1 Introduction

Historical buildings, archaeological sites and artefacts constitute material cultural heritage of the past centuries. Technological advancement of the last decades allowed one to develop the process of documentation and popularize the cultural heritage in the form of spatial objects and structures by means of reality modelling. At present, scientists may obtain accurate 3D models in a short time and distribute them by means of various digital products (Blanco-Pons et al. 2019). Technologies based on images (photogrammetry) and laser scanning (terrestrial laser scanning—TLS) are the most popular, providing dense point clouds for 3D models. TLS was used in multiple projects on cultural heritage to obtain accurate documentation of complex monuments or places (Lerma et al. 2011; Pritchard et al. 2017; Gomes et al. 2014). This technique requires expensive equipment and a long time for processing data. Unlike TLS, photogrammetry provides one with a low-budget method of creating high quality 3D models (Themistocleous 2016; Koutsoudis et al. 2014). Even ultralight unmanned aerial vehicles (ULDs) are an excellent low-cost alternative and enable the acquisition and development of accurate 3D models of cultural heritage objects (Bakirman et al. 2020). Based on the example of the conducted study in which the 3D model of the old wage hall of the Zeche “Bonifacius” in Essen (Germany) was created using an ultralight unmanned aerial vehicle, it was shown that microdrones are an alternative to the national database containing airborne laser scanning (ALS), and the obtained data are valuable point clouds and a source for texturing 3D models (Weißmann et al. 2022). Currently, technological progress makes it possible to use photogrammetric data not only to make advanced and detailed 3D models of cultural heritage objects, but also to later implement these data in the VR environment. An example of such research is the visualisation of the Al Zubarah fortress in Qatar (Kersten et al. 2021), or The Imperial Cathedral (Kaiserdom) of Königslutter (Walmsley and Kersten 2020). The proposed methodology is valuable as a new form of spatial data transmission and can be used by museums or other cultural and educational institutions. The increasing availability of UAV technology makes it possible to register even the most endangered objects of cultural heritage. The 3D modelling carried out on the example of three objects located in Iraq, i.e. Rabban Hormizd Monastery, Taq Kasra, and the Great Mosque of Samarra, proved that it is possible even with the use of crowdsource data (Alsadik 2022). 3D models obtained with these methods constitute a valuable additional source of information about the building, particularly when no appropriate technical data or documentation exist, and a new source of information for both scientists (experts) and society (non-experts) (Blanco-Pons et al. 2019).

Apart from methods of obtaining data, one also needs to consider methods of visualizing spatial information in the effective and attractive way. The methods of popularizing historical and geographical research in scientific literature take advanced forms of presentation beyond classical scientific publications in journals or monographs (Medyńska-Gulij et al. 2021; Lorek 2016, 2021). Every year, interactive multimedia portals, websites and social platforms are gaining more and more popularity as platforms transmitting scientific knowledge (Maiellaro and Varasano 2017). The combination of traditional methods of visualizing research results with a visual-narrative storytelling is recognized in broadly understood science as an interesting form of an interdisciplinary approach for scientific publications (Roth 2021). YouTube website is one of the most important pillars of modern culture, which plays many functions, including educational, informative or promotional. Publishing research results in such an important medium is a form of engaging more users to study a given research topic and a form of presenting research results in an international platform.

Virtual content may take different multimedia forms, such as: photos, videos, audio, 3D models. The use of 3D models, which makes it possible to demonstrate historical objects or entire spatial structures in the new dimension, e.g. Virtual Reality (Halik and Smaczyński 2018), deserves particular attention. Such presentation method enhances historical value of the source, highlights its crucial elements, and complements descriptive sources. Reconstructed historical space, combined with 3D technology, allows one to demonstrate the previously unknown spatial relations with the close environment (i.e. relief, land cover, other objects) (MacEachren 1994).

In the research the authors used archival maps, the unique source of information about the environment and landscape, which presents individual elements of space and their traits that frequently may be classified according to the measurement levels (Bertin 1983). Such measurable assessment of the informative potential of the map determines the opportunities to use the map in further research (Lorek and Medyńska-Gulij 2020). Furthermore, the specificity of this depiction of historical information allows one to construe individual objects in the wide context of the entire space (the layout of objects on the map, spatial relations). The suitable selection of maps in terms of their accuracy level (scale) or period of time they present makes it possible to recreate historical conditions of the environment and significant changes that were occurring at that time.

