied among the Brazilian natural regions. The classifications were generated by the Code Editor of Google Earth Engine, a platform capable of processing satellite images in the cloud in a distributed and fast way, allowing high performance tools to interpret and analyze a range of information, which is visualized in the form of maps. Based on the visual analysis and the application of statistical tests of global accuracy and accuracy by class, the data that best fit the context of the Caatinga was identified as the more appropriate to proceed with the mapping of the soil cover in the biome. The letter SC-24-V-C presented Tau values for the classifications from the TOA data and with data corrected by FLAASH and QUAC, of 54.13%, 39.13% and 58.10%, respectively. SD-23-X-D presented results for the same index of, respectively, 55.45%, 68.90% and 64.90%. This showed that the atmospheric correction data, in general, presented a better fit to the context of the Caatinga compared to the top-of-atmosphere reflectance data. In addition, among the data used, FLAASH presented a greater inconsistency, showing to be sometimes the best and in other times the worst for each image. And it is still quite complex to be used if compared to QUAC, which is faster in processing time and had a better performance. The data without atmospheric correction did not show significant differences in comparison with the corrected data. And, even though it had lower values, it showed almost identical results in both analysis, which means that the data should not be dispensed, being necessary only adjustments in the decision tree parameters to achieve more efficient results. In the end, despite the variability of the data, it was found that Google Earth Engine is a highly effective tool to proceed with the classification of the ground cover.

Il telerilevamento ha reso possibile l’applicazione di tecnologie in grado di ridurre gli impatti sull’ambiente. Tra queste, la missione Sentinel-2 di ESA è in grado di fornire informazioni per contrastare il cambiamento climatico. In questo elaborato, l’utilizzo di una action cam per il monitoraggio di appezzamenti agricoli correlata all’utilizzo di bande Sentinel-2 e degli indici NDVI e NDWI hanno permesso di valutare l’andamento fenologico in una serie di appezzamenti della Provincia di Ravenna. Il primo risultato è stato quello di ottenere ground truth (GTs) a basso costo che potessero integrarsi al successivo utilizzo del satellite attraverso il software Google Earth Engine. Secondariamente, le GTs accoppiate al dato satellitare hanno permesso di identificare alcune fasi fenologiche (maturazione e fioritura) nel mais, girasole, soia e sorgo. Gli indici NDVI e NDWI descrivono bene la curva di crescita delle piante e in particolare la loro deviazione standard appare utile per identificare alcune variazioni di fenologia legata ad un non omogeneo accrescimento delle piante. L’applicazione di una tecnica di Machine Learning, attraverso il Random Forest Classifier, per creare un modello predittivo delle colture ha permesso di tentare l’identificazione delle colture nell’intera provincia di Ravenna. Le colture sono risultate ben riconosciute nell’area di raccolta delle GTs, mentre l’assenza di verità a terra nell’area vasta non ha permesso di validare i risultati della classificazione. I risultati ottenuti pongono le basi per integrare la conoscenza a terra ottenuta con metodi di rilievo a basso costo con l’informazione del programma Copernicus, per proseguire verso una precoce identificazione delle colture agricole in un’ottica di gestione sostenibile delle produzioni e dei loro sottoprodotti.

La tesi ha l'obbiettivo di testare le capacità del software Google Earth Engine nel telerilevamento satellitare del dissesto e più precisamente nell'individuazione di frane. L'analisi svolta considera frane recenti note avvenute nella provincia di Forlì-Cesena e cerca di riconoscerne l'attivazione attraverso il confronto di immagini satellitari Sentinel nel tempo. Lo scopo è definire le caratteristiche (tipologia, dimensioni) dei fenomeni effettivamente riconoscibili con questa tecnologia, nella prospettiva che essa possa trovare applicazione in ambiti quali lo studio dell'occorrenza delle frane e il loro monitoraggio.

