Explore chapters and articles related to this topic
Infrastructure Monitoring and Sustainable Building and Construction in Civil
Published in Parveen Berwal, Jagjit Singh Dhatterwal, Kuldeep Singh Kaswan, Shashi Kant, Computer Applications in Engineering and Management, 2022
Parveen Berwal, Jagjit Singh Dhatterwal, Kuldeep Singh Kaswan, Shashi Kant
The estimated spatial resolution can be retrieved from the EXIF file, but the operating length should be specified. Usually, it’s not always possible or feasible to measure the camera light intensity. The rebuilt 3D cloud and camera positions from the Motion Structure (SfM) issue are then employed to determine the tasks are handled and the focal length. To achieve this purpose, many photos of the item overlapped from various viewpoints are first collected. The SfM technique helps to increase a sparse 3D cetane number and view characteristics from multi-viewpoint views concurrently. This investigation is carried out using the SfM system established by Snavely et al. (2006). In this approach, in each image SIFT key dots are recognized and all pictures are compared. To eliminate outliers, the RANSAC algorithm is utilized. This combination is intended for recovering focal length, center, alignment, and radial lens distortions characteristics (two components are calculated for each view and corresponds to a 4th order construct for frequency region) and 3D scene structure. The capability of transferring is dubbed this great optimization procedure [169].
Wheel hop estimation on rough roads
Published in Maksym Spiryagin, Timothy Gordon, Colin Cole, Tim McSweeney, The Dynamics of Vehicles on Roads and Tracks, 2018
H.A. Hamersma, T.R. Botha, P.S. Els
The RANdom SAmple Consesus (RANSAC) algorithm (Fischler and Bolles, 1981) was used to fit a plane to the points of the reference surface. The RANSAC algorithm is an iterative procedure to estimate the parameters of a model from a set of data which is corrupted by outliers. Upon completion of the iterative process, the model with the highest number of inliers is used to fit the plane in a least squares sense. This algorithm thus allows the model to be based on average value of inliers only. The equation of the plane is given in Equation 9: () ax+by+cz+d=0
Feature Extraction from LiDAR Data in Urban Areas
Published in Jie Shan, Charles K. Toth, Topographic Laser Ranging and Scanning, 2017
The RANSAC algorithm introduced by Fischler and Bolles is a general robust approach to estimate model parameters [3]. Instead of using the largest amount of data to obtain an initial solution and then attempting to eliminate the invalid data points, RANSAC uses the smallest feasible data set and enlarges this set with consistent data when possible. With applications to roof facet detection, classical RANSAC would be formulated as follows (cf. Algorithm 14.1): (i) randomly select a set of N ∈ ℕ planar surfaces ℘ within a LiDAR point cloud S and keep a count of the number of points (also called supports) of Euclidean distances from the associated planes ℘ less than a critical distance dcr. (ii). A least square estimation of the final plane (℘final) is performed with the set of supports (ℳk) of maximum cardinal. (iii) The set ℳk is then removed from the initial point cloud S. The algorithm runs until card(S) < 3, where card(S) is the cardinal of set S.
Automatic Detection of Surface Damage in Round Brick Chimneys by Finite Plane Modelling from Terrestrial Laser Scanning Point Clouds. Case Study of Bragança Dukes’ Palace, Guimarães, Portugal
Published in International Journal of Architectural Heritage, 2023
Jesús Balado, Lucía Díaz-Vilariño, Miguel Azenha, Paulo B. Lourenço
The modelling of a building can be performed manually, although it is a tedious task and the tendency is to automate the process as much as possible. The most common modelling method is through RANSAC and its variants (Núñez Andrés et al. 2012). In Barazzetti (2016), the point cloud is converted to a NURB network and then modelled on surfaces. In Hess et al. (2015), structure from motion (i.e. the process of estimating the 3-D structure of a scene from a set of 2-D images) is used to locate the main constructive elements. Intersections between elements are also important, as they give coherence to the final model (Lemos 2007). Many times, manual support of the modelling process is necessary. In Rodríguez-Moreno et al. (2018), two modelling methods are proposed, in which the non-automatic consists of generating polylines manually and the automatic generates a mesh. In Hess et al. (2015), modelling is based on line detection and subsequent manual correction. In the case of complex structures, it is also possible to model by zones and then integrate all data (Pierdicca et al. 2016). Generated models can also be updated. In Aoki et al. (2005), parameters are extracted by inverse eigen-sensitivity method to update a chimney model. Conventional chimneys are not usually elements that are represented in most levels of building detail, following the CityGML standard (Dore and Murphy 2012). Their small size makes them difficult to acquire and correct modelling of roofs is normally prioritized (Huang, Brenner, and Sester 2011; Lafarge and Mallet 2012; Sampath and Shan 2010).
An automatic calibration algorithm for endoscopic structured light sensors in cylindrical environment
Published in Nondestructive Testing and Evaluation, 2023
Mohand Alzuhiri, Zi Li, Jiaoyang Li, Adithya Rao, Yiming Deng
The results of the sensor calibration for five rings with the conventional calibration process are shown in Table 5. The table shows the parameters of five rings with projection angles that range from 37.8 to 57.7 degrees. The camera is assumed to be at the origin, and the projector is placed approximately 60 mm behind the camera. The table also shows that the values of increase with the increase of the ring number, which is consistent with the projection pattern design, where ring number 1 refers to the smallest ring in the pattern with the smallest projection angle. The slight shift in the z-component of the vertex () is related to the refraction experienced by the projected cone, which increases with the decrease in the projection angle. The results in Table 5 show inconsistent changes with abrupt increments and decrements in the absolute value of , which is assumed to be decreasing with the increase of the projection angle. One of the reasons that affect the calibration accuracy is the existence of outliers; therefore, random sample consensus (RANSAC) was applied. RANSAC is an iterative method that estimates the model parameters in the existence of outliers by separating them from inliers with repeated random sub-sampling [38]. The results after applying RANSAC are shown in Table 6. The results show a continuous reduction in the absolute value of that is associated with the increase of the projected angle, which is consistent with the effect of refraction.
Machine learning based anomaly detection and diagnosis method of spinning equipment driven by spectrogram data
Published in The Journal of The Textile Institute, 2022
Chen Shen, Bing Chen, Lianqing Yu, Fei Fan
Random sample consensus (RANSAC) is a re-sampling modeling method commonly used in image data processing. It generates candidate solutions by using the minimum number of data points to estimate the underlying model parameters. Considering the data characteristics of WSG and the requirements of abnormality detection, a method of reducing the dimensionality based on this idea is proposed as a benchmark. RANSAC has a non-linear conversion unit, which is more refined. However, since the selected dimension needs training, the expression ability is limited by the versatility of the training sample.