Historical maps, particularly those made in topographic scales, constitute a rich and useful source of spatial information about, among others, the occurrence of historical land use structures (Wästfelt 2020) or the layout of individual topographic objects. Retrospective study, supported by historical maps and up-to-date cartographic sources (Statuto et al. 2017) or information collected in the field, allows one to identify historical structures and forms (or their relics) in modern landscape, and reconstruct the no longer existing ones. Such data are usually used for local studies (Lorek 2011; Lorek et al. 2018) and confronted with other sources of spatial information. According to Wilson (2005), in some cases historical maps may be the unreliable source of knowledge in drawing conclusions on a large scale. Hence, according to the methodology of using historical maps in research, one needs to compare the information included in maps with other graphical or textual sources (Buczek 1996; Scharfe 1972). Researchers also have to learn the map’s characteristic and circumstances in which it was created, as in many cases that affected what was depicted on the map (Harley 1989; Edney 1993, 1996).

Wilson (2005) distinguished two attitudes that can be applied by researchers in collecting spatial information: the geohistorical and the geocomputational one. The geohistorical attitude is largely based on the analysis of archival data and previous studies. The geocomputational attitude is used mostly by computers for spatial data analysis or process modelling (Geographic Information System) (Horbiński and Lorek 2020). Presently, combining both methods (Wästfelt 2020; Lorek and Horbiński 2020) for retrospective study is becoming a common practice (Karsvall 2013; Lorek 2021). It constitutes the optimum combination that makes use of both attitudes, indicating their complementary nature.

The compilation of the above research approaches allows contemporary researchers to distinguish the landscape features characteristic of a given period, often visible only in retrospect, e.g. pre-industrial landscape (cultural landscape). The concept of ‘landscape’ is understood by the authors as a comprehensive reflection of natural phenomena and human activity processes (Myga-Piątek 2001; Degórski 2005; Taylor 2013). In addition, the cartographic research methodology adopted in these studies allowed to refer also to the topographic landscape, which is defined by specific spatial objects recorded on the map (Lorek 2021). In this context, the windmill and its immediate surroundings (other topographic objects and relief) were considered. The above approach clarifies and refines the traditional definition of landscape to information about space that can be obtained from a topographic map.

The windmill is an example of topographic objects that were abundant in the past in this region. According to the research by Lorek (2021), the windmill belongs to a group of objects that formed the cultural landscape of Wielkopolska in the nineteenth and twentieth centuries. Over the next years, as a result of economic changes, these objects lost their importance and disappeared from the space. Currently, relatively few windmills are legally protected, despite the fact that they are important objects of cultural heritage of the region. Grass roots activities were undertaken by scientists, institutions and enthusiasts to protect the regional windmills, including compiling an inventory, collecting archival materials, and creating old windmill database (Mosakowski and Brykala 2019; Mosakowski et al. 2020).

2 Research Aim

In the article, the authors expand the retrospective study, adding the method of visualisation and the way of sharing the results. The main aim of this article is to develop and apply the methodology of obtaining, processing and publishing data for the purposes of documentation and promotion of cultural heritage (example of a windmill). Particular emphasis is placed on activities in the field of acquiring and processing detailed measurement data for the aim of creating a 3D model and a numerical terrain model (DEM). To achieve the set aim, the authors used low-level aerial photogrammetry, which allows obtaining detailed image data. Moreover, the authors’ aim was to present the data to the widest possible audience. For this aim, the authors used multimedia and the Internet, which allow for universal access for the public user around the world. Due to the geohistorical and geocomputational approach adopted by the researchers, the research undertaken is interdisciplinary. This means that to achieve the main goal, the scientific foundations developed in the field of geodesy, cartography, archeology, and geography, with particular emphasis on landscape research, will be used.

3 Study Area and Material Selection

Windmill from western Poland (Wielkopolska–Greater Poland) has been selected as a representative topographic structures of this region. The space presented, along with multiple windmills, created a landscape of Greater Poland characteristic of the 19th and the early twentieth century, the relics of which currently constitute a part of legally protected cultural heritage. Few of these objects have survived, the majority of them have not, and some have left traces indicating their location, which gives a foundation for the research on their reconstruction in the landscape.

