Development and virtual representation of 3D models of Cultural Heritage (CH) objects has triggered great interest over the past decade. The main reason for this is the rapid development in the fields of photogrammetry and remote sensing, laser scanning, and computer vision. The advantages of using 3D models for restoration, preservation, and documentation of valuable historical and architectural objects have been numerously demonstrated by scientists in the field. Moreover, 3D model visualization in virtual reality has been recognized as an efficient, fast, and easy way of representing a variety of objects worldwide for present-day users, who have stringent requirements and high expectations. However, the main focus of recent research is the visual, geometric, and textural characteristics of a single concrete object, while integration of large numbers of models with additional information—such as historical overview, detailed description, and location—are missing. Such integrated information can be beneficial, not only for tourism but also for accurate documentation. For that reason, we demonstrate in this paper an integration of high-resolution spherical panoramas, a variety of maps, GNSS, sound, video, and text information for representation of numerous cultural heritage objects. These are then displayed in a web-based portal with an intuitive interface. The users have the opportunity to choose freely from the provided information, and decide for themselves what is interesting to visit. Based on the created web application, we provide suggestions and guidelines for similar studies. We selected objects, which are located in Bulgaria—a country with thousands of years of history and cultural heritage dating back to ancient civilizations. The methods used in this research are applicable for any type of spherical or cylindrical images and can be easily followed and applied in various domains. After a visual and metric assessment of the panoramas and the evaluation of the web-portal, we conclude that this novel approach is a very effective, fast, informative, and accurate way to present, disseminate, and document cultural heritage objects.

Aerial or satellite images are conventionally used for geospatial data collection. However, unmanned aerial vehicles (UAVs) are emerging as a suitable technology for providing very high spatial and temporal resolution data at a low cost. This paper aims to show the potential of using UAVs for map creation and updating. The whole workflow is introduced in the paper, using a case study in Rwanda, where 954 images were collected with a DJI Phantom 2 Vision Plus quadcopter. An orthophoto covering 0.095 km² with a spatial resolution of 3.3 cm was produced and used to extract features with a sub-decimetre accuracy. Quantitative and qualitative control of the UAV data products were performed, indicating that the obtained accuracies comply to international standards. Moreover, possible problems and further perspectives were also discussed. The results demonstrate that UAVs provide promising opportunities to create high-resolution and highly accurate orthophotos, thus facilitating map creation and updating.

Within the past years, the development of high-quality Inertial Measurement Units (IMU) and GNSS technology and dedicated RTK (Real Time Kinematic) and PPK (Post-Processing Kinematic) solutions for UAVs promise accurate measurements of the exterior orientation (EO) parameters which allow to georeference the images. Whereas the positive impact of known precise GNSS coordinates of camera positions is already well studied, the influence of the angular observations have not been studied in depth so far. Challenges include accuracies of GNSS/IMU observations, excessive angular motion and time synchronization problems during the flight. Thus, this study assesses the final geometric accuracy using direct georeferencing with high-quality post-processed IMU/GNSS and PPK corrections. A comparison of different data processing scenarios including indirect georeferencing, integrated solutions as well as direct georeferencing provides guidance on the workability of UAV mapping approaches that require a high level of positional accuracy. In the current research the results show, that the use of the post-processed APX-15 GNSS and IMU data was particularly beneficial to enhance the image orientation quality. Horizontal accuracies within the pixel level (2.8 cm) could be achieved. However, it was also shown, that the angular EO parameters are still too inaccurate to be assigned with a high weight during the image orientation process. Furthermore, detailed investigations of the EO parameters unveil that systematic sensor misalignments and offsets of the image block can be reduced by the introduction of four GCPs. In this regard, the use of PPK corrections reduces the time consuming field work to measure high quantities of GCPs and makes large-scale UAV mapping a more feasible solution for practitioners that require high geometric accuracies.

There is a growing demand for cheap and fast cadastral mapping methods to face the challenge of 70% global unregistered land rights. As traditional on-site field surveying is time-consuming and labor intensive, imagery-based cadastral mapping has in recent years been advocated by fit-for-purpose (FFP) land administration. However, owing to the semantic gap between the high-level cadastral boundary concept and low-level visual cues in the imagery, improving the accuracy of automatic boundary delineation remains a major challenge. In this research, we use imageries acquired by Unmanned Aerial Vehicles (UAV) to explore the potential of deep Fully Convolutional Networks (FCNs) for cadastral boundary detection in urban and semi-urban areas. We test the performance of FCNs against other state-of-the-art techniques, including Multi-Resolution Segmentation (MRS) and Globalized Probability of Boundary (gPb) in two case study sites in Rwanda. Experimental results show that FCNs outperformed MRS and gPb in both study areas and achieved an average accuracy of 0.79 in precision, 0.37 in recall and 0.50 in F-score. In conclusion, FCNs are able to effectively extract cadastral boundaries, especially when a large proportion of cadastral boundaries are visible. This automated method could minimize manual digitization and reduce field work, thus facilitating the current cadastral mapping and updating practices.

Fit-for-purpose land administration aims to align cadastral policies, administration and technology selection with the prevailing societal needs and capacity within a country context. It seeks to support delivery of more rapid and low-cost cadastral boundary mapping – and ultimately more widespread land tenure security. The philosophy suggests that when high positional accuracies are not the primary concern, but when time pressure is, high-resolution satellite images can serve as a source for creating cadastral boundary information. This paper explores the potential of mean-shift segmentation plug-in in QGIS to semi-automatically extract cadastral boundaries in rural areas – based on the land cover information from WorldView-2 satellite images. The segmentation gives a vector file satisfying many cadastral boundary requirements and ready to be used in a GIS environment. The buffer overlay method was used to assess the quality of extracted boundaries. For a non-vegetated terrain having visible boundaries, the approach could be taken as an alternative to support existing, relatively slow, boundary mapping approaches by minimizing the effects of manual digitization and surveying.

Cadastral boundaries are often demarcated by objects that are visible in remote sensing imagery. Indirect surveying relies on the delineation of visible parcel boundaries from such images. Despite advances in automated detection and localization of objects from images, indirect surveying is rarely automated and relies on manual on-screen delineation. We have previously introduced a boundary delineation workflow, comprising image segmentation, boundary classification and interactive delineation that we applied on Unmanned Aerial Vehicle (UAV) data to delineate roads. In this study, we improve each of these steps. For image segmentation, we remove the need to reduce the image resolution and we limit over-segmentation by reducing the number of segment lines by 80% through filtering. For boundary classification, we show how Convolutional Neural Networks (CNN) can be used for boundary line classification, thereby eliminating the previous need for Random Forest (RF) feature generation and thus achieving 71% accuracy. For interactive delineation, we develop additional and more intuitive delineation functionalities that cover more application cases. We test our approach on more varied and larger data sets by applying it to UAV and aerial imagery of 0.02–0.25 m resolution from Kenya, Rwanda and Ethiopia. We show that it is more effective in terms of clicks and time compared to manual delineation for parcels surrounded by visible boundaries. Strongest advantages are obtained for rural scenes delineated from aerial imagery, where the delineation effort per parcel requires 38% less time and 80% fewer clicks compared to manual delineation.