Mapping the Earth’s surface and its rapid changes with remotely sensed data is a crucial tool to un-derstand the impact of an increasingly urban world population on the environment. However, the impressive amount of freely available Copernicus data is only marginally exploited in common clas-sifications. One of the reasons is that measuring the properties of training samples, the so-called ‘fea-tures’, is costly and tedious. Furthermore, handling large feature sets is not easy in most image clas-sification software. This often leads to the manual choice of few, allegedly promising features. In this Master’s thesis degree project, I use the computational power of Google Earth Engine and Google Cloud Platform to generate an oversized feature set in which I explore feature importance and analyse the influence of dimensionality reduction methods. I use Support Vector Machines (SVMs) for object-based classification of satellite images - a commonly used method. A large feature set is evaluated to find the most relevant features to discriminate the classes and thereby contribute most to high clas-sification accuracy. In doing so, one can bypass the sensitive knowledge-based but sometimes arbi-trary selection of input features.Two kinds of dimensionality reduction methods are investigated. The feature extraction methods, Linear Discriminant Analysis (LDA) and Independent Component Analysis (ICA), which transform the original feature space into a projected space of lower dimensionality. And the filter-based feature selection methods, chi-squared test, mutual information and Fisher-criterion, which rank and filter the features according to a chosen statistic. I compare these methods against the default SVM in terms of classification accuracy and computational performance. The classification accuracy is measured in overall accuracy, prediction stability, inter-rater agreement and the sensitivity to training set sizes. The computational performance is measured in the decrease in training and prediction times and the compression factor of the input data. I conclude on the best performing classifier with the most effec-tive feature set based on this analysis.In a case study of mapping urban land cover in Stockholm, Sweden, based on multitemporal stacks of Sentinel-1 and Sentinel-2 imagery, I demonstrate the integration of Google Earth Engine and Google Cloud Platform for an optimised supervised land cover classification. I use dimensionality reduction methods provided in the open source scikit-learn library and show how they can improve classification accuracy and reduce the data load. At the same time, this project gives an indication of how the exploitation of big earth observation data can be approached in a cloud computing environ-ment.The preliminary results highlighted the effectiveness and necessity of dimensionality reduction methods but also strengthened the need for inter-comparable object-based land cover classification benchmarks to fully assess the quality of the derived products. To facilitate this need and encourage further research, I plan to publish the datasets (i.e. imagery, training and test data) and provide access to the developed Google Earth Engine and Python scripts as Free and Open Source Software (FOSS).
Kartläggning av jordens yta och dess snabba förändringar med fjärranalyserad data är ett viktigt verktyg för att förstå effekterna av en alltmer urban världsbefolkning har på miljön. Den imponerande mängden jordobservationsdata som är fritt och öppet tillgänglig idag utnyttjas dock endast marginellt i klassifikationer. Att hantera ett set av många variabler är inte lätt i standardprogram för bildklassificering. Detta leder ofta till manuellt val av få, antagligen lovande variabler. I det här arbetet använde jag Google Earth Engines och Google Cloud Platforms beräkningsstyrkan för att skapa ett överdimensionerat set av variabler i vilket jag undersöker variablernas betydelse och analyserar påverkan av dimensionsreducering. Jag använde stödvektormaskiner (SVM) för objektbaserad klassificering av segmenterade satellitbilder – en vanlig metod inom fjärranalys. Ett stort antal variabler utvärderas för att hitta de viktigaste och mest relevanta för att diskriminera klasserna och vilka därigenom mest bidrar till klassifikationens exakthet. Genom detta slipper man det känsliga kunskapsbaserade men ibland godtyckliga urvalet av variabler.Två typer av dimensionsreduceringsmetoder tillämpades. Å ena sidan är det extraktionsmetoder, Linjär diskriminantanalys (LDA) och oberoende komponentanalys (ICA), som omvandlar de ursprungliga variablers rum till ett projicerat rum med färre dimensioner. Å andra sidan är det filterbaserade selektionsmetoder, chi-två-test, ömsesidig information och Fisher-kriterium, som rangordnar och filtrerar variablerna enligt deras förmåga att diskriminera klasserna. Jag utvärderade dessa metoder mot standard SVM när det gäller exakthet och beräkningsmässiga prestanda.I en fallstudie av en marktäckeskarta över Stockholm, baserat på Sentinel-1 och Sentinel-2-bilder, demonstrerade jag integrationen av Google Earth Engine och Google Cloud Platform för en optimerad övervakad marktäckesklassifikation. Jag använde dimensionsreduceringsmetoder som tillhandahålls i open source scikit-learn-biblioteket och visade hur de kan förbättra klassificeringsexaktheten och minska databelastningen. Samtidigt gav detta projekt en indikation på hur utnyttjandet av stora jordobservationsdata kan nås i en molntjänstmiljö.Resultaten visar att dimensionsreducering är effektiv och nödvändig. Men resultaten stärker också behovet av ett jämförbart riktmärke för objektbaserad klassificering av marktäcket för att fullständigt och självständigt bedöma kvaliteten på de härledda produkterna. Som ett första steg för att möta detta behov och för att uppmuntra till ytterligare forskning publicerade jag dataseten och ger tillgång till källkoderna i Google Earth Engine och Python-skript som jag utvecklade i denna avhandling.