The windmill in question was recorded on the Messtischblatt topographic map (German: Meßtischblätter), sheet number 3572 (Fig. 1), made at the scale of 1: 25,000. The process of creating this series of maps began in 1870s, and the area of Greater Poland was measured for the first time between 1888 and 1889 (Jankowska and Lisiewicz 1998). The sheets were measured with the plane table surveying method, with the employment of the dense triangulation network (Kraus 1969). Accurate measurements allowed one to use contour lines to present the relief. Even though the first edition of the map was not in colour, the manner of depicting the content made the map highly informative (Lorek and Medyńska-Gulij 2020). The blue colour was introduced to some sheets to highlight the network of surface waters (Baumgart 1944), and the brown colour—to mark the relief (Konias 2010). As far as the legend is concerned, the simplified version of it, containing a collection of the basic symbols used on the sheets of a given series, was placed in selected Messtischblatt sections in the field outside the frame (marginalia). Furthermore, a separate analysis with a collection of conventional symbols, along with descriptions that ordered the content according to several categories, such as waters, roads with rail and borders, land cover etc., was also worked out.

Fig. 1
figure 1

Messtischblatt of 1911 (Sheet 3572), Source: http://maps.mapywig.org/m/German_maps/series/025K_TK25/3572_Witkowo_1911.jpg

The section of the map selected for the study depicted, in terms of administrative units, the area of the Witkowo commune and the surrounding territories located in the north–east part of the Greater Poland province (Fig. 2). The map was made in 1911 and, apart from receiving a number, it was named “Witkowo” after the largest settlement in the area presented. A fragment of the space depicted on the sheet constitutes an example of the agricultural land use in Greater Poland at the turn of the nineteenth and twentieth centuries when windmills were the typical elements of the landscape (Lorek 2009). Many of them have not survived until this day, however, the ones that have are the relics of historical, prevailing sectors of the local economy, and nowadays also the elements of cultural heritage. The existing windmills are diverse in terms of condition and level of preservation, hence, the researchers found motivation to examine the subject. The utilitarian goal was to preserve the historical structure with the employment of modern technologies of processing and visualising spatial data. The authors came to the conclusion that the post mill, which dominated in the area, would be the most representative structure. The post mill is situated in the village of Kamionka (the eastern part of the map sheet). The mill has been faithfully renovated and made available to tourists in 2014 (Fig. 3).

Fig. 2
figure 2

Location of the area analysed

Fig. 3
figure 3

(Source: Google Maps); B the windmill after the renovation, October 2020

A The windmill before the renovation, July 2012

In the first edition of the Meßtischblätter map two symbols were used to mark the post mill. In the short version of the legend included in some sheets both symbols were captioned collectively as “Windmühle” (Fig. 4). Only in the separate legend, published as a separate document, the signatures of the post mill were distinguished in terms of quality. In the commentary the symbol with the triangular basis was selected as the one denoting a post mill (Bock-Windmühle), and the other signature with the circular basis denoted a smock mill (Hollander Windmühle). The difference between these two types of mills included in the legend consisted in their construction. The name of a post mill originated from the fact that it was mounted on a wooden post, around which it could be turned to bring the sails into the wind. In the smock mill only the upper, topped with the roof part with the sails rotated, and the construction was mounted on the still body (Mosakowski et al. 2020).

Fig. 4
figure 4

The fragments of the Messtischblatt map of 1911 (Sheet 3572) (A the existing windmill in the village of Kamionka, B the no longer existing windmill on the route between the village of Folwark and Trzuskołoń) and legends (C)

4 Results

4.1 Gathering Low-Level Aerial Images

The authors decided to obtain spatial data by means of the close-range photogrammetry technique. The minor goal of the research was to create a cartometric 3D model of a selected windmill. Considering the size and geometry of the structure selected, it was determined that multirotor UAV will be the best method of obtaining images. Furthermore, to be able to accurately register the windmill from all sides and angles, and to work out its 3D model afterwards, the Structure-from-Motion (SfM) algorithm was employed (Westoby et al. 2012; Smaczyński and Horbiński, 2020). To obtain image data, the DJI Phantom Advanced+platform, equipped with a camera with the 1″ CMOS matrix and the 20 Mpx lens. Although the platform is equipped with the Global Navigation Satellite Systems (GNSS), the researchers decided that the aerotriangulation process of the images obtained would take place with the employment of GCPs laid out around the structure. Considering the size of the tested object, it was decided that four ground control points (GCPs)—points 1–4, would be used for the photogrammetric development. The location of the GCPs around the tested windmill is shown in Fig. 5.

