Content uploaded by Carlos Cabo
Author content
All content in this area was uploaded by Carlos Cabo on Jun 30, 2017
Content may be subject to copyright.
A preview of the PDF is not available
Algorithms for automatic feature detection from Mobile Laser Scanning data
Abstract and Figures
Mass data collection systems, both based on images and LiDAR (Light Detection and Ranging), gather field information that can often be visually interpreted. However, this visual inspection is very time and resource intensive, to a point that frequently it is not worth using this technology. For this reason, there is an increasing trend towards the automatization of feature interpretation and identification processes, which are materialized through the creation of algorithms that perform these tasks from images and point clouds. The design of these algorithms consists of: (i) the formulation of accurate abstractions of the goal features (i.e. attributes and descriptors that allow their identification and separation from other features), and (ii) the computation or implementation, involving the transformation of the aforesaid abstractions into a set of computer instructions.
In the last few years, the use of terrestrial laser scanners have increased substantially, especially with respect to mobile terrestrial laser scanners. These systems gather hundreds of thousands of points per second on the objects and surfaces around them while the mobile platform is moving, and they overcome some of the disadvantages of the static terrestrial laser scanners and aerial LiDAR. Usually, mobile laser scanners provide point clouds with centimetric accuracy, with a highly variable distribution and point density (according to distances and incidence angles to the surfaces and objects). The point clouds are affected by the occlusions caused by the perspective obtained from a specific mobile point, which is usually close to the ground.
In this PhD thesis, two data structures for reducing the effects of the large amount of data and the heterogeneity of the point distribution are developed: (i) voxelization, and (ii) line clouds. These two structures allow the creation of reduced and homogenized versions of the point cloud, and have the advantage of being reversible. That is, the transformation of a point cloud into any of these structures produces a simplified version of the data on which the feature detection algorithms can be applied. Afterwards, the simplified structures (voxels and/or lines) are labelled, and the transformation is reversed in order to recover the original points. These points inherit the labels from the simplified structures.
Three algorithms for automatic feature detection are developed based on the two simplification and homogenization structures: (i) an algorithm for automatic detection of pole-like street furniture objects (e.g. traffic signs, traffic lights or lampposts), (ii) an algorithm for automatic surface detection (e.g. façades, walls or panels); and (iii) and algorithm for automatic delineation of road and street edges. For each one of the algorithms, the abstractions, implementation and validation tests are described and specified.
In the validation tests, the algorithm for automatic detection of pole-like objects is able to identify more than 92% of the goals in the test point clouds, with actual poles accounting for 84% of the detected features. For the second algorithm, the results show that more than 90% of the points belonging to each one of the 27 test surfaces are assigned to them, and 99.9% of the points assigned to each surface are correct. The third algorithm is able to delineate more than 97% of the test road surface, and 98% of the surface delineated by the algorithm belongs to the actual road. The three algorithms overcome many of the drawbacks of previous algorithms, and they surpass their performance in terms of detection and success rates.
Figures - uploaded by Carlos Cabo
Author content
All figure content in this area was uploaded by Carlos Cabo
Content may be subject to copyright.
ResearchGate has not been able to resolve any citations for this publication.
Accurate road asphalt extent delineation is needed for road and street planning, road maintenance, and road safety assessment. In this article, a new approach for automatic roadside delineation is developed based on the line clouds concept. The method relies on line cloud grouping from point cloud laser data. Using geometric criteria, the initial 3D LiDAR point data is structured in lines covering the road surface. These lines are then grouped according to a set of quasi-planar restriction rules. Road asphalt edge limits are extracted from the end points of lines belonging to these groups. Finally a two-stage smoothing procedure is applied to correct for edge occlusions and other anomalies. The method was tested on a 2.1 km stretch of road, and the results were checked using a RTK-GNSS measured dataset as ground truth. Correctness and completeness were 99% and 97%, respectively.
