Applied Artificial Intelligence

An International Journal

ISSN: (Print) (Online) Journal homepage: www.tandfonline.com/journals/uaai20

Development Of A Vision- based Anti-drone

Identification Friend Or Foe Model To Recognize
Birds And Drones Using Deep Learning

Yasmine Ghazlane, Maha Gmira & Hicham Medromi

To cite this article: Yasmine Ghazlane, Maha Gmira & Hicham Medromi (2024) Development
Of A Vision- based Anti-drone Identification Friend Or Foe Model To Recognize Birds
And Drones Using Deep Learning, Applied Artificial Intelligence, 38:1, 2318672, DOI:
10.1080/08839514.2024.2318672

To link to this article: https://doi.org/10.1080/08839514.2024.2318672

© 2024 The Author(s). Published with

license by Taylor & Francis Group, LLC.

Published online: 17 Feb 2024.

Submit your article to this journal

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=uaai20
APPLIED ARTIFICIAL INTELLIGENCE
2024, VOL. 38, NO. 1, e2318672 (29 pages)
https://doi.org/10.1080/08839514.2024.2318672

Development Of A Vision- based Anti-drone Identification

Friend Or Foe Model To Recognize Birds And Drones Using
Deep Learning
a,b
Yasmine Ghazlane , Maha Gmiraa, and Hicham Medromib,c
a
School of Digital Engineering and Artificial Intelligence, Euromed Research Center, Euromed University,
Fes, Morocco; bResearch Foundation for Development and Innovation in Science and Engineering
(FRDISI), Casablanca, Morocco; cNational Higher School of Electricity and Mechanic (ENSEM), Hassan II
University, Casablanca, Morocco

ABSTRACT ARTICLE HISTORY

Recently, the growing use of drones has paved the way for Received 13 March 2023
limitless applications in all the domains. However, their mali­ Revised 24 November 2023
cious exploitations have affected the airspace safety, making Accepted 1 February 2024
them double-edged weapons. Therefore, intelligent anti-drone
systems capable of recognizing and neutralizing airborne tar­
gets become highly required. In the existing literature, most of
the attention has been centered on recognizing drones as
unique airborne target, whereas the real challenge is to distin­
guish between drones and non-drone targets. To address this
issue, this study develops an Identification Friend or Foe (IFF)
model able to classify the aerial targets in foe or friend cate­
gories by determining whether the aerial target is a drone or
bird, respectively. To achieve this objective, artificial intelligence
and computer vision approaches have been combined through
transfer learning, data augmentation and other techniques in
our model. Another contribution of this work is the study of the
impact of depth on the classification performance, which is
demonstrated through our experiments. A comparison is per­
formed based on eight models, where EfficientNetB6 shows the
best results with 98.12% accuracy, 98.184% precision, 98.115%
F1 score and 99.85% Area Under Curve (AUC). The computa­
tional results demonstrate the practicality of the developed
model.

Introduction
Recently, drones have become widespread in several domains thanks to the
progress made in Artificial Intelligence (AI), security and transportation fields
(Gupta, Kumari, and Tanwar 2021; Kumar et al. 2021; Kurt et al. 2021;
Serrano-Hernandez, Ballano, and Faulin 2021; Spanaki et al. 2021). Thus,
they are considered a revolution of the 21st century. With the rapid develop­
ment of intelligent techniques that range from image and speech recognition

CONTACT Yasmine Ghazlane [email protected] School of Digital Engineering and Artificial

Intelligence, Euromed Research Center, Euromed University, Fes, 30110, Morocco
© 2024 The Author(s). Published with license by Taylor & Francis Group, LLC.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/
licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly
cited. The terms on which this article has been published allow the posting of the Accepted Manuscript in a repository by the author(s)
or with their consent.
e2318672-2 Y. GHAZLANE ET AL.

to self-decision-making in autonomous robots, drones have benefited from it.

Moreover, they are demonstrating their high ability to provide real-time and
cost-effective solutions in a variety of areas including security, surveillance and
smart city monitoring (Kangunde, Jamisola, and Theophilus 2021; Lin Tan
et al. 2021).
Nonetheless, they may be exploited maliciously and anarchically for
miscellaneous hostile objectives such as spying, smuggling drugs and
contraband, cyber-attacks and terrorism. Their overnight invasion and
unrestricted deployment are threatening public safety and security.
Thus, drone activities need to be controlled properly and efficiently.
This can be ensured only by the anti-drone systems, which are cutting-
edge devices relying on a reverse working process to take down the aerial
target lawfully and safely (U.S.C. Title 49 - TRANSPORTATION, n.d.). In
fact, the anti-process goes through four main phases, namely, Detection,
Identification, Tracking and Interception (DITI) as explained in
(Yasmine, Maha, and Hicham 2022).
The two first phases aim to recognize the airborne target in real-time and
with high confidence scores. The detection and identification seek to depict
and confirm a foreign presence in the sky based on the generated signals (Park
et al. 2021; Taha and Shoufan 2019) and to label the intruder object to “Friend
or Foe” category. Knowing the degree of lethality and hazard of the potential
target, the identification should be done accurately for safety and security
issues, since false alarm responses due to missed or confused detections may
lead to an unsuccessful anti-drone process and a doomed interception that
could have weighty damages for both the target and the environment under
operation.
In general, within safeguarded areas and critical infrastructures, the use
of drones improperly or maliciously can pose significant privacy and
security risks (Kolamunna et al. 2021). As drones have become more
prevalent among civilians, the related technical, security, and public safety
issues must be addressed, regulated, and prevented (Al-Sa’d et al. 2019).
Thus, anti-drone systems are used to counter potential threats posed by
unauthorized or malicious drone activity, especially in highly sensitive
locations such as airfields and military bases (Allahham, Khattab, and
Mohamed 2020). The effectiveness of an anti-drone system often relies on
the incorporated recognition module to provide comprehensive coverage
and accurate identification of drones. In addition, the analysis of the
drone’s payload allows to determine the lethality degree of the drone.
Furthermore, the anarchic and rogue drone deployment poses potential
collisions, environmental disturbances, airspace distress, privacy threats,
security concerns and safety issues. Therefore, an Automatic Drone
Identification (ADI) protection mechanism is proposed in (Svaigen et al.
2021) to detect and avoid malicious drones using the generated acoustic
 APPLIED ARTIFICIAL INTELLIGENCE e2318672-3