Fig. 5
figure 5

Ground control points used in the aerotriangulation process

Such attitude allowed one to conduct the triangulation process of the images obtained as accurately as possible (Nex and Remondino 2014; Padró et al. 2019). Measuring GNSS RTK is one of the methods of measuring GCPs (Clapuyt et al. 2016). For the measurement the authors used the GNSS Trimble R10 model 2 receiver. Furthermore, during the measurement the corrections, made available by the reference station, were used. Processing the photogrammetric images obtained took place in the Agisoft Metashape Professional software (Fig. 6B). The attitude adopted by researchers allowed them to carry out the aerotriangulation process based on GCPs and obtain the RMSE value. This made it possible to define the deviation of the original tie-up points from the corresponding points calculated on the basis of the generated model. For calculating the RMSE, the following formula was used (Herrero-Tejedor et al. 2020; Smaczyński and Horbiński 2021):

Fig. 6
figure 6

A Image from the UAV, B image from the AgiSoft MetaShape Professional software, C image from the SkechUp

$$\mathrm{RMSE}=\pm \sqrt{\sum_{i=1}^{n}\frac{{\left({X}_{i, est}-{X}_{i, in}\right)}^{2}+{\left({Y}_{i, est}-{Y}_{i, in}\right)}^{2}+{\left({Z}_{i, est}-{Z}_{i, in}\right)}^{2}}{n},}$$

where

X/Y/Zi,in—the actual value of a particular coordinate.

X/Y/Zi,est—the estimated value of this coordinate.

On the basis of the obtained results, it was concluded that the ground control point whose total error of aerotriangulation process had the lowest value was point 3, and the point burdened with the greatest error—point 1 (Table 1).

Table 1 Calculations of error on the basis of the chosen ground control points

The value of mean square error of 1.3 cm (Table 1), calculated during the aerotriangulation process, made it possible to align the images obtained and was considered very good (Smaczyński and Horbiński 2020), which resulted in adopting the 3D model created for further research.

4.2 Manual 3D Reconstruction Based on UAV Point Cloud

In the next stage the authors used the manual visualisation technique during 3D modelling with faithful recreation of the current condition of the windmill, considering digital photogrammetry based on factual geometric data (Specht et al. 2016). The technique presented demonstrates methodological attitude that allows one to visualise the accurate digital representation of the structure in its current condition. Thanks to the SketchUp program the windmill was modelled, and main characteristics and properties of the building depicted were recreated (Fig. 6C). During the reconstruction of the windmill the 3D model from the automatic processing of the point cloud obtained from the UAV was used as a point of reference (Hwang and Chu 2016). The point cloud generated from UAV modelling included a lot of imprecision related to incomplete recreation of some windmill elements. Therefore, researchers decided to reconstruct the structure manually (Carvajal-Ramírez et al. 2016). Modelling in the SketchUp program, compared to automatic modelling, is characterized by the following advantages: easiness (intuitive environment), low time consumption (fast creation process), low work consumption (little work needed) (Cichociński et al. 2016). The disadvantages of the digital reconstruction can be observed at texture mapping for individual elements of the structure because of the large number of architectural elements (Singh et al. 2013). Referring to high elementary accuracy of the structure, the impossibility of a good reconstruction through the automation of the 3D model was highlighted. The choice of modelling method should depend on the purpose the visualisation is supposed to serve.

