China 
Africa 
Explore chapters and articles related to this topic
Registration for Super-Resolution: Theory, Algorithms, and Applications in Image and Mobile Video Enhancement
Published in Peyman Milanfar, Super-Resolution Imaging, 2017
Patrick Vandewalle, Luciano Sbaiz, Martin Vetterli
After image registration, we reconstruct a high-resolution image from the set of images using a nonuniform interpolation method implemented in Matlab [12]. Assuming the PSF is very narrow, and can be approximated by a Dirac, we compute the precise locations of all pixel coordinates on the high-resolution grid. Next, we perform a Delaunay triangulation using the Quickhull algorithm [1]. The high-resolution pixel values are then non-uniformly interpolated using bicubic interpolation. Such a reconstruction method provides good precision, with very low computational complexity. For more advanced reconstruction methods, we refer the reader to other chapters in this book.
Morphological analysis of nanoparticle agglomerates generated using DEM simulation
Published in Particulate Science and Technology, 2022
Another method to obtain the volume of agglomerate is convex hull method. It is defined as the smallest convex polygon enclosing any shape. The convex hull is the smallest polyhedral formed by joining the coordinates of the center of the particles around the boundary as displayed in Figure 3(c). “Divide and conquer based Quickhull algorithm” as implemented in MATLAB® is used to get the convex hull (Barber, Dobkin, and Huhdanpaa 1996). Volume of the convex hull thus obtained is used as the volume of agglomerate in Equation (6). The porosity of the agglomerate under study is found to be using this method. It can be noted that porosity obtained from radius of gyration is high as compared to that of convex hull method. This is because radius of gyration method gives large volume of agglomerate which leads to high porosity in comparison to convex hull method. Similar trend of difference in porosity with these methods are noted in Dadkhah, Peglow, and Tsotsas (2012).
Railway vehicle bearings risk monitoring based on normal region estimation for no-fault data situations
Published in Journal of Transportation Safety & Security, 2021
Yuan Zhang, Yong Qin, Yan-ping Du, Lei Zhu, Xiu-kun Wei
The Quickhull algorithm first selects the leftmost, rightmost, uppermost, and lowermost points and then must form a convex quadrilateral (or triangle). The points in this quadrangle must not be on the convex hull. The remaining points are then divided into four parts by the closest edge and repeated. Considering that in most cases, the convex hull consists only of some points in the point set, and the remaining points exist inside the convex hull, the Quickhull algorithm continuously produces areas that do not require research by detecting some special points (such as the furthest point). Thus, efficiency can be improved by excluding nonconvex hull vertices and reducing the number of analytical points, so its computational complexity is low (Barber et al., 1996; Jiang, 2002).