signals. Another paper (Casabianca and Zhang 2021) leverages the acoustic
signal to develop a CNN in order to improve the drone’s detection
performance.
Actually, an anti-drone should be able to recognize the main types of
airborne targets and especially distinguish between the prevalent ones; which
are the drones and birds. Since they share the same airspace and altitudes;
mainly the low altitude airspace up to 32 000 ft as an upper limit. At this
altitude, recognizing the flying targets turns out to be a real challenge regard­
ing their similarities which increases the likelihood of false detections.
It is a significant task for an anti-drone system to distinguish accurately
between drones and birds. Bird flights can cause far-reaching impacts within
safeguarded areas and critical infrastructures such as airports and military
bases. Thus, it is of high importance to effectively distinguish between birds
and drones to avoid doomed interception and collateral damages where anti-
drone measures are deployed. For instance, bird strikes with deployed drones
and airplanes can result in severe damage to the environment, which poses
a safety concern and a potential accident. The collision of birds with drones
can damage the frame and the parts of the drone, compromising flight safety.
Efforts to mitigate any potential negative impacts on birds involve the use of
an appropriate recognition model within the anti-drone system. The birds can
be considered as false-positive detections for anti-drone if the used recognition
model determines the bird as a drone and thus the system is falsely triggered
causing a failed neutralization. Due to the complexity of the system, this might
cause unnecessary alerts or interruptions. As anti-drone technologies aim to
detect and mitigate threats from drones, they must distinguish them from
harmless airborne objects, such as birds.
Furthermore, detecting the identity of the airborne targets is crucial since
the drones and birds have similar movement behaviors, radar-cross sections,
very close flying altitudes and speeds, and also their radar signals share similar
signal amplitude as well as the fluctuation of time series and spectrum struc­
ture (Gong et al. 2019). Moreover, when the distance increases, the generated
radio and acoustic signals cannot be recognized and distinguished properly
(Alaparthy, Mandal, and Cummings 2021; Fuhrmann et al. 2017; Patel,
Fioranelli, and Anderson 2018; Torvik, Olsen, and Griffiths 2016). For this
reason, a thorough study of the detection and identification model is of great
importance in reinforcing aerial security and building effective anti-drone
systems. Based on the research work done in [10,17–19], the visual-based
detection is the most advantageous in view of the quality and quantity of
information delivered by the Electro-Optical (EO) and Infrared (IR) sensors.
Recent visual recognition tasks with image detection and tracking use
Convolutional Neural Networks (CNN) as a foundational part to process
and extract the visual features from the input images to provide
a probability distribution over a set of categories (Isaac-Medina et al. 2021).
e2318672-4 Y. GHAZLANE ET AL.

Further, visual detection methods using suitable AI algorithms are often

considered one of the most accurate means for anti-drone detection due to
several reasons such as:

● Adaptability and flexibility of deployment in various locations and

weather conditions.
● Reduced false alarms since the visual detection methods are the most
suitable to distinguish between drones and other flying objects, reduce
false alarms and misidentification of non-threatening airborne targets,
e.g. birds.
● Real-time and effective identification of the visual methods allows the
anti-drone automated system to visually confirm the presence of a drone
or not rapidly with the use of a suitable pre-trained model based on the
related requirements and challenges. Thus, the incorporated identifica­
tion models can rapidly identify the type of the airborne intruder and
assess its potential threat level.

Furthermore, the collected images and videos are more reliable to

recognize and to fit the intruder into its defined class « drone or
bird » by gathering the visual cues; such as the appearance feature,
e.g., colors, contour lines, geometric forms and edges, etc., and the
motion across consecutive frames (Shi et al. 2018). Recent advances in
AI and computer vision have reinforced the visual detection module in
the anti-drone systems, in terms of accuracy and processing time. The
integration of AI algorithms has enabled to take a significant step
forward in making the anti-drone system intelligent, flexible and
autonomous.
To address the existing challenges, we have combined AI techniques,
computer vision approaches and deep learning to develop a novel recogni­
tion model aiming to distinguish between the prevalent aerial targets in
the sky.
In this work, we develop an advanced detection model able to distinguish
between drones and birds. Actually, computational experiments are conducted
by training, validating and testing the model on a real-world dataset. The
computational results demonstrate the practicality of the proposed model.
The main novelty and contributions of this paper can be summarized as
follows:

● Proposing a backbone model suitable for anti-drone deployment.

● Developing an Identification Friend or Foe (IFF) backbone model that
serves as a binary classification backbone model in an anti-drone system
to recognize the types of the aerial targets.
● Study the effect of the depth of the models on the performance.
 APPLIED ARTIFICIAL INTELLIGENCE e2318672-5

In the following, Section II provides a summary of the state-of-the-art studies

on airborne target recognition, whereas the problem-solving methodology is
presented in Section III. Section IV highlights the proposed system setup with
the used techniques and parameters. The experimental results and the com­
parison with the benchmark are detailed in Section V. Finally, we conclude
and address the advantages and limitations of the proposed model as well as
the perspective research directions in Section VI.

Related Work
To date, the recent advances in AI and Machine Learning (ML) have signifi­
cantly accelerated the improvements done on both drones and anti-drone
systems. Thus, the anti-drone systems have become intelligent and autono­
mous in terms of decision-making. Moreover, their capabilities are enhanced
with the fusion of AI approaches with deep learning algorithms, computer
vision techniques and more precisely image recognition.
The four phases of the DITI process aim to automate some or all the operations
to improve the system performance using AI techniques. Thus, the anti-drone
becomes a cutting-edge device by adopting and fusing knowledge from AI,
Internet-of-Things and robotics to respond effectively to the security threats
posed by the malicious drones (Choi 2022; Ding et al. 2018). Since the anti-
drone relies mainly on the detection model used, it is important to select suitable
algorithm and methodology, optimal parameters and appropriate data to achieve
satisfactory results.
One cannot talk of AI and computer vision without talking of image recogni­
tion and classification algorithms and CNN architectures. The achieved advances
have reinforced the effectiveness of the visual detection in anti-drone systems, in
terms of accuracy and processing time. The drone recognition have shifted from
using traditional methods that use low-level handcrafted features and classical
classifiers (Ganti and Kim 2016; Gökçe et al. 2015; Lai, Mejias, and Ford 2011;
Rozantsev, Lepetit, and Fua 2016; Unlu, Zenou, and Rivière 2018; Wang et al.
2018; Wu, Sui, and Wang 2017) to more automated ones represented most of all
by deep neural networks (Wang, Fan, and Wang 2021) considered as a “black-
box” solution for most of the problems (Osco et al. 2021; Seidaliyeva et al. 2020).
In fact, the anti-drone system effectiveness depends mainly on the reliability and
validity of the results delivered by the recognition model.
For this reason, several later research studies have focused on developing
novel drone detection strategies by leveraging different AI approaches with the
purpose of reaching high-level confidence results.
In the following, we review the existing models in the literature as single and
binary airborne target recognition models, which include detection and
classification.
e2318672-6 Y. GHAZLANE ET AL.