The detection of different road furniture such as curb, street floor and sidewalk from point clouds is important in many applications such as road maintenance and city planning. In this paper a pipeline for point cloud processing to detect the road curb from unorganized point clouds captured from a mobile terrestrial laser scanner is proposed. The proposed pipeline utilizes a covariance- based procedure to perform a 3D segmentation of point clouds. Features such as the road curb can be extracted by analyzing the local neighborhood of every point. This is done by computing the surface normal direction and the normalized eigenvalues. These parameters can be used to extract the ground objects, such as curb, street floor and sidewalk. The curb can be isolated from the rest of the ground objects based on the previous parameters in addition to elevation gradient within the local neighborhood. A 2D image processing scheme is also presented to find the curbs as edges in a generated 2D height image. The results show successful detection rates of 78% and 94% using 3D and 2D approaches respectively.
Mobile LiDAR technology is currently one of the attractive topics in the fields of remote sensing and laser scanning. Mobile LiDAR enables a rapid collection of enormous volumes of highly dense, irregularly distributed, accurate geo-referenced data, in the form of three-dimensional (3D) point clouds. This technology has been gaining popularity in the recognition of roads and road-scene objects. A thorough review of available literature is conducted to inform the advancements in mobile LiDAR technologies and their applications in road information inventory. The literature review starts with a brief overview of mobile LiDAR technology, including system components, direct geo-referencing, data error analysis and geometrical accuracy validation. Then, this review presents a more in-depth description of current mobile LiDAR studies on road information inventory, including the detection and extraction of road surfaces, small structures on the road surfaces and pole-like objects. Finally, the challenges and future trends are discussed. Our review demonstrates the great potential of mobile LiDAR technology in road information inventory.
Road safety inspection is currently carried out by time-consuming visual inspection. The latest mobile mapping systems provide an efficient technique for acquiring very dense point clouds along road corridors, so that automated procedures for recognizing and extracting structures can be developed. This paper presents a framework for structure recognition from mobile laser scanned point clouds. It starts with an initial rough classification into three larger categories: ground surface, objects on ground, and objects off ground. Based on a collection of characteristics of point cloud segments like size, shape, orientation and topological relationships, the objects on ground are assigned to more detailed classes such as traffic signs, trees, building walls and barriers. Two mobile laser scanning data sets acquired by different systems are tested with the recognition methods. Performance analyses of the test results are provided to demonstrate the applicability and limits of the methods. While poles are recognized for up to 86%, classification into further categories requires further work and integration with imagery.
The demand for 3D maps of cities and road networks is steadily increasing and mobile mapping systems are often the preferred geo-data acquisition method for capturing such scenes. Manual processing of point clouds is labour intensive and thus time consuming and expensive. This article focuses on the state of the art of automatic classification and 3D mapping of road objects from point clouds acquired by mobile mapping systems and considers the feasibility of exploiting scene knowledge to increase the robustness of classification. Copyright © 2017, G eomares Publishing, The Netherlands All rights reserved.
The negative impact of road accidents cannot be ignored in terms of the very sizeable social and economic loss. Road infrastructure has been identified as one of the main causes of the road accidents. They are required to be recorded, located, measured, and classified in order to schedule maintenance and identify the possible risk elements of the road. Toward this, an accurate knowledge of the road edges increases the reliability and precision of extracting other road features. We have developed an automated algorithm for extracting road edges from mobile laser scanning (MLS) data based on the parametric active contour or snake model. The algorithm involves several internal and external energy parameters that need to be analyzed in order to find their optimal values. In this paper, we present a detailed analysis of the snake energy parameters involved in our road edge extraction algorithm. Their optimal values enable us to automate the process of extracting edges from MLS data for tested road sections. We present a modified external energy in our algorithm and demonstrate its utility for extracting road edges from low and nonuniform point density datasets. A novel validation approach is presented, which provides a qualitative assessment of the extracted road edges based on direct comparisons with reference road edges. This approach provides an alternative to traditional road edge validation methodologies that are based on creating buffer zones around reference road edges and then computing quality measure values for the extracted edges. We tested our road edge extraction algorithm on datasets that were acquired using multiple MLS systems along various complex road sections. The successful extraction of road edges from these datasets validates the robustness of our algorithm for use in complex route corridor environments.