4.3 Editing the Film

The final result was the presentation of a three-dimensional model of the windmill in the form of a short film. The authors used the Camtasia 2020 program, thanks to which, with the previously collected data and prepared visualisation, a short history and spatial location of the windmill were presented. This form of presentation allows for a broader look at the historical and spatial relations of the 3D object with cartographic maps (Caquard and Cartwright 2014; Thöny et al. 2018; Medyńska-Gulij et al. 2021). The movie, which lasts 2 min and 11 s, can be divided into several sequences, each connected with each other by appropriate transitions. At the beginning, a sheet of the Messtischblatt topographic map was presented and the exact location of the visualised windmill was indicated. Then, with the use of free available programs for visualisation of point clouds, such as Cloud Compare or Fugro Viewer, and with the use of the free Potree cloud viewer, the profile and the elevation model of the terrain were presented (Fig. 7). During the analysis of the area presented in the video, it was noticed that the discussed windmill was located above the dominant elevation of the terrain. The median height of the area shown in the film was 107.32 m above sea level, while the median area under the windmill was 110.37 m above sea level. The last stage was to embed the modelled windmill on a topographic map sheet in three-dimensional space and an in-depth presentation of the study. The film was uploaded to the YouTube channel 1 March 2022: https://youtu.be/RLt_kjG3wOM.

Fig. 7
figure 7

A frame of the film showing the spatial and height profile of the windmill location

5 Discussion

The research presented in this article is a case study. It should be emphasized that the methodology of acquiring image data, activities related to the design of the photogrammetric raid as well as the establishment and measurement of the GCPs must each time be developed for a given object. High geometric accuracy and high quality of textures obtained during the research support protective activities with regard to cultural heritage objects. The methodology may differ depending on the type of windmill, and above all its location, size, and geometry (Smaczyński and Horbiński 2021). The authors see the possibility of using additional techniques in measurements, e.g. laser scanning, to increase the detail and accuracy of the studies (Brabyn and Mark 2011). Moreover, as noted by Brabyn and Mark (2011), data integration improves the ability to analyse and modify the point of view on a given problem. The new quality of spatial data becomes the basis for conducting innovative scientific research. Research by Herrero-Tejedor et al. (2020) proved that the combination of photogrammetric data and laser scanning (TLS) allows for the improvement of an image detail. In their work, the researchers attempted to develop a methodology for acquiring and processing photogrammetric data and laser scanning during the registration of the atomic garden in Finca El Encı’n (Madrid). The study leads to a conclusion that compared to other techniques (e.g. terrestrial photogrammetry), laser scanning made it possible to increase the level of image detail especially when registering leaves, branches and small plants. In the study in question, TLS technology would enable more accurate mapping of small elements of the windmill structure, i.e. structure of the propellers.

The integration of measurement data obtained with the use of several measurement techniques may be of particular value. Data acquired with the use of various techniques and measurement methods can be particularly valuable when preparing detailed measurement documentation for a given object. In practice, such documentation allows to register the condition of a given object and enables the protection of other cultural heritage assets. In addition, these data can be used to develop advanced geovisualisations presented as short films that can be used for educational purposes and promotion of tourist products. Films of a historical and geographic nature posted on video sharing platforms and social media are an interesting and open form of sharing information with a wider audience (Beautemps and Bresges 2021). The short film was created to show the activities related to the reconstruction of the object in a fragment of the former space. The specificity of the film development was related to archival and contemporary source materials and the models generated on their basis (terrain model, object model). Based on the example of one windmill, a method of landscape reconstruction was proposed using multimedia by embedding a 3d model of the object on an old map into the location marked over 100 years ago. With the use of animation, in the final sequences of the film the old object is presented from different perspectives (bird’s eye view) at various levels of detail (scale) (Stams 2001; Lorek 2016). This form of space presentation can be developed using other tools for editing film shots and allows to create and compile multiple layers of time by selecting archival maps.

Archival maps play an important role in the reconstruction of the former state of space (topographic landscape) because they enable the consideration of both individual objects and entire structures, such as road networks or distribution of buildings (Stams 2001). The specificity of the presented research is related in particular to the operation of combining old and current spatial data. Research with the use of archival source materials requires knowledge of the circumstances of the creation of old maps to correctly interpret the content in the context of considering the ongoing changes. The old maps show not only the structure of the landscape, but also reflect the standards and relations in the social and cultural sphere in particular periods of time (Edney 1993; Kent 2009). In the presented research, the reference point was the topographic map from the beginning of the twentieth century which recorded the features of the old landscape and changes resulting from the industrialization and urbanization processes in this part of Europe (Osterhammel 2013; Bürgi et al. 2017). In the conducted research, a sheet that belongs to the publishing series (Messtischblätter) was used, which is important because it enables the comparison and verification of the localization of individual objects between sheets in different areas (Lorek and Horbiński 2020).