Single Target Recognition Models

The single airborne target recognition refers to drone recognition as a single

target. Indeed, the proposed methods train the models only on one target class,
i.e., drones. The latter is a limitation for anti-drone deployment in view of the
omnipresence of the birds in the sky.
In (Garcia, Min Lee, and Kim 2020), the authors used Faster R-CNN with
ResNet-101 to detect drones, reaching an accuracy of 93.40%. However, the
developed model is not trained on the bird dataset and thus it is unable to
properly identify birds. Using visible-band and infrared datasets, the paper
(Isaac-Medina et al. 2021) has shown that the precision counts more than the
time when detecting small drones with the selected Faster R-CNN that yields
the best precision rate. The proposed model didn’t find the optimal perfor­
mance compromise to detect different sizes of drones. Further, the authors in
(Behera and Bazil Raj 2020) have trained Yolov3 during 150 epochs to detect
drones. However, the overall performance does not fulfill the anti-drone
requirements. Further, the paper (Ashraf, Sultani, and Shah 2021) has demon­
strated that a two-stage approach performs better than the one-stage one for
drone detection thanks to its small memory consumption, fast speed and
excellent capability of capturing detailed spatio-temporal information.
Another drone detection method presented in (Zeng et al. 2021) has tested
three CNN namely, R-CNN, YOLOv3, SSD with different backbone architec­
tures, i.e., VGG16, ResNet 50, DarkNet 53 and DenseNet 201. Results showed
that all the models achieve high results rates as long as the dataset is repre­
sentative and varied. However, there is a need to finetune the model on the
anti-drone research task. Deep reinforcement learning was also used and
implemented with Transfer Learning (TL) for drone detection and guidance
as proposed in (Çetin, Barrado, and Pastor 2020). TL improves the overall
performance since it was faster in reaching the threshold reward around 30K
steps with a higher asymptotic performance. Another paper (Çetin, Barrado,
and Pastor 2021) has shown that the combination of Darknet-53 and
EfficientNet-B0 shows good results in terms of precision measured.
Recently, the authors in (Fujii, Akita, and Ukita 2021a) have proposed to
use CenterNet model to detect only birds without considering drone class.
They have shown that the selection of appropriate data augmentation techni­
ques has a direct impact on the detection performance, e.g., random flip,
random scaling, random cropping. The developed model achieved 72.13
mAP.In (Zheng et al. 2021), the authors have focused on proposing a dataset
that covers a wide range of practical scenarios with viewing angles, different
background scenes, relative distance, flying altitude, and lightning conditions.
Eight representative deep learning models have been evaluated to assess the
key challenges in the problem of air-to-air UAV detection. RetinaNet model
showed the best performance. However the ground-to-air detection is not
 APPLIED ARTIFICIAL INTELLIGENCE e2318672-7

considered. Based on an extensive experimentation comparison, the paper

(Zhao et al. 2022) has selected Faster R-CNN with VGG16 as a backbone. The
selected model is fused with a tracking algorithm to detect and track the target
with to keep it within the field of view. Similarly, the paper (Ge et al. 2022) has
integrated a deep Neural Network (DNN) as visual target detector with the
Kalman Filter tracking model.
Recently, the paper (Lee, Chin, and Ho 2023) has proposed a novel method
based on the use of pre-trained model and Yolov4 algorithm with the Kalman
Filter to track the target motion. The model performance achieved an average
precision of 90.04%. However, the model is developed based on only the drone
class.

Binary Target Recognition Models

Binary recognition models make reference to drone and bird recognition

models based on a twofold input dataset.
In (Mahdavi and Rajabi 2020), the authors have shown that CNN models
outperform Support Vector Machine (SVM) and K-Nearest Neighbors (KNN)
in detecting drones using extracts from films in which drones and birds
appear, especially when increasing the convolutional layers and epochs that
has direct impact on the accuracy. However, the used dataset lacks of varia­
bility with respect to the most encountered drones and birds in the sky. The
authors in (Upadhyay, Murthy, and Raj 2021) have used a 15-layer series CNN
to classify the airborne targets as drones and birds. The model achieved 87.5%
accuracy. However, the developed model is not suitable for real-time deploy­
ment due to the included data acquisition module and the processing time of
the developed model. In order to upgrade the visual automated methods
integrated into the anti-drone system, the paper (Oh, Lee, and Kim 2019)
has proposed to examine the most suitable model for drone classification by
comparing the classification accuracy, processing time, and network size of the
representative CNN models. The experimental results have shown that the
shallow and simple structure models produce better training results when
classifying a small number of labels. Knowing that time is a significant chal­
lenge when detecting airborne targets, there is a need to minimize the infer­
ence time without affecting the performance of the model.
The work reported in (Samadzadegan et al. 2022) has implemented binary
detection and recognition with YOLO v4 to make accurate distinction. Their
proposed method sequentially achieved 84% mAP, 81% IoU, and 83% accu­
racy, which can be improved to reach higher detection performance that is
highly required to reinforce airspace safety. The work reported in (Pawełczyk
and Wojtyra 2020) addresses a new drone dataset tested on MobileNetv1
model, resulting in an accuracy and F1 that does not exceed 70% and 60.2%,
respectively.
e2318672-8 Y. GHAZLANE ET AL.

In 2019, IEEE International Conference on Advanced Video and

Signal-based Surveillance (AVSS) has launched the first edition of
Drone-vs-Bird Detection Challenge, whose primary objective was to
correctly detect drones and birds appearing in video sequences without
false alarms.
The winning team of the first edition (Coluccia et al. 2019) has
proposed the use of two separate CNN; U-Net and ResNet. The pro­
posed model has reached 0.73 F1 rate; exceeding thus the other algo­
rithms. In the second edition (Coluccia et al. 2021), the combination of
Cascade R-CNN architecture with ResNeXt as backbone obtained the
best overall average precision performance of 80%. The paper (Coluccia
et al. 2022) presents the results of the 5th edition of the “Drone-vs-
Bird” detection challenge with the use a new dataset comprising 77
video samples of eight commercial drone and two fixed wings. The
first ranking team proposed the use of Yolov5 model integrating syn­
thetic data, negative images during training to reduce false-positive
predictions and a scoring method to assess the performance of a track
(Akyon, Altinuc, and Temizel 2022). The team achieved the best per­
formance in terms of the adopted performance metrics: 0.910 recall and
0.796 AP. In order to integrate these models as recognition modules
within anti-drone systems, there is a need to adopt the models to the
anti-drone requirements and deployments. Also, it is important to
address the overfitting challenge which has not been investigated and
detailed accordingly in the existing related work. Table 1 highlights the
main contributions as well as the advantages and limitations of the
mentioned research works in this section.
In this paper, we have focused and worked on the existing technical and
conceptual gaps in the detection of aerial targets to respond to the existing
needs. Therefore, we have designed a binary classification backbone model for
an anti-drone system to determine whether the aerial target is a drone or
a bird. Indeed, drones and birds are the most encountered airborne targets in
both urban and rural areas. We have experimented an advanced model by
selecting a representative dataset, integrating the most recent computer vision
techniques and comparing several deep learning models to find the best
combination.
Using a backbone model in anti-drone system is significantly important
for an effective target detection and identification. The combination of
this backbone model with the real-time detection model generates results
which are assessed and considered during the tracking and interception
phases.
In the following, we are going to develop an Identification Friend or Foe
(IFF) model to determine if the aerial target is a drone or bird using transfer
learning, optimal hyperparameters and suitable state-of-the art architectures.
 APPLIED ARTIFICIAL INTELLIGENCE e2318672-9