Today, people are used to having technology as a constant aid. This also sets expectations that information should always be available. This, together with ongoing climate change that has led to more natural disasters, has laid the foundation for the need to change the methodology for how geographical data is collected, compiled and visualized when used for crisis communication. This study explores how authorities, at present, communicate with the public during a crisis and how this can be done in an easier and more comprehensible way, with the help of Geographical Information Systems (GIS). The goal is to present a new way of collecting, compiling and visualizing geographical data in order to communicate, as an authority, with the public during a crisis. This has been done using a case study with focus on wildfires. Therefore, most of the work consisted of the creation of a prototype, CMAP – Crisis Management and Planning, that visualizes fire-related data. The basic work of the prototype consisted of determining what data that exists and is necessary for the information to be complete and easily understood together with how the data is best implemented. The existing data was retrieved online or via a scheduled API request. Eventrelated data, which is often created in connection with the event itself, was given a common structure and an automatic implementation into the prototype using Google Fusion Tables. In the prototype, data was visualized in two interactive map-based sections. These sections focused on providing the user with the information that might be needed if one fears that they are within an affected location or providing the user with general preparatory information in different counties. Finally, a non-map-based section was created that allowed the public to help authorities and each other via crowdsource data. This was collected in a digital form which was then directly visualized in the prototype’s map-based sections. The result of this showed, among other things, that automatic data flows are a good alternative for avoiding manual data handling and thus enabling a more frequent update of the data. Furthermore, it also showed the importance of having a common structure for which data to be included and collected in order to create a communication platform. Finally, by visualizing of dynamic polygon data in an interactive environment a development in crisis communication that can benefit the public’s understanding of the situation is achieved. This thesis is limited to the functionality and layout provided by the Google platform, including Google Earth Engine, Google Forms, Google Fusion Tables etc
I dagens samhälle är människan van vid teknik som ett ständigt hjälpmedel. Detta sätter också förväntningar på att information alltid ska vara tillgänglig och uppdaterad. Detta tillsammans med pågående klimatförändringar som lett till fler och svårare naturkatastrofer har lagt grunden till att det finns ett behov av att förändra hur man samlar in, sammanställer och visualiserar geografiska data som används för kommunikation i en krissituation. Denna studie utforskar hur myndigheter, i dagsläget, kommunicerar med allmänheten vid en krissituation och hur detta kan göras på ett enklare och mer givande sätt med hjälp av GIS. Målet är att visa ett nytt sätt att samla in, sammanställa och visualisera geografiska data för att, som myndighet, kommunicera med allmänheten under en kris. Detta har gjorts som i en fallstudie med fokus på skogs- och gräsbränder. Merparten av arbetet bestod därför av framtagande av en prototyp, CMAP – Crisis Management and Planning som visualiserar brandrelaterade data. Grundarbetet till prototypen bestod av att fastställa vilken data som finns och är nödvändig för att informationen skulle bli lättförstådd och komplett samt hur denna bäst implementeras. Den existerande data som implementerades hämtades online eller via ett schemalagt anrop av APIer. Händelserelaterade data skapas ofta i samband med själva händelsen och därför skapades en gemensam struktur och direktimplementation till prototypen för denna data med hjälp av Google Fusion Tables. I prototypen visualiserades data i två interaktiva kartbaserade sektioner. Dessa sektioner fokuserade kring att förse användaren med den information som kan behövas om man befarar att man befinner sig på en drabbad plats eller att förse användaren med allmän förberedande information inom olika län. Slutligen skapades även en icke kartbaserad sektion som möjliggjorde att allmänheten kan hjälpa myndigheter och varandra genom ”crowdsource” data. Denna samlades in i ett digitalt formulär som sedan direkt visualiserades i