Search

High-resolution remote sensing imagery plays a strategic role in critical applications such as water resource management, water quality monitoring, and emergency responses to natural disasters for the quick and accurate identification and extraction of water bodies. However, traditional water body extraction methods present various challenges, particularly in the selection of image texture and characteristic features. In this study, a methodology is proposed that combines geometric features extracted from point cloud data with spectral information obtained from aerial photographs to more effectively define and extract the boundaries of water bodies. The geometric features generated from three-dimensional (3D) structure tensors are merged with the spectral information produced by the sensor system, and the well-known Random Forest (RF) classifier suitable for high-dimensional data, speed, and resistance to overfitting is used for training in water body detection. The effectiveness of the methodology developed in Matlab has been tested over four different locations in Turkey with varying topographic and vegetative characteristics. When the accuracy analysis of the detected water body boundaries is evaluated through the F-Score, the following were obtained: 85.7% for Study Area-1, 76.6% for Study Area-1 River, 93.7% for Study Area-2, 94.9% for Study Area-3, and 73.6% for Study Area-4. The study demonstrates that the presented methodology is applicable across different spatial scales and sensor types and carries potential for comprehensive uses in environmental and hydrological research. [ABSTRACT FROM AUTHOR]

With the development of technology, objects in the real world are transferred to the virtual environment, making it easier to manage all data flows and processes of the target object. The most important feature of digital twins is that they reflect the real world as accurately and up-to-date as possible. The use of LiDAR and photogrammetry has become very popular in the creation of three-dimensional point clouds of physical environments. Besides being able to represent the real world in detail, one of the most useful features of digital twins is that point clouds are suitable for direct use. In this way, point clouds can be used directly as a three-dimensional geometric model for a digital twin without the need for many processing steps. To update the three-dimensional geometric model, it is necessary to rescan the changed areas, and then perform change detection operations that provide a point cloud of these objects within the change. In this study, the importance and direct use of point clouds in the production of digital twins were examined, and for this purpose, photogrammetric studies were carried out on three sculptures with irregular geometrical structures and complex structures. As a result of the data obtained in the study, it was observed that the models were missing, especially in the parts where depth perception is difficult, and suggestions were made. As a result, it has been found that point clouds produced by SfM-based photogrammetric software provide a great advantage in fast and low-cost data collection to create and update the digital twin. [ABSTRACT FROM AUTHOR]

In recent years, the use of cold-formed steel (CFS) in low and medium-rise buildings has become widespread. CFS members have a high strength-to-weight ratio, and since the construction of the structures performed using these elements takes a short time, it offers an effective solution in terms of meeting the requirements of rapid construction. CFS construction has advantages as well as disadvantages, and one of these disadvantages is that the geometric imperfections that occur in the member during the manufacturing, transportation, and installation processes affect the element's behavior. This research focuses on accurately detecting and quantifying the geometric imperfections found in C-sectioned CFS members. Local and global imperfections in CFS members are determined using the improved automatic geometric imperfection detection and quantification method. The results obtained are compared with a previously developed, literature-based geometric imperfection detection and quantification method. As a result of this study, it is observed that the maximum and average geometric imperfection values calculated by the improved geometric imperfection detection and quantification method for all elements decreased by 50% or more, except for the geometric imperfections whose formulation remain same and which are not directly affected by the initial ideal geometric model placement process. It has been verified that the improved method accurately detects both local and global geometric imperfections. [ABSTRACT FROM AUTHOR]

It has been important in recent years to documention an underwater object as a three-dimensional and to transfer it to digital platforms. For this purpose, the photogrammetry method has been tried underwater and the term underwater photogrammetry has made ground in the literature. The most important reason for choosing a photogrammetry method is that it can product data in less time than traditional methods, and contribute positively in terms of time and cost. However, because divers that collect photographic data underwater can stay underwater for a limited duration, the photographing process is repeated and the process is prolonged. In order to minimize this disadvantage, data acquisition studies by videogrammetry method have started to be examined. In this study, a comparative analysis of three-dimensional model creation was made by placing a test object in a pool and using both photography and videogrammetry methods. As a result of the study, the photography method was accepted as a reference to test the accuracy of the video method. The accuracy study was carried out by comparing the three-dimensional point clouds acquired from both data sets, and the mean square error was determined to be 3.24 cm. The utility of the underwater video technique was explored in this study, and it was established that the video method was adequate in terms of accuracy but unsatisfactory visually. [ABSTRACT FROM AUTHOR]

