Orthophotomaps are now an irreplaceable source of topographic data acquisition, which also can be used in the preparation of navigational charts. As they are maps for special applications, they shall have a specified charting accuracy. The aim of this study is to evaluate the accuracy of the shoreline mapping using aerial photographs and satellite images. This evaluation was based on the statistical analysis related to the accuracy of the vectorization of the inland waters shoreline

Observational sensors, such as CCTV cameras and radars, are part of the RIS system. These sensors should provide information about the current situation on the waterway and to a greater extent – aquatory. Barriers for the observational sensor, in the sense of reducing the view field, are various objects in the aquatory. This is one of the main factors that make it difficult to determine the proper location of the sensor. Currently, in order to increase the efficiency of determining the locations of the observational sensors are used different tools for performing spatial analysis. This applies to both the initial geographical identification as well as detailed analysis. Analysis efficiency, in turn, depends on the model of the environment, represented in the form of the Digital Surface Model. Given the complexity of some geographic features, it should be used the appropriate methodology for the visibility analysis, as well as the preparation of the input material. The following work includes the theme of visibility analysis on inland waters, with special emphasis on the complex objects. They include bridges, cranes, buildings, high vegetation and other engineering structures

By applying digital model of the sea bottom it is possible to model information de-pending on dynamical phenomena, such as squat. This article presents a conception of data modeling , which allows to visualize safe depth area depending on ship speed. That kind of information can be implemented in numerical charts.

Przedstawiono wykorzystanie sztucznych sieci neuronowych do budowy modelu powierzchni dna. Budowę neuronowego modelu powierzchni dna przeprowadzono na przykładzie perceptronu wielowarstwowego. Możliwości rekonstrukcji powierzchni dna zilustrowano na przykładzie różnych typów danych i kształtów powierzchni. Podano również przykłady zastosowania neuronowego modelu dna.

Model powierzchni dna jako element numerycznej mapy morskiej powinien charakteryzować się dolnym odwzorowaniem kształtu. W pracy porównano w)>hrane metody interpolacji powierzchni w aspekcie zastosowania ich do budowy numerycznej powierzchni dna.

Shoreline mapping is one of the key stages in navigational charting. In terms of navigation, the shoreline marks the boundary of a river, which is often equivalent to the navigable water area. In cartographic terms, it is an important topological element between different objects that are adjacent to it. Currently, topographic objects are often mapped using photogrammetric materials obtained from various altitudes – satellite, airborne or low, which is associated with the use of an airborne UAV. Depending on the type of materials, the shoreline can be obtained in vector form with differing situational accuracy and differing degree of detail. In addition to the standard methods of processing vector data, the research in this paper also included the use of sonar images, enabling the detection of the shoreline with the use of a surveying hydrographic unit. On the basis of the collected photogrammetric and sonar images of different spatial resolution, an analysis of the accuracy of shoreline mapping was performed in terms of the situational accuracy and the level of detail in its representation. The results of the research provided the basis for the determination of dedicated remote sensing materials enabling the development of maps for inland navigation.

The shoreline is an important geographical zone, and knowledge of its accurate location can be crucial for coastal management and mapping. The ever-increasing number of aerial and satellite sensors is leading to research related to the development of new methods for the automatic extraction of the shoreline. Currently, there is a lot of research in this area with different research methodologies. In this paper, an analysis of shoreline extraction methods was carried out. Based on the analysis undertaken, current research processes in this field can be verified. This enabled the further evaluation of the research methodologies studied, including the identification of basic assessment elements for shoreline extraction accuracy. Practical aspects of this work include the ability to establish the correct methods to assess the accuracy of extracted shorelines for both research and production processes related to data extracted from remotely sensed images.

This article presents a comparative analysis of the creation of digital bathymetric models using multispectral bands. The case analyzed involves images acquired from low altitude using an unmanned aerial vehicle in an area of shallow inland waters. The use of an underwater photogrammetric network enables georeferencing of the datasets at a level no worse than 0.013 m (mean RMSE). The green and red bands make it possible to generate a dense point cloud in the full bottom area and create DBMs, the accuracy of which is examined on the basis of 22 bottom check points and the obtained accuracies for the RMS error are no worse than 0.24 m. The results of this study indicate that the green and red bands can be used for bathymetry acquisition in shallow and very shallow waters, which can be applied to work that requires accurate bathymetry reconstruction in these types of water bodies.