prototypens kartbaserade delar. Resultatet av detta visade bland annat att automatiska dataflöden är ett bra alternativ för att slippa manuell hantering av data och därmed möjliggöra en mer frekvent uppdatering. Vidare visade det även på vikten av att ha en gemensam struktur för vilken data som ska inkluderas och samlas in för att skapa en kommunikationsplattform. Slutligen är visualisering av dynamiska polygondata i en interaktiv miljö en utveckling av kriskommunikation som kan gynna förståelsen för situationen hos allmänheten. Studien är begränsad till att skapa en plattform baserad på den inbyggda funktionaliteten och designen som erbjuds i Googles plattform, detta inkluderat Google Earth Engine, Google Formulär, Google Fusion Tables etc.

O reconhecimento de padrões com recurso a redes neuronais é cada vez mais utilizado, numa tentativa de dotar as máquinas computacionais de inteligência e capacidade de aprendizagem. Neste trabalho pretende-se reconhecer determinados elementos naturais como ’água’, ’casas’ ou ’estradas’, em imagens aéreas provenientes do Google EarthTMe do Google MapsTM, recorrendo à utilizando redes neuronais para o efeito. Foram realizadas experiências com quatro conjuntos de imagens utilizados para o treino da rede neuronal, com variação de número de neurónios, e foram analisados os erros de classificação testando cinco novos conjuntos de imagens. Foram ainda realizadas diversas experiências quanto aos métodos de extração de características e à aplicação de operadores morfológicos com diferentes elementos estruturais visando o pós-processamento dos resultados obtidos.
Pattern recognition using neural networks is increasingly used in an attempt to provid to machines computational intelligence and learning ability. This work aims to recognize certain natural elements like ’water’, ’house’ or ’roads’ in aerial images from Google Earth TMand Google Maps TM, resorting to using neural networks for the purpose. Experiments were performed with four sets of images used for training the neural network, with varying number of neurons, and analyzed the classification errors by testing five new sets of images. Were also carried out several experiments on methods of feature extraction and application of morphological operators with different structural elements aimed at the post-processing of results.

碩士
國立中央大學
國際永續發展碩士在職專班
107
Forests act as a reservoir of biodiversity, shelter to wildlife, carbon sinks, and mitigates climate change. Despite their importance, they have been threatened by human activities, which is the main driver of deforestation. The objectives of this study are the assessment and analysis of deforestation in the Paraguayan Chaco, South America, by means of satellite image classification and analysis through the cloud-based Google Earth Engine (GEE) platform. Landsat images from the years of 2013 and 2017 were classified by a pixel-based supervised Random Forest Classifier. The classification results from different land-cover in the study area were then use to assess the deforestation, which expose a forest cover loss, in 2013 the forest cover was 172,862 km2 and 163,875.5 km2 in 2017; this revealed a cover loss of 8,986.3 km2 in 4 years. Furthermore, the classification results obtained in GEE platform were validated with validation points, the classification overall accuracy obtained was 0.86 for 2013 and 0.87 for 2017. The results indicate that GEE perform a rapid image processing and it is an effective reliable platform for deforestation assessment within the Chaco area. The results shown that the deforestation process leaves geometry rectangular features in the land surface and they are easily visualized, this suggests that there is a large scale clearing and heavy machinery is use. The main driver of deforestation in the Chaco is cattle ranching, and the results of this study indicate that the deforestation in continue advancing. During the years of study, Paraguay increase the amount of meat exportation, situating the country within the top 10 major countries worldwide exporter of meat, and this coincide with the increase of deforestation in the Chaco, this problem is likely to continue, because Paraguay set a country goal of reaching more cattle heads by 2020. Moreover, the results revealed that the deforestation process was spread within and in the buffer zones of the national forest protected areas, this suggest that there is lack of compliance with law were it stated that the protected areas are forbidden to use for economic purposes. Indigenous communities were also affected by the deforestation in the study period, threatening their home land and heritage.

