Landuse change detection in a surface coal mine area using multi-temporal high-resolution satellite images

Urban sprawl can negatively impact the archaeological record of an area. In order to study the urbanisation process and its patterns, satellite images were used in the past to identify land-use changes and detect individual buildings and constructions. However, this approach involves the acquisition of high-resolution satellite images, the cost of which is increases according to the size of the area under study, as well as the time interval of the analysis. In this paper, we implemented a quick, automatic and low-cost exploration of large areas, for addressing this purpose, aiming to provide at a medium resolution of an overview of the landscape changes. This study focuses on using radar Sentinel-1 images to monitor and detect multi-temporal changes during the period 2015–2020 in Limassol, Cyprus. In addition, the big data cloud platform, Google Earth Engine, was used to process the data. Three different change detection methods were implemented in this platform as follow: (a) vertical transmit, vertical receive (VV) and vertical transmit, horizontal receive (VH) polarisations pseudo-colour composites; (b) the Rapid and Easy Change Detection in Radar Time-Series by Variation Coefficient (REACTIV) Google Earth Engine algorithm; and (c) a multi-temporal Wishart-based change detection algorithm. The overall findings are presented for the wider area of the Limassol city, with special focus on the archaeological site of “Amathus” and the city centre of Limassol. For validation purposes, satellite images from the multi-temporal archive from the Google Earth platform were used. The methods mentioned above were able to capture the urbanization process of the city that has been initiated during this period due to recent large construction projects.

Download Full-text

DEEP BAYESIAN ACTIVE LEARNING IN HIGH-RESOLUTION SATELLITE IMAGES FOR CHANGE DETECTION IN URBAN AND SUBURBAN AREAS

ISPRS Annals of Photogrammetry Remote Sensing and Spatial Information Sciences ◽

10.5194/isprs-annals-v-3-2021-175-2021 ◽

2021 ◽

Vol V-3-2021 ◽

pp. 175-180

Author(s):

L. Ragia ◽

A. Panagiotopoulou

Keyword(s):

High Resolution ◽

Active Learning ◽

Change Detection ◽

Satellite Images ◽

Robust Statistics ◽

Training Set ◽

Suburban Areas ◽

Training Images ◽

High Resolution Satellite Images ◽

Zero Test

Abstract. In this work the problem of change detection in high-resolution (HR) satellite images is addressed. The active learning (AL) algorithm Bayesian active learning disagreement (BALD) is applied on WorldView images of urban and suburban areas in the island of Crete, Greece. Comparisons with results from random sampling (RS) on AL are carried out. Several cases of selecting different amounts of images in the training set of a convolutional neural network (CNN) are experimented. The results show that the validation accuracy of classification as changed or unchanged of the BALD algorithm is superior to that of the RS algorithm. Indeed, the BALD algorithm achieves zero test error against the test errors 34.6% and 38.5% of the RS algorithm. Actually, as the amount of training images increases, the accuracy also increases. Interesting experiments could be executed in the future utilizing estimators from robust statistics inside the AL acquisition function framework. Up to now in the literature no other work has appeared to present deep AL on WorldView images for change detection.

Download Full-text

PROPOSAL OF CHANGE DETECTION IN LAND COVER FOR PIPELINE MONITORING FROM RAPIDEYE AND IKONOS IMAGES

Brazilian Journal of Geophysics ◽

10.22564/rbgf.v32i4.553 ◽

2014 ◽

Vol 32 (4) ◽

pp. 655 ◽

Cited By ~ 1

Author(s):

Paulina Setti Riedel ◽

Mara Lúcia Marques ◽

Mateus Vidotti Ferreira ◽

Marcelo Elias Delaneze

Keyword(s):

High Resolution ◽

Land Cover ◽

Change Detection ◽

Satellite Images ◽

Bare Soil ◽

Reference Limit ◽

Object Based ◽

High Resolution Satellite Images ◽

Pipeline Monitoring ◽

Structural Preservation

ABSTRACT. The goal of this study was to improve and evaluate the applicability of a methodological procedure of pipeline monitoring to reveal indicators of thirdparty activities that may interfere with the structural preservation of pipes and environmental damages. The procedure was developed from the technique of changedetection through object-based classification of land cover, using high resolution satellite images applied to a section of the Guararema-Mauá – São Paulo pipeline, Brazil. In the seven-month monitoring period performed with RapidEye imaging, an area of 2.024 km2 was identified as area of change, corresponding to 3.30% of thetotal area analyzed. For the monitoring performed with Ikonos imaging during a four-month period, changes were detected in an area of 0.187 km2, which correspondedto 1.92% of the total area analyzed. The main changes in land cover were from Bare Soil to Grassland, due to changes related to the different stages of agriculturalactivity and reforestation areas, as well as the natural regeneration of vegetation over the pipeline and solid waste landfill. The results of the change detection of landcover from object-based classification were close to the technique reference limit for areas with great complexity and diversity of space occupation.Keywords: structural preservation of pipes, object-based classification, high resolution satellite images. RESUMO. Este estudo teve por objetivo avaliar a aplicabilidade de um procedimento metodológico de monitoramento de faixas de dutos que revelem indicativos deatividades de terceiros que podem interferir na integridade estrutural dos dutos e provocar danos ambientais. O procedimento foi desenvolvido a partir da técnica dedetecção de mudanças na cobertura da terra pela classificação baseada no objeto, com utilização de imagens orbitais de alta resolução. Este procedimento foi empregadoem um trecho da faixa de dutos Guararema-Mauá – SP, no monitoramento realizado por meio de imagens RapidEye. Em um período de sete meses, foram identificados 2,024 km2 como área de mudança, que corresponde a 3,30% do total da área analisada. Para o monitoramento realizado a partir da imagem Ikonos, com período de quatro meses, foi identificada como mudança uma área de 0,187 km2, correspondendo a 1,92% do total da área analisada. As principais mudanças ocorridas foramentre Solo Exposto e Vegetaçao Rasteira, devido às alterações ocorridas nos estágios de cultivo agrícola e áreas de reflorestamento, como também, estão associadas às áreas de regeneração da vegetação da faixa de dutos e aterro sanitário. Os resultados da detecção de mudanças da cobertura da terra pela classificação baseada no objeto atingiram acertos próximos ao limite de para esta técnica, em áreas com grande complexidade e diversidade de ocupação do espaço.Palavras-chave: integridade estrutural dos dutos, classificação baseada no objeto, imagens orbitais de alta resolução.

Download Full-text