4 / 16
E u r o S c i C o n J o i n t E v e n t o n
Laser Optics & Photonics and
Atomic & Plasma Science
American Journal of Computer Science and Information Technology
ISSN: 2349-3917
J u l y 1 6 - 1 7 , 2 0 1 8
P r a g u e , C z e c h R e p u b l i c
Page 20
Laser Optics & Photonics and Atomic & Plasma Science 2018
L
ight detection and ranging (LIDAR) presents a series of unique challenges, the
foremost of these being object identification. Because of the ease of aerial
collection and high range resolution, analysts are often faced with the challenge
of sorting through large datasets and making informed decisions across multiple
square miles of data. This problem has made automatic target detection in
LIDAR a priority. We propose a novel algorithm with the overall goal of automatic
identification of five object classes within aerially collected LIDAR data: ground,
buildings,vehicles,vegetationandpowerlines.Themainobjectiveofthisresearchis
addressed as two specific tasks viz. region segmentation andobject classification.
The segmentation portion of the algorithm uses a progressive morphological
filter to separate the ground points from the object points. The object points
are then examined and a Normal Octree Region Merging (NORM) segmentation
process is applied. This new segmentation technique, based on surface normal
similarities, subdivides the object points into clusters. Next, for each cluster of
object points, a Shape-based Eigen Local Feature (SELF) is computed. Finally, the
features are used as the input to a cascade of classifiers, where four individual
support vector machines (SVM) are trained to distinguish the object points into
the remaining four classes. The ability of the algorithm to segment points into
complete objects and also classify each point into its correct class is evaluated.
Both the segmentation and classification results are compared to datasets which
have been manually ground-truthed. The evaluation demonstrates the success of
the proposed algorithm in segmenting and distinguishing between five classes of
objects in a LIDAR point cloud. Future work in this direction includes developing a
method to identify the volume changes in a scene over time in an effort to provide
further contextual information about a given area.
Biography
Vijayan Asari is the University of Dayton Ohio Research Schol-
ars Endowed Chair in Wide Area Surveillance and a Professor
with the Department of Electrical and Computer Engineering.
He is also the Director of the Center of Excellence for Computer
Vision and Wide Area Surveillance Research (Vision Lab). He is
the Senior Member of IEEE since 2001 and Senior Member of
the SPIE. He co-organized several IEEE and SPIE conferences
and workshops. He is also a Member of IEEE Computational
Intelligence Society (CIS); IEEE Systems, Man and Cybernetics
Society (SMC) Technical Committee of Human Perception in
Vision, Graphics and Multimedia; IEEE Internet of Things (IoT)
Community; Society for Imaging Science and Technology
(IS&T); IS&T Data Analytics and Marketing Task Force; Insti-
tute for Systems and Technologies of Information, Control and
Communication (INSTICC); and American Society for Engineer-
ing Education (ASEE).
vasari1@udayton.eduLIDAR data analysis for automatic region
segmentation and object classification
Vijayan Asari and Nina Varney
University of Dayton, USA
Vijayan Asari et al., Am J Compt Sci Inform Technol 2018, Volume 6
DOI: 10.21767/2349-3917-C1-001