Land Cover Classification by using Sentinel-2 Images: A case study in the city of Rome

Nezhad, Meysam Majidi; Heydari, Azim; Fusilli, Lorenzo; Laneve, Giovanni

doi:10.11159/iceptp19.158

Cited by 13 publications

(9 citation statements)

References 28 publications

(30 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is evident that having a detailed and updated depiction of land cover is critical for estimating fire risk. In view of this fact, GIS can be used to classify land cover and vegetation from satellite imagery with the implementation of machine learning algorithms [31,32]. Various algorithms have been used by different studies for land classification, such as k-means clustering [33], maximum likelihood classification [34], and support-vector machines (SVM) [35].…”

Section: Introductionmentioning

confidence: 99%

Fire Risk Probability Mapping Using Machine Learning Tools and Multi-Criteria Decision Analysis in the GIS Environment: A Case Study in the National Park Forest Dadia-Lefkimi-Soufli, Greece

Maniatis

Doganis²,

Chatzigeorgiadis

2022

Applied Sciences

View full text Add to dashboard Cite

Fire risk will increase in the upcoming years due to climate change. In this context, GIS analysis for fire risk mapping is an important tool to identify high risk areas and allocate resources. In the present study, we aimed to create a fire risk estimation model that incorporates recent land cover changes, along with other important risk factors. As a study area, we selected Dadia-Lefkimi-Soufli National Forest Park and the surrounding area since it is one of the most important protected areas in Greece. The area selected for the case study is a typical Mediterranean landscape. As a result, the outcome model is generic and can be applied to other areas. In order to incorporate land cover changes in our model, we used a support vector machine (SVM) algorithm to classify a satellite image captured in September 2021 and an image of the same period two years ago to obtain comparable results. Next, two fire risk maps were created with a combination of land cover and six other factors, using the analytic hierarchy process (AHP) on a GIS platform. The results of our model revealed noticeable clusters of extreme high risk areas, while the overall fire risk in the National Park Forest of Dadia-Lefkimi-Soufli was classified as high. The wildfires of 1st October 2020 and 9th July 2021 confirmed our model and contributed to quantification of their impact on fire risk due to land cover change.

show abstract

Section: Introductionmentioning

confidence: 99%

Fire Risk Probability Mapping Using Machine Learning Tools and Multi-Criteria Decision Analysis in the GIS Environment: A Case Study in the National Park Forest Dadia-Lefkimi-Soufli, Greece

Maniatis

Doganis²,

Chatzigeorgiadis

2022

Applied Sciences

View full text Add to dashboard Cite

show abstract

“…The specific machine learning algorithms used to perform supervised classification for species mapping purposes differs among studies. For example, Supported Vector Machines (SVM) [31,32], Convolutional Neural Networks (CNN) [33] and Random Forests (RF) [34,35] are some of the most frequently used techniques. Nonetheless, of these most popular techniques, RF is typically preferred [5], due to its processing efficiency, robustness and ease of implementation [36][37][38].…”

Section: Introductionmentioning

confidence: 99%

Forest Land Cover Mapping at a Regional Scale Using Multi-Temporal Sentinel-2 Imagery and RF Models

2021

View full text Add to dashboard Cite

Over the last several decades, thanks to improvements in and the diversification of open-access satellite imagery, land cover mapping techniques have evolved significantly. Notable changes in these techniques involve the automation of different steps, yielding promising results in terms of accuracy, class detection and efficiency. The most successful methodologies that have arisen rely on the use of multi-temporal data. Several different approaches have proven successful. In this study, one of the most recently developed methodologies is tested in the region of Galicia (in Northwestern Spain), with the aim of filling gaps in the mapping needs of the Galician forestry sector. The methodology mainly consists of performing a supervised classification of individual images from a selected time series and then combining them through aggregation using decision criteria. Several of the steps of the methodology can be addressed in multiple ways: pixel resolution selection, classification model building and aggregation methods. The effectiveness of these three tasks as well as some others are tested and evaluated and the most accurate and efficient parameters for the case study area are highlighted. The final land cover map that is obtained for Galicia has high accuracy metrics (an overall accuracy of 91.6%), which is in line with previous studies that have followed this methodology in other regions. This study has led to the development of an efficient open-access solution to support the mapping needs of the forestry sector.

show abstract

“…Frequently updated surface maps critical Geographic information systems (GIS) techniques are one of the most commonly used tools for land use classification in urban areas [2,6]. It is not only fast and automatic tools at the same time free from statistical assumption and manipulation of statistic numbers [3,[10][11][12][13][14]. For the analysing LULC changing used methods of Normalize Different Vegetation Index (NDVI) and Normalize Different Buildup Index (NDBI) algorithm to separate green and bare land areas and residential, infrastructure areas by using Landsat images.…”

Section: Introductionmentioning

confidence: 99%

Applying remote sensing techniques to monitor green areas in Tashkent Uzbekistan

et al. 2021

View full text Add to dashboard Cite

Land use and land cover (LULC) change are one of the most important signals of regional environmental monitoring and study. Recently, the pull of capital cities has snowballed, an increasing number of people moving to the cities, especially in developing countries. Consequently, as more people arrive at cities, the more pressure will be on land. Land price getting high and constructions try using open green areas. A wide variety of green areas of different sizes will be solve many urban diseases and ecological problems at the same time improve the quality and life of urban residents, as urban green area provides various ecosystem services. The green area includes parks, woodlands, nature reserves and bare lands. With the population increase and expansion of cities, an increasing amount of open area, woodland and bare land has been converted into construction land, buildings due to the increasing demands and residential land. For the accuracy assessment, we applied an automatically supervised classification using the software QGIS 3.18. The reference values were based on ground truth data and visual interpretation.

show abstract

Land Cover Classification by using Sentinel-2 Images: A case study in the city of Rome

Cited by 13 publications

References 28 publications

Fire Risk Probability Mapping Using Machine Learning Tools and Multi-Criteria Decision Analysis in the GIS Environment: A Case Study in the National Park Forest Dadia-Lefkimi-Soufli, Greece

Fire Risk Probability Mapping Using Machine Learning Tools and Multi-Criteria Decision Analysis in the GIS Environment: A Case Study in the National Park Forest Dadia-Lefkimi-Soufli, Greece

Forest Land Cover Mapping at a Regional Scale Using Multi-Temporal Sentinel-2 Imagery and RF Models

Applying remote sensing techniques to monitor green areas in Tashkent Uzbekistan

Contact Info

Product

Resources

About