Table 1. Comparative table of the related work.

Proposed method Size of the
Paper and AI approaches dataset Advantages Limitations
(Garcia, Min Lee, Combining Faster NA Good detection Assessing the model with
and Kim R-CNN with performance only one metrics.
2020) a ResNet-101 as Model unable to
backbone detect birds
network.
(Isaac-Medina Using Yolov3 and NA Cross modality detection The dataset includes only
et al. 2021) DETR algorithms with optical and infrared one target class.
signals
(Behera and Yolov3 NA Explainable insight related
The detection is relatively
Bazil Raj to the model used low for anti-drone
2020) system requirement
(Ashraf, Sultani, Two-stage model NPS and FL Using a segmentation-based Lack of the used
and Shah drone dataset approach with pixel and experimental and
2021) channel-wise attention computational
parameters
(Zeng et al. Faster R-CNN, SSD, 13 803 Using CNN for object Need to finetune the
2021) YOLOv3 images detection backbone. selected model and
the used dataset on
the research task
Çetin, Barrado, Reinforcement AirSim Use of TL to improve the Bird class is not
and Pastor learning environments detection performance considered during the
(2020) development of the
model
(Çetin, Barrado, Darknet-53 and 2,000 images Real-time deployment No sufficient learned
and Pastor EfficientNet-B0 knowledge
2021)
(Fujii, Akita, and CenterNet 21 837 Use of varied bird dataset
Drone target class is not
Ukita 2021b) images considered
(Mahdavi and 8-layers CNN model 712 images Experimental comparisons The used dataset is not
Rajabi 2020) with SVM and KNN representative
(Upadhyay, 15-layer CNN model Micro- Detecting drones and birds Not suitable for real-time
Murthy, and doppler using the generated radar deployment
Raj 2021) signals signals
(Oh, Lee, and Alexnet, Resnet, 7000 images Experimental comparison of Time consuming model
Kim n.d..) Squeeznet and several deep CNN models.
other models
(Samadzadegan Yolov4 10000 images Efficient detection in Relatively low achieved
et al. 2022) crowded backgrounds confidence scores
and high similarity
between drones and
birds
(Pawełczyk and MobilNetv1 56 821 Use of a pre-trained model Need to include more
Wojtyra 2020) images on COCO dataset and confidence metrics
finetuning it on the used and improve the
dataset detection results.
(Lee, Chin, and Yolov4 9 332 images Use of pre-trained model Considering only classes
Ho 2023) with the detection and of drone category
interception modules
(Zheng et al. RetinaNet 13 000 Comparing eight deep Not including ground-to-
2021) images learning models air detection
(Zhao et al. Faster-RCNN-VGG16 10 000 Fusing detection and Considering only classes
2022) images tracking algorithms of drone category
(Ge et al. 2022) DNN with Kalman NA Effective detection in Tracking information is
filter challenging scenarios not investigated
e2318672-10 Y. GHAZLANE ET AL.

Proposed Methodology
In most previous studies, most of the attention has been centered on improv­
ing detecting drones as unique targets while the real challenge consists of
distinguishing drones and non-drone objects.
In this study, we develop a proper model able to recognize and distin­
guish drones and birds. Also, we propose to enhance the recognition
system by developing a twofold dataset representing the largest combina­
tion of backgrounds and foregrounds, in different environments and
weather conditions.
Some of the challenges in visual-based drone detection and recognition are:

(i) Resemblance in physical and behavioral attributes of drones and birds

at high altitudes and long distances.
(ii) Existence of drones in a populated background, changing weather
conditions, varying resolutions, and different illumination situations.
(iii) The difficulty of identifying drones and birds to their small size.

To address these challenges, different data augmentation techniques are

applied along with transfer learning to improve visual recognition.

Data Acquisition

In supervised learning, all is about learning from the data and the model are
data driven (Taha and Shoufan 2019). The training process learns the input
output relationship and maps the function that binds the output to the inputs
thanks to the acquired knowledge going from the obvious to underlying
patterns that results in mapping a function between the input and the output.
Once trained, the model makes prediction by assigning the unseen data to the
category it belongs to using the acquired knowledge.
Thus, dataset preparation is a crucial task for our supervised image recogni­
tion task. In fact, we are looking for both high quality and large quantity of
data for creating a model with minimal bias and variance, representing a wide
range of real cases.
In this work, we have carefully collected images with different types of
drones and birds with the purpose to highlight the maximum possible combi­
nation of the most encountered aerial targets in different environments. We
have selected 20 000 instances mainly from (Caltech-UCSD Birds-200-2011, n.
d; Pawełczyk and Wojtyra 2020). The dataset includes different types and
categories of drones and birds captured in different altitudes, weather condi­
tions and locations. It is important to note that the selected dataset ensures
that the trained model can adapt to a variety of situations and environments.
 APPLIED ARTIFICIAL INTELLIGENCE e2318672-11

Figure 1. A selection of bird and drone images from the collected twofold dataset.

Further, to overcome overfitting and to improve the generalization ability of

the model, we have used three data augmentation techniques through chan­
ging brightness range, applying zoom effect and horizontal flip on the images.
In this way, each image generates three augmented ones, added to the original
one, resulting in increasing the whole database four times. To the best of our
knowledge, we have used the largest database of images of drones and birds for
a binary classification model.
The Figure 1 shows a selection of drones and birds images used.
The selection of appropriate training, validation, and testing configurations
is an important step to ascertain optimal performance without overfitting.
This is performed by dividing the dataset into 3 sets: training, validation, and
testing. Assuming that there is a need for hyperparameter tuning, a validation
set creation is necessary aside from the testing set.
The training, validation, and testing were chosen, and the probability was
set to 0.7, 0.2, and 0.1. A probability of 0.7 means that 70% of the images will be
tagged for training, 20% of the images will be assigned for validation, and the
remaining 10% of the images will be tagged for testing.

