Planar Shape Detection and Regularization in Tandem

Abstract

We present a method for planar shape detection and regularization from raw point sets. The geometric modelling and processing of man-made environments from measurement data often relies upon robust detection of planar primitive shapes. In addition, the detection and reinforcement of regularities between planar parts is a means to increase resilience to missing or defect-laden data as well as to reduce the complexity of models and algorithms down the modelling pipeline. The main novelty behind our method is to perform detection and regularization in tandem. We first sample a sparse set of seeds uniformly on the input point set, and then perform in parallel shape detection through region growing, interleaved with regularization through detection and reinforcement of regular relationships (coplanar, parallel and orthogonal). In addition to addressing the end goal of regularization, such reinforcement also improves data fitting and provides guidance for clustering small parts into larger planar parts. We evaluate our approach against a wide range of inputs and under four criteria: geometric fidelity, coverage, regularity and running times. Our approach compares well with available implementations such as the efficient random sample consensus-based approach proposed by Schnabel and co-authors in 2007. In this work, we focus on the automated detection and regularization of primitive shapes from unorganized point clouds. Point cloud is the de facto standard data format for surface reconstruction

Incremental and Enhanced Scanline-Based Segmentation Method for Surface Reconstruction of Sparse LiDAR Data

“…These make the planar surface look foggy and thick. To regularize the point cloud, planar shapes are estimated using Region Growing or RANSAC and the points on the estimated planes are redirected using preset regularity relationships, such as parallel, orthogonal or coplanar [18,19]. These methods are mainly used to process dense and uniform point cloud.…”

Scan Profiles Based Method for Segmentation and Extraction of Planar Objects in Mobile Laser Scanning Point Clouds

“…They are more robust to noise and perform better than the edge based approaches (Nurunnabi et al, 2015). Oesau et al (2016) proposed a method for planar object detection for unorganized point clouds. Their method uses the K-nearest neighbour (KNN) for determining the neighbours for normal vectors estimation.…”

“…They can be divided into five groups: edge based (Huang & Menq, 2001), robust model fitting (Fischler & Bolles, 1981;Hough, 1962;Oesau et al, 2016;Vosselman & Maas, 2010), scan line based (Jiang & Bunke, 1994;Sithole & Vosselman, 2003) and region/surface growing based (Nurunnabi et al, 2012;Nurunnabi et al, 2015;Oesau et al, 2016;Rabbani, van den Heuvel, & Vosselmann, 2006;Vosselman & Maas, 2010).…”

Object Classification via Planar Abstraction

“…Related works differ greatly in the way they detect the planar shapes, depending on the defects in the input point data. Region growing is very efficient in point clouds structured as range images (Boulch et al, 2014, Holz and Behnke, 2012, Oesau et al, 2016, but are not suited to unstructured point clouds due to missing neighborhood linkage. The Hough transform (Hough, 1962, Davies, 2005, popular for detection of primitive shapes in images, is now commonly used for plane detection in point clouds.…”