Page 66 - 2023-Vol19-Issue2

Received: 28 February 2023 | Revised: 4 April 2023 | Accepted: 4 April 2023

DOI: 10.37917/ijeee.19.2.8 Vol. 19 | Issue 2 | December 2023

Open Access

Iraqi Journal for Electrical and Electronic Engineering

Original Article

Feature Deep Learning Extraction Approach for Object

Detection in Self-Driving Cars

Namareq Odey *, Ali Marhoon
1Electrical Engineering Department, University of Basrah, Basrah, Iraq

Correspondance
* Namareq Odey
Electrical Engineering department
University of Basrah, Basrah, Iraq
Email: namareqalmotter22@gmail.com

Abstract
Self-driving cars are a fundamental research subject in recent years; the ultimate goal is to completely exchange the
human driver with automated systems. On the other hand, deep learning techniques have revealed performance and
effectiveness in several areas. The strength of self-driving cars has been deeply investigated in many areas including
object detection, localization as well, and activity recognition. This paper provides an approach to deep learning; which
combines the benefits of both convolutional neural network CNN together with Dense technique. This approach learns
based on features extracted from the feature extraction technique which is linear discriminant analysis LDA combined
with feature expansion techniques namely: standard deviation, min, max, mod, variance and mean. The presented
approach has proven its success in both testing and training data and achieving 100% accuracy in both terms.

Keywords
SELF-DRIVING CARS, FEATURE EXPANSION, DEEP LEARNING, CNN, LDA.

I. INTRODUCTION task of the perception system of a vehicle computer is to
reveal nearby objects simultaneously with making the optimal
Autonomous Vehicle(AV) performs an important field both decision like path planning as well as collision avoidance.
in modern mechanical and electrical engineering as well as Autonomous vehicles have gained explosive improvement
intelligent transportation systems. The field of self–driving during the past decade [6].
cars is evolving very fast, it utilizes a combination of sensors,
actuators, machine learning systems, and complex and pow- Self-driving cars are under the spotlight and gained tremen-
erful algorithms to implement software and travel between dous improvement throughout the past decade [7], despite the
destinations without human interference. technological improvement, autonomous vehicle systems are
still far away from being trustworthy and reliable. Several ac-
On the other hand, deep learning is an improved subdo- cidents in self-driving cars are caused by misclassification or
main of machine learning. It enhances the modeling of com- not recognized objects [8]; therefore, increasing the capability
plex relationships and concepts using multiple levels of repre- of object detection in these systems is a very high priority.
sentation. Supervised and unsupervised learning algorithms
are used to construct successively higher levels of abstraction, Recent studies in deep learning have boosted the perfor-
defined using the output features from lower levels. mance of object detection algorithms to a higher level [9],[10].
This paper introduces an approach of deep learning mimics
However, object detection using deep learning is at the the human brain in case of learning based on features instead
core of a great number of computer vision applications like of images, to be used in Pedestrians, Cars, bakers, trucks and
face detection [?], video surveillance [1], optical character traffic lights recognition and identification aiming to develop
recognition [2], and object counting/tracking [3],[4]. autonomous driving technologies.

More specifically, in self-driving cars [5], the ultimate

This is an open-access article under the terms of the Creative Commons Attribution License,
which permits use, distribution, and reproduction in any medium, provided the original work is properly cited.
©2023 The Authors.
Published by Iraqi Journal for Electrical and Electronic Engineering | College of Engineering, University of Basrah.

https://doi.org/10.37917/ijeee.19.2.8 |https://www.ijeee.edu.iq 62