碩士
國立彰化師範大學
資訊管理學系所
102
The purpose of this project is to investigate Soil and Water Conservation Bureau to initiating Google Earth Enterprise's critical success factors by Delphi technique.The trends of Geographic Information Systems was being an early development of a unique style to the harmonious development of the recent, building unified related software from the previous stand-alone version evolved to a network version and cloud computing. In year 2012 has began to import Google Earth Enterprise Platform after a carefully evaluation by Soil and Water Conservation Bureau to execute global imaging (aerial photos, satellite photos, aerial unmanned vehicles photographs, site survey photographs), vector, terrain, 3D models, video, realtime monitoring screen, project management and graphic display of such information in order to reach the dataset, data sharing and data consistency purposes. SWCB has over 20 business units on-line in the GIS system currently, exchange of information between different systems, through GEE and gradually establish with service oriented architecture and modular platform for GIS integration services. In addition, in order to enrich GEE showcase in the future at home and abroad to provide interfacing multi collected the Open Data Information. The relevant information system which produced by SWCB (Such as rural information, mudslide disaster prevention information, etc.), it can also be collated and provided to government agencies or private institutions to utilize, other than purpose of sharing resources, it is also through various pipeline will promote the desired information about SWCB for promotion and marketing.

Groundwater is a vital non-renewable resource that is being over exploited at an alarming rate. In Saudi Arabia, the majority of groundwater is used for agricultural activities. As such, the mapping of irrigated lands is a crucial step for managing available water resources. Even though traditional in-field mapping is effective, it is expensive, physically demanding, and spatially restricted. The use of remote sensing combined with advanced computational approaches provide a potential solution to this scale problem. However, when attempted at large scales, traditional computing tends to have significant processing and storage limitations. To address the scalability challenge, this project explores open-source cloud-based resources to map and quantify center-pivot irrigation fields on a national scale. This is achieved by first applying a land cover classification using Random Forest which is a machine learning approach, and then implementing a circle detection algorithm. While the analysis represents a preliminary exploration of these emerging cloud-based techniques, there is clear potential for broad application to many problems in the Earth and environmental sciences.

Remote sensing has enabled unprecedented earth observation from space and has proven to be an invaluable tool for agricultural applications and crop management practices. Here we detect seasonal metrics indicating the start of the season (SOS), the end of the season (EOS) and maximum greenness (MAX) based on vegetation spectral signatures and the normalized difference vegetation index (NDVI) for a time series of Landsat-8, Sentinel-2 and PlanetScope imagery of potato, wheat, watermelon, olive and peach/apricot fields. Seasonal metrics were extracted from NDVI curves and the effect of different spatial and temporal resolutions was assessed. It was found that Landsat-8 overestimated SOS and EOS and underestimated MAX due to its low temporal resolution, while Sentinel-2 offered the most reliable results overall and was used to classify the fields in Aljawf. Planet data reported the most precise SOS and EOS, but proved challenging for the framework because it is not a radiometrically normalized product, contained clouds in its imagery, and was difficult to process because of its large volume. The results demonstrate that a balance between the spatial and temporal resolution of a satellite is important for crop monitoring and classification and that ultimately, monitoring vegetation dynamics via remote sensing enables efficient and data-driven management of agricultural system

Trabalho de projeto de mestrado em Sistemas de Informação Geográfica (Tecnologias e Aplicações), Universidade de Lisboa, Faculdade de Ciências, 2021
A área de deteção remota é uma área em desenvolvimento devido a sua capacidade em adquirir remotamente dados da superfície e da atmosfera da Terra ou de qualquer outro planeta. A quantidade massiva de dados e de imagens de satélite existentes, tem contribuído para uma melhoria e para um aumento da qualidade de informação de observação da Terra. As metodologias científicas tais como a classificação supervisionada por imagens de satélite têm sido aplicadas na análise de dados de deteção remota. O presente estudo tem como objetivo avaliar os resultados obtidos dos processos de classificação por diferentes algoritmos, na utilização de diferentes tipos de dados e otimizá-los através da fusão dos mesmos (SAR e multiespectrais), com recurso a plataforma cloud Google Earth Engine. Recorreu-se à classificação supervisionada de imagens de satélite por cinco abordagens distintas, para a classificação de culturas de Inverno, com seis classificadores: Classification and