Transfer Learning and Pre-Trained Models

Transfer Learning Process
To avoid training models from scratch and to address the problem of data
scarcity and resource inefficiency, the visual-based detection methods
leverage massively the Transfer Learning (TL) approach. Accordingly,
e2318672-12 Y. GHAZLANE ET AL.

TL is incorporated to improve the performance of an original model, the

speed and accuracy of the new recognition tasks using the knowledge
gained from a training recognition source on a similar target task (Hua
et al. 2021). In visual automated detection, the transfer learning addresses
cross domain learning problems by extracting visual features from data in
a related source task and transferring them on a target one (Ling, Zhu,
and Li 2015); which improves significantly the generalization ability of
feature extraction process of the pre-trained model (Exploring the
Efficacy of Transfer Learning in Mining Image-Based Software Artifacts
| Journal of Big Data | Full Text, n.d.; Yang et al. 2021). Indeed, the
acquired knowledge is transferred from a pre-trained model on a large
dataset source; for instance ImageNet, to a new target task. The use of
pre-trained models assures the extraction of the fundamental visual fea­
tures on the new target domain, in this case, airborne targets model. This
is achieved by applying the preliminary learned parameters from the pre-
trained model on the new one which reduces data requirement; so that
the final trained model is both accurate and fast (Zhu et al. 2021). The
used ImageNet dataset was created to further the visual recognition tasks,
and it represents the largest database spanning for 1000 object classes
(Krizhevsky, Sutskever, and Hinton 2017).
In our binary classification problem, there are two unseen target classes:
the drone and birds classes. The used transfer learning is established by
freezing much of the convolutional layers by using them as feature extractors
without changing the learned parameters, then fine-tune the last ones on our
targeted dataset by updating their weights during backpropagation to custo­
mize the model with the new recognition categories (Imai, Kawai, and
Nobuhara 2020). Figure 2 highlights the process of knowledge transfer
applied on our classification problem. Figure 2(a) shows the training phase
in which the parameters are trained and fine-tuned. On the other hand,
Figure 2(b) illustrates the testing phase which provides the class probability
of the target.

Backbone Pre-trained Models

So far, there are several powerful deep learning models used for visual object
recognition tasks that achieved outstanding performance. We are going to test
four architectures that have already demonstrated their effectiveness, with
different depths. Further, these models depend mostly upon the provided
visual information and their resolution (Kannojia and Jaiswal 2018), which
is satisfied by our dataset. The pre-trained models are selected on the basis of
their architecture and performance in image recognition since their learned
parameters and knowledge are transferred and fine-tuned on the drone-bird
recognition task.
 APPLIED ARTIFICIAL INTELLIGENCE e2318672-13

Figure 2. Transfer learning framework used from pre-trained models to our target model; divided
on a) a training and b) a testing phases.

We implement two versions of each pre-trained model using different

number of layers so that the depth is increased to analyze the difference and
study the impact. The selected architectures are VGG, ResNet, DenseNet and
EfficientNet.

System Setup
In this section, we detail the used setup and configuration of the proposed
model, adjusted according to the anti-drone requirements and needs.

Experimental Design
The proposed models are trained on a laptop a NVIDIA® GeForce RTX 3050,
Intel® Core i7 -11,370 H Processor 3.3 GHz 12 M with 16GB of memory and
desktop NVIDIA Quadro P4000, Intel(R) Xeon(R) W-2155@ 3.30 GHZ with
e2318672-14 Y. GHAZLANE ET AL.

32GB of memory and Windows as OS. Our experiments are executed using
Tensorflow deep learning framework.
During our training process, the parameter and hypermarameters are care­
fully selected and then tested to be in accordance with the binary classification
task. Table 2 details the used parameters and hyperparameters. Actually, the
parameters are determined from the general context. Meanwhile, the hyper­
parameters are adjustable parameters that are tuned several times before
finding the optimal ones.
Moreover, layer regularization techniques are used to avoid the overfitting
risk and to speed up the training process, which results in lesser variance. In
this work, we have found that dropping randomly 20% of the neurons at each
iteration is the optimal value for our problem.
Since, we are developing a backbone model to be integrated within the anti-
drone system, it’s important to save the best performing weights of the model
for each epochs under h5 files by using ModelCheckpoint technique. Further,
the EarlyStopping allows to stop training when the performance is getting
worse, especially when the model stops improving with the purpose to retain
the optimal generalization performance. After several trials, the early stopping
patience is adapted to each set of epochs when the approximation and com­
plexity errors get close to each other detect and the dominance of the variance
part begins as explained in details in (Prechelt n.d.).

System Model
The overall research flow diagram with the significant followed steps is illu­
strated in Figure 3. We start by feeding the selected dataset to the model which
is augmented and split into three sets: training, validation and testing. After,
the training process starts using the specified parameters and techniques. The
validation process is done in parallel to assess the reliability of tuned hyper­
perparameters. Thereafter, we test the model on an unseen set of images to get
the performance of the model with numerical confidence scores and visual
results.

Evaluation Protocol

In this paper, we are using predefined metrics to assess the proposed models.

Used Metrics
In order to evaluate the performance of the proposed classification models and
methods, we adopt accuracy, precision, F1 score and confusion matrix as
performance metrics. Indeed, these metrics allow better assessment of the
models and methods in the case of binary classification problems since they
rely on the verified and missed detection with True Positives (TP) which are
 APPLIED ARTIFICIAL INTELLIGENCE e2318672-15

Figure 3. Proposed research methodology.

Table 2. Parameters and hyper-parameters setting.

Parameters Value/type
CNN Parameters Epochs [20, 30, 40, 50]
Batch size 16
Image size 224, 224, 3
CNN classes 2
Loss Binary cross entropy
Activation Relu and Sigmoid
Optimizer Adam
Dataset size 28 600 images
Metrics Accuracy, F1 score, precision, confusion matrix
CNN Hyperparameters Learning rate 0.3
Drop factor 0.0001
Learning rate
Fine-tuning layers GlobalAveragePooling, Dropout and Dense
Decay Learning rate/epochs
Dropout 0.2
e2318672-16 Y. GHAZLANE ET AL.

drones, True Negatives (TN) representing the Birds, False Positives (FP) and
Negatives (FN) are the falsely classified drones and birds. Table 3 details the
relation between true and predicted labels.
The used metrics are defined by the following equations.
The accuracy indicates the number of successful detected drones and birds.
P
TP þ TN
Accuracy ¼ P (1)
TP þ TN þ FP þ FN
The precision and recall show the validity of the positive detected aerial
targets.
P
TP
Precision ¼ P (2)
TP þ FP
P
TP
Recall ¼ P (3)
TP þ FN
The F1 score measures the proposed models’ accuracy. Usually, it is used in the
case of binary classification, with positive and negative samples.
precision � recall
F1 score ¼ 2: (4)
precision � recall
The confusion matrix presents the rate of true positive and false positive in the
true and predicted classes of the two classes in a matrix format.