The aim of the study was to investigate and implement an augmented reality-based platform and appropriate digital content for personalized non-formal and free learning of architectural and cultural heritage, which has lost its physical or structural integrity. It was aimed to raise awareness about cultural heritage and its appreciation by future generations. Developing a digital memory and making cultural heritage accessible with the help of new technologies were the main concerns of the study. In this respect, a study on a mobile, geolocation-based augmented reality application, which have simultaneous localization and mapping function, was conducted. The Temple of Augustus, which is an important archaeological site in Ankara, was chosen as a sample structure for the study. A digital 3D model of the building was prepared. This model was superimposed and aligned with the point cloud produced by the augmented reality platform with the help of photographs taken from the ruins. This overlap served as a marking function, so that even if the angle of the camera used by the application on mobile devices changes or the structure leaves the field of view and re-enters, the digital model placed on the site can be viewed at the right angle, with scaled size and material mapping. In this study, the possibilities of perceiving the geometric and semantic knowledge of the building as a whole, and experiencing that digital content interactively were investigated. [ABSTRACT FROM AUTHOR]

As 3-dimensional point cloud can be obtained easily with laser scanning technology, three-dimensional face recognition have become popular against the facial recognition performed using 2D images that has many limitations. In this study, the facial data of 10 people were modeled in 3D using a laser scanner. A total of 30 point clouds were taken from 10 people-two natural facial expressions and one laughing facial expression. The algorithm consists of three steps. In the first step, 3D points are defined on the point clouds using ISS and LSP methods. In the second step, key points were described using PFH and FPFH methods to obtain feature histogram. In the third step, the key points in different point clouds were matched using the feature histograms via the Kullbeck-Leiber Divergence method. For accuracy analysis, point clouds are registered with Iterative Closest Point (ICP) method. For accuracy assessment, the Euclidean distance between the matching points was calculated. The ISS algorithm finds about 25% less points than the LSP algorithm. The correct matching rate for PFH is up to 60%, while FPFH histograms are around 25%-30%. [ABSTRACT FROM AUTHOR]

Making inventories of cultural heritages is the most important issue for protecting and transferring of these historical artefacts from past to the future. Creation of these inventories is quite crucial for restoration of worn artefacts and recovering artefacts, which have lost attributes over time, with collected data. During the creation of these inventories, documentation of these artefacts should be done without damaging them. Laser scanning measurement system has been widely used in the documentation of cultural heritage with the use of state-ofart- technology in this field. Laser scanning is a technology that can collect three-dimensional location data quickly, reliable and detailed for many different applications. Today, laser scanners are an important research topic in especially photogrammetry and remote sensing study areas. In this study, modeling facilities have been exposed with terrestrial laser scanner technology to make three dimensional inventories of historic artefacts. In this context, the filtering effects on the models have been investigated. For this purpose, two different datasets collected with different scanners has been used. The first dataset belongs to Church of Tavra Creek, which is located in Sivas Temeltepe. Second dataset is a point cloud which belongs to stone building in Mersin. Within the scope of this study, coloring point cloud, registration, filtering, surface modeling, and deviation analysis have been performed on data sets, respectively. Noise reduction and smoothing filters have been used to filter data. Comparisons of the obtained models have been utilized by deviation analysis. Accordingly, the effects of filtering methods on the creation of historical inventories have been examined. [ABSTRACT FROM AUTHOR]

Keypoints which is detected automatically from images enable conjugate points creation between photogrammetric images, and dense point cloud can be generated by proceeding the photogrammetric process. The dense point cloud data includes many spatial information related to imaging area. On the other hand topographic changes can be detected by comparing two periods of point clouds. In this study urban changes in Mevlana region of Konya city was visualized by comparing three periods of point clouds belong the year 1951, 1975 and 2010. The urban changes were estimated with the vertical distances between compared point clouds. In addition, orthophoto images were created for analysing the related changes. [ABSTRACT FROM AUTHOR]

