Temporal analyses and multi-temporal 3D reconstruction are fundamental for the preservation and maintenance of all forms of Cultural Heritage (CH) and are the basis for decisions related to interventions and promotion. Introducing the fourth dimension of time into three-dimensional geometric modelling of real data allows the creation of a multi-temporal representation of a site. In this way, scholars from various disciplines (surveyors, geologists, archaeologists, architects, philologists, etc.) are provided with a new set of tools and working methods to support the study of the evolution of heritage sites, both to develop hypotheses about the past and to model likely future developments. The capacity to “see” the dynamic evolution of CH assets across different spatial scales (e.g. building, site, city or territory) compressed in diachronic model, affords the possibility to better understand the present status of CH according to its history. However, there are numerous challenges in order to carry out 4D modelling and the requisite multi-data source integration. It is necessary to identify the specifications, needs and requirements of the CH community to understand the required levels of 4D model information. In this way, it is possible to determine the optimum material and technologies to be utilised at different CH scales, as well as the data management and visualization requirements. This manuscript aims to provide a comprehensive approach for CH time-varying representations, analysis and visualization across different working scales and environments: rural landscape, urban landscape and architectural scales. Within this aim, the different available metric data sources are systemized and evaluated in terms of their suitability.

The study presents the analysis of the possible use of limited number of the Sentinel-2 and Sentinel-1 to check if crop declarations that the EU farmers submit to receive subsidies are true. The declarations used in the research were randomly divided into two independent sets (training and test). Based on the training set, supervised classification of both single images and their combinations was performed using random forest algorithm in SNAP (ESA) and our own Python scripts. A comparative accuracy analysis was performed on the basis of two forms of confusion matrix (full confusion matrix commonly used in remote sensing and binary confusion matrix used in machine learning) and various accuracy metrics (overall accuracy, accuracy, specificity, sensitivity, etc.). The highest overall accuracy (81%) was obtained in the simultaneous classification of multitemporal images (three Sentinel-2 and one Sentinel-1). An unexpectedly high accuracy (79%) was achieved in the classification of one Sentinel-2 image at the end of May 2018. Noteworthy is the fact that the accuracy of the random forest method trained on the entire training set is equal 80% while using the sampling method ca. 50%. Based on the analysis of various accuracy metrics, it can be concluded that the metrics used in machine learning, for example: specificity and accuracy, are always higher then the overall accuracy. These metrics should be used with caution, because unlike the overall accuracy, to calculate these metrics, not only true positives but also false positives are used as positive results, giving the impression of higher accuracy. Correct calculation of overall accuracy values is essential for comparative analyzes. Reporting the mean accuracy value for the classes as overall accuracy gives a false impression of high accuracy. In our case, the difference was 10–16% for the validation data, and 25–45% for the test data.

Classification of remote sensing images using machine learning models requires a large amount of training data. Collecting this data is both labor-intensive and time-consuming. In this study, the effectiveness of using pre-existing reference data on land cover gathered as part of the Land Use–Land Cover Area Frame Survey (LUCAS) database of the Copernicus program was analyzed. The classification was carried out in Google Earth Engine (GEE) using Sentinel-2 images that were specially prepared to account for the phenological development of plants. Classification was performed using SVM, RF, and CART algorithms in GEE, with an in-depth accuracy analysis conducted using a custom tool. Attention was given to the reliability of different accuracy metrics, with a particular focus on the widely used machine learning (ML) metric of “accuracy”, which should not be compared with the commonly used remote sensing metric of “overall accuracy”, due to the potential for significant artificial inflation of accuracy. The accuracy of LUCAS 2018 at Level-1 detail was estimated at 86%. Using the updated LUCAS dataset, the best classification result was achieved with the RF method, with an accuracy of 83%. An accuracy overestimation of approximately 10% was observed when reporting the average accuracy ACC metric used in ML instead of the overall accuracy OA metric.

The specific aim of the European project named "Cultural Heritage Through Time" (CHT2) and reported in this paper is to fully integrate the fourth dimension (4D) into Cultural Heritage studies for analysing structures and landscapes over time. Krakow - the Fortress City (Poland) is the one of four case studies of the CHT2, which are used for the time varying reconstruction, analysis, visualization, and preservation. The goal of the project is produce time-varying 3D products, from landscape to architectural scales, to envisage and analyse lost scenarios or visualize changes due to anthropogenic activities or intervention, pollution, wars, earthquakes, or other natural hazards. The results of the case studies will be the basis for web publication.

The article discusses a method for classifying land cover types in rural areas using a trained neural network. The focus is on distinguishing agriculturally cultivated areas and differentiating bare soil from quarry areas. This distinction is not present in publicly available databases like CORINE, UrbanAtlas, EuroSAT, or BigEarthNet. The research involves training a neural network on multi-temporal patches to classify Sentinel-2 images rapidly. This approach allows automated monitoring of cultivated areas, determining periods of bare soil vulnerability to erosion, and identifying open-pit areas with similar spectral characteristics to bare soil. After training the U-Net network, it achieved an average classification accuracy of 90% (OA) in the test areas, highlighting the importance of using OA for multi-class classifications, instead of ACC. Analysis of our main classes revealed high accuracy, 99.01% for quarries, 92.3% for bare soil, and an average of 94.8% for annual crops, demonstrating the model's capability to differentiate between crops at various growth stages and assess land cover categories effectively.

The goal of the research was to perform tests aimed at assessing possibilities of utilising multi-temporal Landsat satellite images for automatic efficient detection of changes (e.g. accumulation and erosion) of the sea coastal zone. The research database was composed of Landsat satellite images, and standardized NDVI vegetation indexes from the years 1998 and 2015, as well as multi-temporal vector maps and aerial orthophotomaps. The result map of Change detection allowed to locate areas, in which diametrical changes in land coverage took place. Satellite images reflecting the condition of the examined area for twenty years enabled outlining changes of Baltic coastal zone, and then determining the rate and extent of transformations of examined part of coast (erosion and accumulation), and also made it possible to trace the migration of dunes. The analysis showed the range of shifting dune displacement in the years 1998-2015 amounts to ca. 180 m. On the examined section of seashore (32 km), the process of erosion and accumulation was detected respectively on the length of 21 km and 10 km along the coast. The changes of accumulation and erosion of the sea coast are easily identifiable and clearly visible. Properly conducted workflow of processing the satellite images have allowed the rapid and efficient detection of changes coastline, dunes and vegetation.