2016
DOI: 10.1590/s1982-21702016000200011
|View full text |Cite
|
Sign up to set email alerts
|

Classificação De Nuvem De Pontos Laser Utilizando O Conceito De Análise De Componentes Principais E O Fator De Não Ambiguidade

Abstract: Resumo:Este artigo apresenta um método que realiza a classificação automática dos pontos amostrados por um sistema de varredura a LASER aerotransportado (SVLA). Nesse método são utilizados os autovalores da matriz de variâncias e covariâncias (MVC). Para o cálculo da MVC considerase uma vizinhança no entorno do ponto de interesse, a qual é determinada com base no conceito de entropia. A classificação é executada comparando os autovalores calculados, referentes a cada ponto e sua vizinhança, com os autovalores … Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1

Citation Types

0
3
0

Year Published

2018
2018
2018
2018

Publication Types

Select...
1

Relationship

1
0

Authors

Journals

citations
Cited by 1 publication
(3 citation statements)
references
References 8 publications
0
3
0
Order By: Relevance
“…In the first three cases, the LiDAR points were classified by the k-means method, considering three different groups of measures, as described in Subsection 2.3 and in Table 2. In the fourth case, the classification step was performed through discriminant analysis, similar to that used by Santos and Galo (2016), using theoretical eigenvalues related to eight different classes, as presented in Gross and Thoennessen (2006). For purposes of comparison, the eight classes were subdivided into two groups: edge and non-edge points.…”
Section: Resultsmentioning
confidence: 99%
See 2 more Smart Citations
“…In the first three cases, the LiDAR points were classified by the k-means method, considering three different groups of measures, as described in Subsection 2.3 and in Table 2. In the fourth case, the classification step was performed through discriminant analysis, similar to that used by Santos and Galo (2016), using theoretical eigenvalues related to eight different classes, as presented in Gross and Thoennessen (2006). For purposes of comparison, the eight classes were subdivided into two groups: edge and non-edge points.…”
Section: Resultsmentioning
confidence: 99%
“…( 1)( 2)In this paper, three different groups of measures were explored. The first is composed of eigenvalues (λ1, λ2, λ3), as performed by Gross and Thoennessen (2006), Yang and Dong (2013), and Santos and Galo (2016). The second is formed by linearity and planarity measures (Lλ, Pλ), and the last is composed only by linearity measure (Lλ).…”
Section: Selection Of Measures Using Principal Component Analysismentioning
confidence: 99%
See 1 more Smart Citation