regression trees (CART), Random Forest (RF), Support Vector Machine (SVM), Maximum Entropy (MAXE), Minimum Distance (MD) e Naive Bayes (NB); uma delas, a abordagem com recurso à fusão de dados SAR e dados multiespectrais. De todas as abordagens testadas, os melhores resultados foram obtidos com o classificador Random Forest (RF), na fusão de dados, com um valor de 76, 2% de exatidão global e de 68,3% de coeficiente kappa com 301 bandas, 102 pertencentes as imagens Sentinel-1 (SAR) com polarização VV e VH e 189 bandas de imagens do Sentinel-2 (multiespectrais) com nove bandas por imagem. Verificou-se que a fusão de dados multiespectrais e SAR beneficiam claramente a classificação efetuada, em parte pelo número de imagens utilizadas, fazendo com que as imagens SAR beneficiem os sistemas óticos principalmente na época de Inverno, onde as imagens óticas são mais limitadas, devido a nebulosidade presente; obtendo-se os valores de exatidão global de 76,2% comparativamente aos resultados individuais de 73% e 70,8%, um da coleção de imagens SAR com 112 bandas, e outro dos sistemas óticos com 189 bandas, respetivamente. O conjunto das imagens SAR em Sentinel-1 (com número de órbita 147 e 52) revelam resultados mais elevados do que as imagens individuais dos sistemas multiespectrais em Sentinel-2, tendo em conta a época de Inverno. A classificação de culturas de Inverno foi efetuada com dados fornecidos pelo Instituto de Financiamento de Agricultura e Pescas (IFAP) com informação geográfica das parcelas correspondente à área do Baixo Alentejo e com identificação da classe e área de cada cultura, num método de classificação supervisionado, por repartição em dados de treino e teste. Este projeto foi realizado com a utilização da plataforma Google Earth Engine (GEE) pelo processamento computacional demonstrado para a analise de um grande volume de dados como as coleções de imagens de satélite por series temporais prontas a usar existentes no catálogo de dados; e pela vasta oferta de funções presentes, entre elas os algoritmos de classificação. Revela-se uma plataforma excecional no processamento, análise e classificação, pela sua versatilidade e performance, tornando-se uma ferramenta imprescindível na área de deteção remota que potencializa a utilização de uma quantidade massiva e heterogénea de dados.
The remote sensing area is an area under development, due to the amount of existing geospatial data. The massive amount of data and existing satellite images has contributed to an improvement and an increase in the quality of Earth observation information. Scientific methodologies such as the classification supervised by satellite images have been applied in the analysis of remote sensing data. This project aimed to evaluate the results obtained from the classification processes by different algorithms, in the use of different types of data and to optimize them through the fusion of them (SAR and multispectral), using the Google Earth Engine cloud platform. Supervised classification of satellite images is done by five different approaches, using the GEE cloud platform for the classification of winter crops, with six classifiers: Classification and regression trees (CART), Random Forest (RF), Support Vector Machine (SVM), Maximum Entropy (MAXE), Minimum Distance (MD) and Naive Bayes (NB); in one of them, approach A5 using fusion of SAR and multispectral data. Of all the approaches, RF demonstrated to have the best results in data fusion with a value of 76, 2% of global accuracy and 68.3% of kappa coefficient, reaching peaks in the A5 approach, in the use of 301 bands, 102 belonging to Sentinel-1 (SAR) images with VV and VH polarization and 189 Sentinel-2 image bands (multispectral) with nine bands per image; it was found that the fusion of optical and SAR data clearly benefits the classification made, in part by the number of images used, making SAR images benefit optical systems mainly in the winter season, where optical images are more limited, due the present cloudiness; obtaining the global accuracy values of 76.2% compared to the individual results of 73% and 70.8%, one from the collection of SAR images with 112 bands, and another from the optical systems with 189 bands, respectively. The set of SAR images in Sentinel-1 (with orbit number 147 and 52) show higher results than the individual images of the multispectral systems in Sentinel-2, taking into account the winter season. The classification of winter crops was carried out using data provided by IFAP with geographic information of the plots corresponding to the area of the lower Alentejo and with identification of the class and area of each crop, in a supervised classification method, by distribution in training and test data. This project was carried out using the Google earth Engine (GEE) platform for the computational processing demonstrated for the analysis of a large volume of data such as collections of satellite images by ready-to-use time series existing in the data catalog, and for the wide range of functions present, among them the classification algorithms. It proves to be an exceptional platform in the processing and analysis of the classification of satellite images, due to its versatility and performance, making it an essential tool in the area of remote sensing, leveraging the use of a massive and heterogeneous amount of data.

