AUTOMATIC POWERLINE SCENE CLASSIFICATION AND RECONSTRUCTION
USING AIRBORNE LIDAR DATA
Gunho Sohn*, Yoonseok Jwa, Heungsik Brian Kim
GeoICT Lab, Geomatics Engineering Program, Earth and Space Science and Engineering Department, York University,
Toronto, Canada - (gsohn, yjwa, hskim,)@yorku.ca
KEY WORDS: Power Line Risk Management, 3D Modeling, Classification, Segmentation, Airborne LiDAR Data
ABSTRACT:
This study aims to introduce new methods for classifying key features (power lines, pylons, and buildings) comprising utility corridor scene using airborne LiDAR data and modelling power lines in 3D object space. The proposed approach starts from PL scene segmentation using Markov Random Field (MRF), which emphasizes on the roles of spatial context of linear and planar features as in a graphical model. The MRF classifier identifies power line features from linear features extracted from given corridor scenes. The non-power line objects are then investigated in a planar space to sub-classify them into building and non-building class. Based on the classification results, precise localization of individual pylons is conducted through investigating a prior knowledge of contextual relations between power line and pylon. Once the pylon localization is accomplished, a power line span is identified, within which power lines are modelled with catenary curve models in 3D. Once a local catenary curve model is established, this initial model progressively extends to capture entire power line points by adopting model hypothesis and verification. The model parameters are adjusted using a stochastic non-linear square method for producing 3D power line models. An evaluation of the proposed approach is performed over an urban PL corridor area that includes a complex PL scene.
its sag position drops down to the underlying terrain surface, 1. INTRODUCTION
which causes a mechanical failure of the PL system. To mitigate
An efficient and reliable monitoring of potential threats on this risk, PL models can be used for conducting thermal re-power line (PL) networks has become critical as electricity rating, which involves calculating its temperature, sag position, demands dramatically increase in today’s life and industry. A and tension under a variety of conditions. Third, as a PL PL network is comprised of conductors, insulators, pylons and structure wears out over time, its management should be carried other associated structures (spacer, dead lines, switch boxes and out in a timely manner to prevent electric failures. For example, etc.). The PL networks are often exposed to potential threats, as the corrosion of a pylon accelerates quickly after 30 years, which are mainly caused by encroaching vegetation, structural older pylons should be inspected regularly within a shorter time changes between “as-built” and “as-is” condition and violating period to identify their inelastic deformations (Smith and Hall, clearance between conductors and assets (Franken, 2003; Sohn 2011). PL models are used as for detecting structural anomalies and Ituen, 2010). To mitigate these risks, a timely and accurate of PL structures by examining movements in their positions monitoring of PL features and their spatial relations is urgently over time. This paper aims to introduce several new methods in required. Traditionally, this utility monitoring task has been classifying PL scenes, pylon localization and 3D PL modelling. achieved primarily relying on on-site manual inspection, which is expensive, time-consuming and often shows inaccurate 2. PREVIOUS RESEARCH results. To overcome this limitation, airborne LiDAR has been lately getting popularity as a primary information source to Despite its significance and demands to develop new methods, detect PL risks. This emerging data acquisition tool provides an there have been not much research efforts that are directly opportunity to classify a utility corridor scene more reliably and related to PL scene classification and modelling using airborne thus generate accurate 3D models of PL features due to LiDAR data. Baillard and Maître (1999) introduced Markov LiDAR’s ability of highly dense and multiple-echo data Random Field (MRF) classifier to identify ground, vegetation acquisition. Associating 3D models of PL features with PL risk and buildings from optical point clouds. Recently, Kim and monitoring tasks would provide several advantages against Sohn (2010) described a PL scene classifier using Random conventional manual on-site investigation. First, vegetation is Forests where a number of features were extracted in a point-dynamically growing towards PL conductors. It is critical to wise manner. The same authors (2011) proposed a Multiple timely identify and predict danger trees to PL structures. If PLs Classifier System (MCS) for classifying PL scenes, which fuses are modelled with high accuracy (a design accuracy of two Random Forests (RF) classifiers respectively trained from conductor’s sag position is in the order of 15 cm), these models two independent feature set (point- and object-based features). can be added to a regular monitoring of vegetation clearance. Calberg et al. (2009) developed a series of binary classifiers, For instance, having accurate PL models can determine each of which can filter out a particular class from unlabelled clearance areas that could be jeopardized by a vegetation LiDAR data remained by a previous classifier. Pylons can be encroachment by simulating a conductor blowout that shows its considered as a vertical object (tree trunk, traffic light lamp post, maximum designed sag and swing position as well as the etc). Although no research publication can be found in pylon potential growth rate of vegetation. Second, PL models provide detection, there has been increasing interests in detecting benefits to investigate violations of ground clearance. The sag vertical objects from ground-based ranging imagery. Lehtomäki positions of conductors vary with atmospheric conditions (i.e., et al. (2010) recognized street posts based on clustering of ambient temperature, snow/ice load, solar radiation and so on) isolated points on horizontal cross-section. Kim and Medioni and electric current flow. If a conductor’s temperature increases, (2011) utilized a liner property measured by analyzing the
英语 ISPRS journal of photogrammetry and remote sensing 国际摄影测量和遥感学会... Mat 英语 Annals of geophysics 地球物理学纪事 英语 英语 Minerals & ...
英语 ISPRS journal of photogrammetry and remote 国际摄影测量和遥感学会志英语 ...(英文版) Technology Beijing, Mineral, 英语 Metallurgy, Mat Annals of ...
英语 ISPRS journal of photogrammetry and remote sensing 国际摄影测量和遥感学会... Mat 英语 Annals of geophysics 地球物理学纪事 英语 英语 Minerals & ...
ISPRS Journal of Photogrammetry and Remote Sensing 64 (2009) 472–481 ...Annals IX Simposio Brasileiro de Sensoriamento Remoto. INPE. Santos, Brazil....
ANNALS OF MATHEMATICS ANNALS OF THE NEW YORK ...ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING ...167 168 169 JOURNAL OF PROTEOME RESEARCH JOURNAL ...
Isprs journal of photogrammetry and remote sensing ...annals of geophysics 0.413 pure and applied ...文档贡献者 kj99102hsl 贡献于2012-02-28 ...
2012年SCI杂志中科院分区表(完整版)_其它_高等教育...ISPRS J PHOTOGRAMM IEEE POWER ENERGY M CEMENT ...ANNALS OF BIOMED APPLIED SOFT INFORMATION & ...
ISPRS J PHOTOGRAMM ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING 1.6740...ANNALS OF GLACIOLOGY 地学 0260-3055 4 0.8020 2012 B SOC GEOL FR ...
ISPRS JOURNAL OF ISPRS J PHOTOGRAMMETRY AND REMOTE...ANNALS OF GLACIOLOGY ANN GLACIOL AUST J EARTH ...2012 0.9837 1546 430 挪威 400 荷兰 347 法国 ...
ISPRS J PHOTOGRAMM ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING 0924-...ANNALS OF THE NEW YORK ACADEMY OF SCIENCES 0077-8923 PHILOS PHILOSOPHICAL T...
热门文档