Building change detection based on remote sensing imagery is a key task for land management and planning e.g., detection of illegal settlements, updating land records and disaster response. Under the post- classification comparison approach, this research aimed to evaluate the feasibility of several classification algorithms to identify and capture buildings and their change between two time steps using very-high resolution images (<1 m/pixel) across rural areas and urban/rural perimeter boundaries. Through an App implemented on the Google Earth Engine (GEE) platform, we selected two study areas in Colombia with different images and input data. In total, eight traditional classification algorithms, three unsupervised (K-means, X-Means y Cascade K-Means) and five supervised (Random Forest, Support Vector Machine, Naive Bayes, GMO maximum Entropy and Minimum distance) available at GEE were trained. Additionally, a deep neural network named Feature Pyramid Networks (FPN) was added and trained using a pre-trained model, EfficientNetB3 model. Three evaluation zones per study area were proposed to quantify the performance of the algorithms through the Intersection over Union (IoU) metric. This metric, with a range between 0 and 1, represents the degree of overlapping between two regions, where the higher agreement the higher IoU values. The results indicate that the models configured with the FPN network have the best performance followed by the traditional supervised algorithms. The performance differences were specific to the study area. For the rural area, the best FPN configuration obtained an IoU averaged for both time steps of 0.4, being this four times higher than the best supervised model, Support Vector Machines using a linear kernel with an average IoU of 0.1. Regarding the setting of urban/rural perimeter boundaries, this difference was less marked, having an average IoU of 0.53 in comparison to 0.38 obtained by the best supervised classification model, in this case Random Forest. The results are relevant for institutions tracking the dynamics of building areas from cloud computing platfo future assessments of classifiers in likewise platforms in other contexts.

Currently, new tools have been implemented that merge high-resolution temporal and spatial images for detection of change land cover. With the purpose of evaluate this type of techniques we generated a time series with Landsat satellite imagery and a time series with simulated images Landsat-MODIS, with the purpose of determining which of the two methods provides the best results in the change quantification in the Sabana of Bogota between 2007 and 2013. The processing consists of (i) Time Series with images Landsat trough BFAST, (ii) getting synthetic images through the ESTARFM algorithm; (iii) time series through BFAST with the use of simulated images. In the time series process, the series incorporating synthetic images and images corrected by the gaps generated the best accuracy indexes (global accuracy: 88.16% y Kappa: 76.52%) with respect to the series that incorporated only the images Landsat (global accuracy: 83% y Kappa: 65.18%); it indicates that densification of time series allow to get the best results in the quantification of changes and dynamics of land cover. The methodology applied represents an advance about generation of synthetic images and monitoring and detection of changes in land cover through time series. This is one of the first studies realized in the country that includes this type of process.

The processes of change detection (CD) in land use and land cover through satellite images acquired in different temporal phases represent a key process to monitor the land, the environment and evaluate disasters. Currently the methodologies used for CD based on pixels do not allow processing a large amount of images massively, so it is not possible to analyze all the available information. Main aim of this research is to evaluate the performance of Deep learning methods for CD, grouped in two approaches, across open mining areas from PlanetScope (PS) image time series of the Cerrejon mine in Colombia. Two approaches for change generation are proposed, one based on post-classification comparison in which two convolutional neural network (CNN) architectures are evaluated: U-Net and Feature Pyramidal Network (FPN) for the classification of mining areas along the time series, for this purpose different models with different hyperparameters were generated and trained to select the most suitable for such process and subsequently perform the difference between the periods of the series; and a second approach based on direct change detection in which a modified U-Net network was evaluated from pairs of images. For this purpose, different models were also trained selecting the most appropriate for the detection of changes for each period of the time series, obtaining a map of changes for each approach, each one the results were validated; the most appropriate approach was number 2 (Direct change detection), with kappa accuracies greater than 0.9 in each period of the time series.

The generation of reliable and up-to-date national land cover information is essential for environmental management, climate action, and territorial planning. In Colombia, the CORINE Land Cover Colombia (CLCC) framework has been the official reference for land cover monitoring since 2000, traditionally updated through expert-based Computer-Assisted PhotoInterpretation (CAPI) at a 1:100,000 scale. However, increasing demands for higher spatial resolution and more frequent temporal updates have made process optimization necessary, driving the incorporation of cloud-based processing and artificial intelligence (AI), including machine learning and deep learning algorithms. This study presents a semi-automated methodology for generating a new generation of harmonized CLCC-compatible raster land cover maps at a 1:50,000 scale—offering four times greater spatial detail than the official vector product—with the capacity for semi-automated annual updates. The approach combines legend harmonization from 55 to 23 classes, historical CORINE Land Cover (CLC) polygon-guided sample generation, spectral stability analysis, and regionalized classification across 190 homogeneous subregions, supported by a reproducible cloud-based architecture. National land cover maps were produced for 2020, 2022, and 2024 with thematic accuracies above 80% and Kappa coefficients up to 0.87, alongside change maps for the 2022–2024 period capturing key dynamics in agricultural frontier expansion, wetland variability, and urban expansion. The resulting products also provide structured inputs for expert-based CAPI workflows, supporting the continuous updating of the official 1:100,000 CLCC map. The results demonstrate the operational capacity of integrating AI, cloud computing, and expert knowledge to strengthen Colombia’s national land cover monitoring system.