In recent years, studies on autonomous vehicles have been progressively increasing and especially, intensive works have been carried out in many areas like obstacle detection, mapping, localization, path planning, navigation, and obstacle avoidance. In this study, it was aimed to develop a new 3D LIDAR system with much lower cost, portable and 360-degree field of view by using standard 2D laser scanner instead of high-costing and large dimension 3D LIDARs which contribute seriously to the autonomy of the vehicle. The acquisition of a new 3D LIDAR system is mainly based on the rotation of the 2D laser scanner about certain axis and the conversion of the distance data published by this laser scanner into point cloud data, which is called as the 3D Cartesian point cluster. When the 3D LIDAR system is examined as a whole, it needs to communicate in coordinately the components it contains in order to operate at the desired efficiency. Therefore, in order to manage this software and hardware communication, ROS (Robot Operating System), which is a software platform that can be controlled by the computer, was used. In addition, a mobile vehicle was designed to be gained motion feature to the system. The 3D LIDAR platform was integrated into the designed mobile vehicle and the new system was made ready for navigation. For navigation, a number of algorithm and ROS nodes, called Navigation Stack, were used to autonomously move the vehicle from one point to another and avoid all obstacles that may occur during movement. Experiments show that the developed 3D LIDAR system provides an economical and effective solution for all directional object detection in autonomous vehicles. The navigation tests also show that the developed system is successful in performing tasks such as obstacle detection and obstacle avoidance in indoor environments. [ABSTRACT FROM AUTHOR]

Rockfalls are one of the geological-based natural disasters and it develops in the control of factors such as lithology, discontinuities, slope, decomposition, roughness and vegetation. There are generally 2 and 3 dimensional solutions for the modeling of rock falls in engineering projects. However, 3D solutions have been used frequently in recent years because they simulate the natural environment more precisely. With the widespread use of unmanned aerial vehicles, very valuable data can be collected about the location and block size of discontinuities on steep slopes, which are very difficult to access as well as topography and vegetation creating a point cloud with orthophotos. In this study, potential rock drops in Kasımlar Village of Sütçüler District in Isparta Province were modeled as 3D in Geographic Information Systems with photogrammetric methods using Unmanned Aerial Vehicle. At the same time, the trajectories of falling blocks belongs to possible rockfalls, trajectory, passing heights and energies are calculated and areas under risk are mapped. According to the results of 3D modeling for four blocks ranging in size from 26 m3 to 355 m3, the maximum kinetic energy is 55170kJ, the maximum passing heightis 11.5 m and the maximum simulated velocity are 24.7m/s.. [ABSTRACT FROM AUTHOR]

Four different variants of building detection are presented. Each variant has a different workflow and is capable of detecting buildings. The first variant of building detection is based on multispectral classification and DSM filtering. In the second variant, DSM blobs, mainly consisting of buildings and trees, are detected by subtraction of the DTM from the DSM. Then, trees are eliminated using NDVI data, derived from unsupervised ISODATA classification of the multispectral images, while small non-building objects are rejected based on size criteria. The third variant uses the planimetric density of raw LIDAR DTM data to detect the above-ground objects. The fourth variant is like the third one, but uses the vertical density of the raw LIDAR data (all points) to distinguish trees and buildings. To improve the results, a combination of the four variants using set intersections and unions is performed. The combination was empirical, with consideration of the datasets used in each variant and the advantages and disadvantages of each variant. In the evaluation, the combination of the four individual results yields 94% correct detections and an omission error of 12% for Zurich airport dataset. [ABSTRACT FROM AUTHOR]

With image processing techniques the improved or different images can be obtained and pattern recognition processes can be done using digital images. These operations begin from capturing the image and continue using the different techniques. Image processing which includes wide area is an important subject of computer science. Moreover, it is one of the significant working areas of reverse engineering. In this paper, the information regarding three different systems which have been developed to obtain 3D point clouds for digital images is given. The algorithm of the system in which point clouds can be obtained using image processing techniques is explained and samples are introduced. 3D point clouds can be obtained from the digital camera or hand held camcorder images which are different shapes with the system which solved the problem. These point clouds can easily be converted into three dimensional solid or surface model in a Computer Aided Design (CAD) program. The system and the other systems are compared and advantages of the system are specified. [ABSTRACT FROM AUTHOR]