Crisis-affected communities and global organizations for international aid are becoming increasingly digital as consequence geotechnology popularity. Humanitarian sector changed in profound ways by adopting new technical approach to obtain information from area with difficult geographical or political access. Since 2011, turkey is hosting a growing number of Syrian refugees along southeastern region. Turkish policy of hosting them in camps and the difficulty created by governors to international aid group expeditions to get information, made such international organizations to investigate and adopt other approach in order to obtain information needed. They intensified its remote sensing approach. However, the majority of studies used very high-resolution satellite imagery (VHRSI). The study area is extensive and the temporal resolution of VHRSI is low, besides it is infeasible only using these sensors as unique approach for the whole area. The focus of this research, aims to investigate the potentialities of mid-resolution imagery (here only Landsat) to obtain information from region in crisis (here, southeastern Turkey) through a new web-based platform called Google Earth Engine (GEE). Hereby it is also intended to verify GEE currently reliability once the Application Programming Interface (API) is still in beta version. The finds here shows that the basic functions are trustworthy. Results pointed out that Landsat can recognize change in the spectral resolution clearly only for the first settlement. The ongoing modifications vary for each case. Overall, Landsat demonstrated high limitations, but need more investigations and may be used, with restriction, as a support of VHRSI.

Our planet is traversing the age of human-induced climate change and biodiversity loss. Projected global warming of 1.5 ºC above pre-industrial levels and related greenhouse gas emission pathways will bring about detrimental and irreversible impacts on the interconnected natural and human ecosystem. A global warming of 2 ºC could further exacerbate the risks across the sectors of biodiversity, energy, food, and water. Time- and cost-effective solutions and strategies are required for strengthening humanity’s response to the present environmental and societal challenges. Coastal seascape ecosystems including seagrasses, corals, mangrove forests, tidal flats, and salt marshes have been more recently heralded as nature-based solutions for mitigating and adapting to the climate-related impacts. This is due to their ability to absorb and store large quantities of carbon from the atmosphere. Focusing on seagrass habitats, although occupying only 0.2% of the world’s oceans, they can sequestrate up to 10% of the total oceanic carbon pool, all the while providing important food security, biodiversity, and coastal protection. But seagrass ecosystems, as all of their blue carbon seascape neighbors, are losing 1.5% of their extent per year due to anthropogenic activities. This has adverse implications for global carbon stocks, coastal protection, and marine biodiversity. Seagrass and seascape recession necessitates their science and policy-based management, protection, conservation which will ensure that our planet will remain within its sustainable boundaries in the age of climate change. The present PhD Thesis and research aim is to develop algorithms for seagrass mapping and monitoring leveraging the recent emergences in remote sensing technology―new satellite image archives, machine learning frameworks, and cloud computing―with field data from multiple sources. The main PhD findings are the demonstration of the suitability of Sentinel-2, RapidEye, and PlanetScope satellite imagery for regional to large-scale seagrass mapping; the introduction and incorporation of machine learning frameworks in the context of seagrass remote sensing and data analytics; the development of a semi-analytical model to invert the bottom reflectance of seagrasses; the design and implementation of multi-temporal satellite image approaches in coastal aquatic remote sensing; and the introduction, design and application of a scalable cloud-based tool to scale up seagrass mapping across large spatial and temporal dimensions. The approaches of the present PhD cover the gaps of the existing scientific literature of seagrass mapping in terms of the lack of spatial and temporal scalability and adaptability; the infancy in seagrass and seascape-related artificial intelligence endeavours; the restrictions of local server and mono-temporal approaches; and the absence of new methodological developments and applications using new (mainly open) satellite image archives. I anticipate and envisage that the near-future steps after the completion of my PhD will address the scalability of the designed cloud-native, data-driven mapping tool to standardise, automate, commercialise and democratise mapping and monitoring of seagrass and seascape ecosystems globally. The synergy of the developed momentum around the global seascape with the technological potential of Earth Observation can contribute to humanity’s race to adapt to and mitigate the climate change impacts and avoid cross tipping points in climate patterns, and biodiversity and ecosystem functions.