Experimentation and Results

In this section, the experiments of the proposed methodology are presented
and explained through numerical results, graphs and prediction visualizations.
Also, we discuss the impact of the depth of the pre-trained models on the
overall performance. At the end, we present ablation experiments and com­
parison between our proposed model and the existing benchmark instances.

Impact of the Depth

Several researches have demonstrated that deeper architectures their ability to
create profound analysis of the abstract representations of the input and to fit

Table 3. True and false responses.

Predicted Labels
Drone Bird
True Labels Drone True Positive (TP) False Negative (FN)
Bird False Negative (FN) True Negative (TN)
 APPLIED ARTIFICIAL INTELLIGENCE e2318672-17

Table 4. Implementation of the models during 20 epochs.

Architecture Accuracy Precision Recall F1 score
VGG16 0.7795 0.7804 0.7795 0.77932
VGG19 0.82692 0.8271 0.82692 0.8269
ResNet50 0.51 0.51 0.51 0.509
ResNet152 0.96654 0.96764 0.96654 0.96652
DenseNet121 0.95654 0.95655 0.95654 0.95654
DenseNet201 0.96769 0.96898 0.96769 0.96767
EfficientNetB1 0.97808 0.979 0.97808 0.97808
EfficientNetB6 0.98115 0.98184 0.98115 0.98115

Table 5. Impact of the depth.

Architecture/
Metric VGG ResNet DenseNet EfficientNet
Accuracy 5.73% 47.23% 1.15% 0.3%
Precision 5.64% 47.29% 1.28% 0.2%
Recall 5.73% 47.23% 1.15% 0.3%
F1 score 5.75% 47.33% 1.15% 0.3%

classifications functions better as pointed out in (Zhong, Ling, and Wang

2019). In fact, deeper models perform better in the presence of larger datasets
(Schindler, Lidy, and Rauber n.d..). Thus, we have tested to deepen the
architectures by increasing the number of Hidden Layers (HL). The selected
architectures are VGG16, VGG19, ResNet50, ResNet152, DenseNet121,
DenseNet201, EfficientNetB1 and EfficientNetB6. The results of Table 4
show that when the number of HL increases, the performance improves
significantly. It is remarked that there is an improvement of the metrics
when the model version gets deeper. According to Table 5, ResNet has proven
to show the largest improvement when increasing the depth of the architecture
with about 47.25% on average. Because DenseNet and EfficientNet have
already an important number of HL with 121 and 237, respectively, they
didn’t show much improvement with only an average of 1% and 0.3% only.
Thus, it is shown that the overall performance depends mostly on the depth of
the model. Indeed, the deeper architectures perform better till a fixed thresh­
old depth where the models’ performance don’t increase much anymore. This
transition is the stagnation phase. Further, we have found that the threshold
depth for this binary classification problem is fixed at 100 layers with a margin
of ±5% common and applied to all the models. Thus, EfficientNet-B1 and
EfficientNet-B6 are retained as the best performing models. This is explained
also by the considerable number of parameters that are used during the
training with trainable and non-trainable parameters. The used parameters
by the pre-trained models are shown in Table 6. The non-trainable parameters
refer to the feature extraction process where the parameters are kept frozen
and used as they were trained on the pre-trained model while the trainable
parameters refer to the parameters used when fine-tuning the model on our
target dataset. The VGG models are fine-tuned in full due to their small
e2318672-18 Y. GHAZLANE ET AL.

Figure 4. Performance progress during 20 epochs of a) EfficientNetB1 and b) EfficientNetB6

models.

number of layers. From this table, it is shown the number of layers and
parameters has a high impact on the performance since the largest models
extract the maximum amount of informative and discriminative features.

Performance Behavior

Founding that we have experimented a stagnation phase starting from

a common threshold depth, we will use graphical analysis of the performance
of EfficientNetB1 and EfficientNetB6 to make better evaluations since their
results are close. Figure 4 shows the performance behaviors in terms of
accuracy, loss and Area Under Curve (AUC) during the first 20 epochs. The
generated graphs have satisfactory behaviors where the accuracy and AUC are
exponentially increasing without overfitting while loss is decreasing promptly
till approaching zero. However, EfficientNetB6 reaches higher values in terms
of accuracy, loss and AUC. This can be explained by its deeper architecture
and significant number of layers and also its computational time which is high
compared to the other models and mostly EfficientnetB1, as it is depicted in
 APPLIED ARTIFICIAL INTELLIGENCE e2318672-19

Figure 5. Evolution of the performance during 50 epochs.

Table 6. The configuration of the pre-trained models.

Model Layers Trainable parameters Non-trainable parameters Total parameters
VGG16 16 3,323,842 0 3,323,842
VGG19 19 2,160,962 0 2,160,962
ResNet50 50 18,064,386 42,880 18,107,266
ResNet152 152 52,749,314 141,184 52,890,498
DenseNet121 121 6,719,810 79,104 6,798,914
DenseNet201 201 17,535,554 220,928 17,756,482
EfficientNetB1 237 5,361,106 55,015 5,416,121
EfficientNetB6 813 36,413,258 211,767 36,625,025

Table 5. This confirms the reliability of the retained model. Further, we have
investigated their behavior during 50 epochs, and it is confirmed that
EfficientNetB6 has the better performance as shown in Figure 5.

Ablation Experiment
As explained earlier, we have selected carefully appropriate Data
Augmentation (DA) and Fine-Tuning (FT) regularization techniques for
developing the aforementioned model. In order to assess and verify their
respective contributions in the overall performance, we have conducted abla­
tion experiments and comparison with the proposed model. The ablation
experiment is presented in Table 7.
We have tested our retained model without DA and FT techniques, and
then with each of them separately to analyze their impact. Based on the
presented results, it can be seen that the integration of each technique
increases significantly the results.
e2318672-20 Y. GHAZLANE ET AL.

Table 7. Ablation experiments.

