The continuous monitoring of tall industrial buildings is necessary to ensure safe operation. With technological advances in terrestrial laser scanning and other non-contact measurement methods, the methods and techniques for assessing the stability of tall industrial chimneys are evolving. This paper presents a method for determining the non-verticality and straightness of chimneys that offers significant advantages over existing methods. Narrow bands of scanned point clouds are processed at selected height intervals. Using the RANSAC method, points that do not belong to the chimney shell are filtered and the centre of the circle or ellipse is adjusted using the least squares method. The proposed method enables the efficient filtering of point clouds due to frequent obstructions on the chimney shell, the determination of the regularity of the chimney shell shape, a mathematical analysis of the chimney axis curvature, and an intuitive graphical representation of chimney non-verticality. The comparison of the results with other studies confirms the efficiency of the method.

This paper presents a photogrammetry-based system for capturing turbulent aerated flow topography in a laboratory environment, especially for complex hydraulic phenomena character-ised by turbulent, non-stationary, and non-homogeneous aerated flows. It consists of ten high-resolution cameras equipped with monochromatic sensors and custom-built LED lights, all synchronised for accurate data acquisition. Post processing involves Structure-from-Motion and Multi-View Stereo techniques to calculate exterior and interior orientation parameters that ensure accurate alignment within a desired coordinate system, and conversion to point clouds. The proposed method showed great potential for capturing free water surface topography of turbulent aerated flows with high spatial and temporal resolution over the entire field of view of the cameras. Due to the unique capabilities of this system, direct comparisons with existing benchmarks were not possible. Instead, average free water surface profiles were derived from selected control cross sections, using 2D LIDAR measurements for verification. Both the LIDAR and photogrammetry averaged profiles showed remarkably good agreement, with deviations within ±20 mm. Validation showed that photogrammetry can be used to measure the complex aerated turbulent free water surface. In this way, this approach, involving consecutive image dataset acquisition at predefined intervals, is proving to be a valuable tool for observing, visualising, analysing, investigating, and gaining a comprehensive understanding of the dynamics of the free water surface.

Deformation analysis is a complex procedure where, based on several periodic geodetic measurements, displacements of points in the geodetic network are detected and determined. On this basis, movements and deformations of the built and natural environment are detected. The article discusses the Squared Msplit estimation, an extension of the maximum likelihood method, which is one of the procedures used in deformation analysis. The equations of the Squared Msplit estimation are derived and the method is presented on 2D geodetic network case study. The effectiveness of the presented method is compared with the results of other deformation analysis approaches performed with the same numerical example. The results obtained using the Squared Msplit estimation slightly differ from the simulated values, with the maximum discrepancy being 11.5 mm at unstable points and 10.4 mm at stable points, which are satisfactory results. The findings indicate that the Squared Msplit estimation provides results comparable to other methods. Therefore, it is considered suitable for deformation analysis and can be regarded as one of the applicable procedures in this field.

In this paper we analyse and compare two modern methods of deformation analysis: the squared Msplit method and the absolute Msplit method. The comparison is made on cases of simulated and actually measured horizontal geodetic networks. The squared Msplit method is based on the assumptions of the least square method, while the absolute Msplit method is based on the assumptions of the least absolute deviation method. For each method, we developed a computer programme and used it to calculate point displacements in both a simulated and a measured 2D geodetic network. The purpose of this article is to compare the displacements. The results of the two methods differ by up to about one centimetre in the simulated network and by less than two centimetres in the actually measured geodetic network.