Points Classification by a Sequential Higher – Order Moments Statistical Analysis of Lidar Data

Abstract

ABSTRACT:The paper deals with a new sequential procedure to perform unsupervised LIDAR points classification by iteratively studying skewness and kurtosis for elevation and intensity point distribution values. After a preliminary local shape analysis of elevation and intensity point distributions, carried out from the original discrete frequencies by a non parametric estimation of the density functions, the procedure starts by choosing the category of data (elevation or intensity) to analyse at first: the choice falls on the category better showing by a testing procedure a bi or a multi clustering distribution. The first point cluster is identified by studying the distribution skewness and kurtosis variations, after removing at each step the largest data values. The selected cluster is furthermore analysed by studying higher order moments behaviour of the complementary data category. This makes possible to find out potential sub clusters of the original selected one, permitting, in this way, a more effective point classification. Successive clusters are identified by applying the same iterative procedure to the still unclassified LIDAR points. For complex point distribution shapes or for the classification of large areas, a progressive analysis method, based on the partition of the entire data set into regular subsets, is proposed. Some real numerical experiments confirm the capability of the method proposed. The classification total errors in the experiments range from a minimum value of 1,2% to a maximum value of 8,9%.

Classification of lidar bare-earth points, buildings, vegetation, and small objects based on region growing and angular classifier

“…They used statistical measures of distribution such as kurtosis and skewness to extract urban objects in scenes with varying terrain and coverage types. Similarly, Crosilla et al (2011) developed a new unsupervised segmentation algorithm based on iteratively studying skewness and kurtosis for elevation and intensity point distribution values. Finally, methods that combined segmentation and progressive densification (e.g.…”

Robust Locally Weighted Regression For Ground Surface Extraction In Mobile Laser Scanning 3D Data

“…Skewness balancing introduced by Bartles et al (2006) is mainly a segmentation algorithm based on the central limit theorem where the statistical measure Skewness is chosen to describe the characteristics of the point cloud distribution and has been used as a termination criterion in a segmentation algorithm. Later this algorithm has been developed combining with the Kurtosis measure by others (Crosilla et al 2011).…”

“…Akel et al (2007) propose an algorithm based on orthogonal polynomials for extracting terrain points from LiDAR data. We know higher order fits may lead to numerical instability (Fan and Gijbels 1996). Skewness balancing introduced by Bartles et al (2006) is mainly a segmentation algorithm based on the central limit theorem where the statistical measure Skewness is chosen to describe the characteristics of the point cloud distribution and has been used as a termination criterion in a segmentation algorithm.…”

“…The chosen family of functions can be overlyrestrictive for some types of data (Avery 2012). Fan and Gijbels (1996) show even a 4th-order polynomial fails to give visually satisfying fits. As an alternative, higher order fits may be attempted, but this may leads to numerical instability.…”

“…To meet the challenges associated with classification/ground filtering (separating ground points) methods have been introduced over the last two decades (e.g. Belton and Bae 2010, Brovelli et al 2004, Crosilla et al 2011, Lindenberger 1993, Kraus and Pfeifer 1998, Nasser et al 2005. A comparative analysis among different methods conducted by the ISPRS Working Group (WGIII/3; Sithole and Vosselman 2004) show that no method is sufficiently good for every dataset.…”