A significant part of the research is based on the development and use of technical issues related to the acquisition, processing and presentation of spatial data. The presented research shows that these issues constitute a plane for the integration of various disciplines and issues. The issue of a historical object (a former windmill), raised in the context of the landscape, with the use of cartographic methodology, determines the interdisciplinary nature of the research. Historical aspects of the pre-industrial state and its transformations were combined as well as a comprehensive approach to the former geographical space. The obtained results also constitute a proposal for the protection and conservation of valuable objects of cultural heritage. The process carried out and the results obtained show the great value of interdisciplinary work, especially in relation to landscape research. Working on the border of various disciplines allows you to create new solutions and provide new information, which would not be possible without their integration and the flow of information between them. The adopted solutions simultaneously develop the workshop of a historian or geographer with new possibilities of researching and presenting changes taking place in space. On the other hand the created products, such as 3D models or multimedia reconstructions, can be used in activities in the field of inventory, preservation or reconstruction of valuable monuments.

In addition to maps, one can also use other archival types of source materials that can expand the ways of recreating and visualizing the old space (Dragan 2016; Herrero-Tejedor et al. 2020; Krejčí and Cajthaml 2022). According to the classification by Lorek (2021), graphic materials such as perspective views, drawings, photographs, postcards, plans, and sketches, can be particularly valuable in the above-mentioned activities. They contain information relating to space and can support activities in the reconstruction of cultural landscapes. Archival topographic maps often make a very good platform for integrating these materials thanks to similar level of accuracy (Lorek 2021; Krejčí and Cajthaml 2022).

6 Conclusion and Future Work

Highlighting as characteristic landscape element as windmills in the landscape of the region is supposed to raise society’s awareness in that field and indicate the necessity to undertake some action to protect and conserve similar forms by local authorities. It is also worth paying attention to the symbolic aspect of the historic mill in the context of the expansion of renewable energy. The objects that used the force of the wind, which used to be common in the region, are now returning in the form of wind turbines. Due to their size, there is a significant interference with the existing landscape. It is related to the creation of a new order, which is significantly different from the traditional image of agricultural rural areas. Touching upon this topic may act as an incentive to the owners of the still existing structures and result in suitable restoration works and recreation of the no longer existing structures. Hence, it is important to keep documentation (photos, surveying data) that will provide data on the current condition of the structure and can be used for renovation and reconstruction. Furthermore, suitably preserved structures can be considered as tourist products and recognisable landmarks of the region. On the examples of other European regions one can observe a correlation between protection and promotion of cultural heritage, and the increase in society’s awareness in the field, which results in undertaking new initiatives regarding tourist development of the region (Sahahiri et al. 2019).

The authors observe potential in the use of 3D models also for the purposes of education. Compared to 2D maps, 3D maps allow one to employ recreated models of the real-life structures on maps instead of using their symbolic pictures (signatures). Such a method of presentation makes it easier for users to get oriented in space (Nurminen and Oulasvirta 2008). For young people it may be an effective and attractive way of passing on knowledge of historical landscapes or spatial relations. Adding extra information, related to modelling the past, future, or unavailable places, to geovisualisations (Cöltekin et al. 2019) may constitute an extra asset that boosts the interest in the topic presented.

The proposed methodology of the procedure allows to combine spatial information generated at different levels of detail. The data obtained from topographic maps ensure visibility of the obtained results, which is particularly important in the reconstruction and research of cultural heritage objects (for instance windmills in the Wielkopolska landscape of the nineteenth century and the beginning of the twentieth century). Data at this level of detail allow for the consideration of the land use structure and spatial relations between individual landscape elements. Moreover, topographic maps are also a source of information about landform features (Lorek 2016). In turn, data obtained and processed with photogrammetry technique used in geodesy show high geometric accuracy (shapes, dimensions) of the generated forms (construction of a 3D model of the windmill). The compilation of the above heterogeneous data and the developed method of their integration increase the value and credibility of the obtained results which can finally be presented to a wide audience in the form of short films posted on the Internet.

The obtained results make it possible to expand the developed methodology and its adaptation to other types of valuable cultural objects, such as churches, castles or palaces. Moreover, the authors consider increasing the research area around the tested object. This will allow for a more accurate characterization of the terrain topography, which can be an important source of data in research on determining spatial relationships and the impact of the tested object on the surrounding space.