To ensure the highest possible temporal resolution of SAR data, it is necessary to use all the available acquisition orbits and paths of a selected area. This can be a challenge in a mountainous terrain, where the side-looking geometry of space-borne SAR satellites in combination with different slope and aspect angles of terrain can strongly affect the backscatter intensity. These errors/noises caused by terrain need to be eliminated. Although there have been methods described in the literature that address this problem, none of these methods is prepared for operable and easily accessible time series analysis in the mountainous areas. This study deals with a land cover-specific local incidence angle (LIA) correction method for time-series analysis of forests in mountainous areas. The methodology is based on the use of a linear relationship between backscatter and LIA, which is calculated for each image separately. Using the combination of CORINE and Hansen Global Forest databases, a wide range of different LIAs for a specific forest type can be generated for each individual image. The algorithm is prepared and tested in cloud-based platform Google Earth Engine (GEE) using Sentinel-1 open access data, SRTM digital elevation model, and CORINE and Hansen Global Forest databases. The method was tested...

Hul’qumi’num communities on south eastern Vancouver Island have concerns about the status and safety of marine foods potentially impacted by environmental change and the urbanization and industrialization of their territories. Collaborative research undertaken with the Hul’q’umi’num’ Lands and Resources Society is part of a broader effort to revitalize cultural practices, language, and food systems. Lhuq’us (the Hul’q’umi’num’ language term for pohrpyra/pyropia spp. (commonly known as red laver or black gold)) is a flavourful and nutritious intertidal seaweed that grows on rocky beaches across the Pacific Northwest. Hul’q’umi’num’ language, cultural values, teachings, and family histories are all interwoven into the harvesting and consumption of lhuq’us in Hul’qumi’num territories. Lhuq’us is one of the species that have been persistently mentioned in conversations with state regulatory agencies and though these concerns have been raised for at least two decades there has been no systematic monitoring of the species. There are two broad streams of inquiry taken by thesis thesis. The first, employing ethnographic methodology including interviews and observant participation, seeks to both document the cultural values, oral histories, lived experiences associated with lhuq’us as well as concerns for the future collaborators have for lhuq’us and lhuq’us beaches. The second stream, based in a geographic approach, asks whether Unoccupied Aerial Vehicle (UAV) technologies could be employed to record the status of lhuq’us as a baseline for monitoring. Two study sites in the Salish sea were surveyed using UAV techniques: ȾEL,IȽĆ and St’utl’qulus. The overall accuracies of the UAV imagery classifications and the particular accuracies of the class representing lhuq’us suggest that UAV technologies paired with Google Earth Engine (GEE) object based image analysis (OBIA) methodologies can effectively detect lhuq’us. There are serious concerns and cultural values and practices deeply interconnected with culturally important species like lhuq’us. Through holding these concerns and values side by side with systematic observation and analyses maps and materials were created which communities can use to assert their rights, enact their own monitoring of territories and re-prioritize environmental decision-making done by federal, provincial, and municipal management agencies.
Graduate