Open AccessFeature curve extraction from data points
Author(s) -
Sushant Gautam,
Vandana Agrawal
Publication year - 2021
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/1136/1/012004
Subject(s) - cluster analysis , intersection (aeronautics) , feature (linguistics) , plane (geometry) , line (geometry) , pattern recognition (psychology) , feature extraction , artificial intelligence , data point , point cloud , line segment , cluster (spacecraft) , computer science , object (grammar) , surface (topology) , mathematics , computer vision , geometry , geography , linguistics , philosophy , cartography , programming language
In the present work, study is done for the extraction of feature curves from data points lying on the surface of object or model. Here, the reconstruction of feature curves is proposed by intersection of plane pairs. These plane pairs approximate the adjacent regions of the feature. Feature of the object may be its edges, corners, holes, sudden jerks, slits etc. Points cloud is generated from the snapshots of the object taken from various viewing angles. Clustering work is extended from K-means clustering to combined K-means and Normal (attribute) approach based clustering as it utilizes the benefits of both K-means and attribute based method and segmented the data points so that a cluster is represented with similar normal vector. Planes are fitted to all clusters based on Least Square Plane Fitting (LSPF) method and line segments from their intersection are identified, highlighted and collected as feature lines i.e. edges and corners based on developed algorithm.