Vol.:(0123456789) 1 3
Journal of Big Data Analytics in Transportation
https://doi.org/10.1007/s42421-019-00005-9
ORIGINAL PAPER
Transfer Learning Using Deep Neural Networks for Classifcation
of Truck Body Types Based on Side‑Fire Lidar Data
Reza Vatani Nezafat
1
· Olcay Sahin
1
· Mecit Cetin
1
Received: 19 February 2019 / Revised: 13 April 2019 / Accepted: 18 April 2019
© Springer Nature Singapore Pte Ltd. 2019
Abstract
Vehicle classifcation is one of the most essential aspects of highway performance monitoring as vehicle classes are needed
for various applications including freight planning and pavement design. While most of the existing systems use in-pavement
sensors to detect vehicle axles and lengths for classifcation, researchers have also explored traditional approaches for image-
based vehicle classifcation which tend to be computationally expensive and typically require a large amount of data for
model training. As an alternative to these image-based methods, this paper investigates whether it is possible to transfer the
learning (or parameters) of a highly accurate pre-trained (deep neural network) model for classifying truck images generated
from 3D-point cloud data from a LiDAR sensor. In other words, without changing the parameters of several well-known
convolutional neural networks (CNNs), such as AlexNet, VggNet and ResNet, this paper shows how they can be adopted to
extract the needed features to classify trucks, in particular trucks with diferent types of trailers. This paper demonstrates the
applicability of these CNNs for solving the vehicle classifcation problem through an extensive set of experiments conducted
on images created based on data from a LIDAR sensor. Results show that using pre-trained CNN models to extract low-level
features within images yield signifcantly accurate results, even with a relatively small size of training data that are needed
for the classifcation step at the end.
keywords Transfer learning · LiDAR · Convolutional neural network · Freight monitoring
Introduction and Background
The exponential growth of technology in the last 50 years
has resulted in an unprecedented increase in the number of
vehicles on the road. Therefore, the need for innovative solu-
tions to increase the performance of transportation systems
is at highest. Nearly all trafc management approaches have
a monitoring mechanism. The purpose of this mechanism
can be an estimation of macroscopic/microscopic trafc
parameters or classifcation of vehicle types, which are use-
ful in many applications such as freight planning, highway
design and maintenance, trafc operations, transportation
planning, and tolling. Some of the commonly used vehi-
cle detection technologies such as inductive magnetic loop
detectors (Mita and Imazu 1995), magnetic sensors (He et al.
2012), acoustic sensors (Wang et al. 2014), infrared sensors
(Tropartz et al. 1999), and weigh-in-motion sensors (Nichols
and Cetin 2007) are already used in practice to count the
number of axels or capture other physical features needed
for the classifcation algorithms.
In addition to the vehicle detection technologies listed
above, surveillance cameras have been another common
option to monitor trafc fow since they have a relatively
low maintenance cost and do not necessitate trafc disrup-
tion during installation. Because of the widespread use of
cameras, researchers within the transportation community
have tried various image/video-based vehicle classifca-
tion approaches (Hsieh et al. 2006). Early studies have tried
heuristic strategies to do classifcation task using low-level
* Reza Vatani Nezafat
rvata001@odu.edu
https://www.linkedin.com/in/rvntri/
Olcay Sahin
osahi001@odu.edu
https://www.linkedin.com/in/olcaysah/
Mecit Cetin
mcetin@odu.edu
https://www.linkedin.com/in/cetin-mecit-09923013/
1
Department of Civil and Environmental Engineering,
Old Dominion University, Kaufman Hall 135, Norfolk,
VA 23529, USA