Test DA FT Accuracy Precision F1 scoore
1 � � 0.93115 0.93949 0.93083
2 yes � 0.96308 0.96531 0.96303
3 � yes 0.97923 0.97994 0.97922
Proposed model yes yes 0.98115 0.98184 0.98115

Table 8. Comparison between our proposed model and the existing ones.
Papers Accuracy Precision Recall F1 score
(Samadzadegan et al. 2022) 83% 84% 84% 83%
(Pawełczyk and Wojtyra 2020) 70% - - 60.2%
(Coluccia et al. 2019) - - 73%
(Coluccia et al. 2021) - 80% - -
(Coluccia et al. 2022) - 0.796 0.910 -
Our model 98.115% 98.184% 98.115% 98.115%

The integration of DA and FT separately and both of them have increased

the results by 3.3%, 4.9% and 5%, respectively, in terms of accuracy. This
validates their important impact on the proposed model.
Considering different altitudes, flight patterns, and environmental condi­
tions, the dataset reflects the diversity of distances encountered in practical
settings. Across these varied distances, our model has been trained to recog­
nize and differentiate drones and birds. As result, we observed that the model’s
accuracy may vary with distance. Model accuracy is enhanced at shorter
ranges, where detailed features can be captured, but decreases slightly at longer
distances due to reduced feature resolution. Therefore, to mitigate the impact
of different distances on classification accuracy, our training process augmen­
ted the dataset to include observations from a variety of distances and incor­
porated techniques to make the model more robust against distance-related
variations.

Comparison with Benchmark

The results of our retained EfficientNetB6 model and the aforemen­

tioned existing models discussed in Section II are presented in
Table 8. To the best of our knowledge, we confirm that our proposed
model achieves the highest confidence score in terms of accuracy, pre­
cision, recall and F1 score and outperforms the other models. This is
explained by integrating different AI approaches during the training
process such as applying data augmentation on the varied dataset,
adopting an appropriate methodology that corresponds to the current
recognition problem as well as using transfer learning approach from
EfficientNet-B6 pre-trained model which is already trained on ImageNet
dataset and then fine-tune it on our airborne target dataset. Also, the
selected combination of the trainable and non-trainable parameters has
 APPLIED ARTIFICIAL INTELLIGENCE e2318672-21

Figure 6. Confusion matrix of EfficientNetB6 model.

proven its effectiveness on our model. Further, the recognition models

are data driven requiring a considerable and satisfactory dataset in
terms of quality and quantity to enhance their performance and make
them more robust.
The combination of all the aforementionned approaches has significantly
enhanced the performance of the selected model when recognizing drone and
birds.
The confusion matrix of the EfficientNetB6 model obtained during
the test step is presented in Figure 6 with the rate of correct and false
prediction.

Prediction Visualization

Finally, we have integrated a visualization technique in order to illustrate how

the models are performing the classification of drones and birds based on the
true and false predictions; which are the key elements behind the adopted
metrics. A selection of the delivered positive and negative responses is pre­
sented in Figure 7.
The performance of the aforementioned models is explained by the rate
of positive and negative responses when predicting the unseen instances.
Hence, the most accurate models predict correctly the true positives and
negatives predictions, and the other way around. The predictions made by
EfficientNetB6 are the most accurate in determining whether the airborne
target is a drone or bird, which is explained by the high achieved con­
fidence scores during the testing process.
e2318672-22 Y. GHAZLANE ET AL.

Figure 7. Samples from the predicted true and false responses.

Gradient-Weighted Class Activation Map (Grad-CAM)

Further, we have integrated prediction visualization techniques to have
more intuitive outputs and to visualize the performance of the models.
The original and predicted aerial targets on the image are highlighted.
Besides, the GRAD-CAM allows to see the activation maps that are
responsible for the prediction ensured by the last Convolution, Pooling
and activation layers. Figure 8 highlights a selection of the detected
birds and drones using Grad-CAM technique.
 APPLIED ARTIFICIAL INTELLIGENCE e2318672-23

Figure 8. Grad-CAM visualization targets.

Discussion
The developed model successfully responds to the anti-drone needs and fulfills the
related challenges and requirements to recognize efficiently and properly the most
encountered airborne targets, which are drones and birds. The conducted experi­
ments have shown that using transfer learning and finetuning significantly
enhances the detection performance. Through the conducted experiments, we
have found that using the backbone model allows for feature extraction in the real-
time detection module, allowing it to focus more on decision-making rather than
raw data processing. This proposed model serves as backbone for real-time
detection during the anti-drone deployment with a complementary real-time
detection model which includes more targets such as airplanes, dayframes and
building (Yasmine, Maha, and Hicham 2023).
e2318672-24 Y. GHAZLANE ET AL.

Conclusion and Perspectives

This paper has proposed an airborne identification friend or foe model aiming to
determine if the aerial target is a drone or bird. This backbone model aims to
improve the overall performance of the anti-drone systems since the whole
process relies on the information delivered by the detection model. Accordingly,
we have developed a binary classification method following an appropriate meth­
odology using transfer learning, fine-tuned regularization techniques and data
augmentation. These latter have enhanced the overall performance.
Also, we have shown that the model depth impacts to a large extent the
overall performance since deeper models perform better. Most importantly,
there is a common threshold depth from which all the models have similar
behavior and then tend to converge with only an increase in the computational
time versus a small improvement in the results. This is the stagnation phase.
Results showed that EfficientNetB6 model achieved the best performance with
98.12% accuracy, 98.184% precision, 98.115% F1 score, and 99.85% AUC. The
EfficientNetB6 results exceed the existing methods in the literature. Finally, it
can be concluded that using pre-trained models and fine-tune them on our
target dataset following the proposed methodology achieves better results
compared with other existing approaches.
Since this proposed model addresses drone and bird classification, the addi­
tion of a complementary real-time tracking model presents an interesting
research avenue. Also, the reliability of the model could potentially be improved
by developing a fine-grained classification model of the types of drones and also
by adding new airborne targets that are susceptible to be assimilated to drones
and birds which is worth considering it in our upcoming research.
In addition, we are going to combine this backbone model with real-time
detection and tracking modules to get the exact locations and predict the
movements of the foe target.

Disclosure Statement
The authors have no relevant financial or non-financial interests to disclose.

Funding
The authors declare that no funds, grants, or other support were received during the prepara­
tion of this manuscript.

ORCID
Yasmine Ghazlane http://orcid.org/0000-0002-9665-1005
 APPLIED ARTIFICIAL INTELLIGENCE e2318672-25

Data Availability Statement

The datasets generated during and/or analyzed during the current study are available from the
corresponding author on reasonable request.

References
Akyon, F. C., S. O. Altinuc, and A. Temizel. 2022. Slicing aided hyper inference and fine-tuning
for small object detection. 2022 IEEE International Conference on Image Processing (ICIP),
966–70. doi:10.1109/ICIP46576.2022.9897990
Alaparthy, V., S. Mandal, and M. Cummings. 2021. Machine learning vs. Human performance
in the realtime acoustic detection of drones. 2021 IEEE Aerospace Conference (50100), 1–7.
doi:10.1109/AERO50100.2021.9438533
Allahham, M. S., T. Khattab, and A. Mohamed. 2020. Deep learning for RF-Based drone
detection and identification: A multi-channel 1-D convolutional neural networks approach.
2020 IEEE International Conference on Informatics, IoT, and Enabling Technologies (ICIoT),
112–17. doi:10.1109/ICIoT48696.2020.9089657
Al-Sa’d, M. F., A. Al-Ali, A. Mohamed, T. Khattab, and A. Erbad. 2019. RF-based drone
detection and identification using deep learning approaches: An initiative towards a large
open source drone database. Future Generation Computer Systems 100:86–97. doi:10.1016/j.
future.2019.05.007.
Ashraf, M. W., W. Sultani, and M. Shah. 2021. Dogfight: Detecting drones from drones videos.
arXiv 2103:17242 [Cs]. http://arxiv.org/abs/2103.17242.
Behera, D. K., and A. Bazil Raj. 2020. Drone detection and classification using deep learning.
2020 4th International Conference on Intelligent Computing and Control Systems (ICICCS),
1012–16. doi:10.1109/ICICCS48265.2020.9121150
Caltech-UCSD Birds-200-2011. n.d. Accessed March 7, 2022. http://www.vision.caltech.edu/
visipedia/CUB-200-2011.html
Casabianca, P., and Y. Zhang. 2021. Acoustic-based UAV detection using late fusion of deep
neural networks. Drones 5 (3):54. doi:10.3390/drones5030054. Article 3
Çetin, E., C. Barrado, and E. Pastor. 2020. Counter a drone in a complex neighborhood area by
deep reinforcement learning. Sensors 20 (8):2320. doi:10.3390/s20082320. Article 8
Çetin, E., C. Barrado, and E. Pastor. 2021. Improving real-time drone detection for
counter-drone systems. The Aeronautical Journal 125 (1292):1871–96. doi:10.1017/aer.
2021.43.
Choi, Y.-J. 2022. Security threat scenarios of drones and anti-drone technology. Academy of
Strategic Management Journal. 21 (1):7.
Coluccia, A., A. Fascista, A. Schumann, L. Sommer, A. Dimou, D. Zarpalas, M. Méndez, D. de
la Iglesia, I. González, J.-P. Mercier, et al. 2021. Drone vs. Bird detection: Deep learning
algorithms and results from a grand challenge. Sensors 21 (8):2824. doi:10.3390/s21082824.
Coluccia, A., A. Fascista, A. Schumann, L. Sommer, A. Dimou, D. Zarpalas, N. Sharma,
M. Nalamati, O. Eryuksel, K. A. Ozfuttu, et al. 2022. Drone-vs-bird detection challenge at
ICIAP 2021. In Image analysis and processing. ICIAP 2022 workshops, ed. P. L. Mazzeo,
E. Frontoni, S. Sclaroff, and C. Distante, vol. 13374, 410–21. Springer International
Publishing. doi:10.1007/978-3-031-13324-4_35.
Coluccia, A., A. Fascista, A. Schumann, L. Sommer, M. Ghenescu, T. Piatrik, G. De Cubber,
M. Nalamati, A. Kapoor, M. Saqib, et al. 2019. Drone-vs-Bird detection challenge at IEEE
AVSS2019. 2019 16th IEEE International Conference on Advanced Video and Signal Based
Surveillance (AVSS), 1–7. doi:10.1109/AVSS.2019.8909876
e2318672-26 Y. GHAZLANE ET AL.

Ding, G., Q. Wu, L. Zhang, Y. Lin, T. A. Tsiftsis, and Y.-D. Yao. 2018. An amateur drone
surveillance system based on the cognitive internet of things. IEEE Communications
Magazine 56 (1):29–35. doi:10.1109/MCOM.2017.1700452.
Exploring the Efficacy of Transfer Learning in Mining Image-Based Software Artifacts | Journal
of Big Data | Full Text. n.d.-b. Accessed July 21, 2022. https://journalofbigdata.springeropen.
com/articles/10.1186/s40537-020-00335-4
Fuhrmann, L., O. Biallawons, J. Klare, R. Panhuber, R. Klenke, and J. Ender. 2017. Micro-
doppler analysis and classification of UAVs at Ka band. 2017 18th International Radar
Symposium (IRS), 1–9. doi:10.23919/IRS.2017.8008142
Fujii, S., K. Akita, and N. Ukita. 2021a. Distant bird detection for safe drone flight and its
dataset. 2021 17th International Conference on Machine Vision and Applications (MVA),
1–5. doi:10.23919/MVA51890.2021.9511386
Fujii, S., K. Akita, and N. Ukita. 2021b. Distant bird detection for safe drone flight and its
dataset. 2021 17th International Conference on Machine Vision and Applications (MVA),
1–5. doi:10.23919/MVA51890.2021.9511386
Ganti, S. R., and Y. Kim. 2016. Implementation of detection and tracking mechanism for small
UAS. 2016 International Conference on Unmanned Aircraft Systems (ICUAS), 1254–60.
doi:10.1109/ICUAS.2016.7502513
Garcia, A. J., J. Min Lee, and D. S. Kim. 2020. Anti-drone system: A visual-based drone
detection using neural networks. 2020 International Conference on Information and
Communication Technology Convergence (ICTC), 559–61. doi:10.1109/ICTC49870.2020.
9289397
Ge, R., M. Lee, V. Radhakrishnan, Y. Zhou, G. Li, and G. Loianno. 2022. Vision-Based Relative
Detection and Tracking for Teams of Micro Aerial Vehicles (arXiv:2207.08301). arXiv. http://
arxiv.org/abs/2207.08301
Gökçe, F., G. Üçoluk, E. Şahin, and S. Kalkan. 2015. Vision-based detection and distance
estimation of micro unmanned aerial vehicles. Sensors (Basel, Switzerland) 15 (9):23805–46.
doi:10.3390/s150923805.
Gong, J., J. Yan, D. Li, D. Kong, and H. Hu. 2019. Interference of radar detection of drones by
birds. Progress in Electromagnetics Research M 81:1–11. doi:10.2528/PIERM19020505.
Gupta, R., A. Kumari, and S. Tanwar. 2021. Fusion of blockchain and artificial intelligence for
secure drone networking underlying 5G communications. Transactions on Emerging
Telecommunications Technologies 32 (1). doi:10.1002/ett.4176.
Hua, J., L. Zeng, G. Li, and Z. Ju. 2021. Learning for a robot: Deep reinforcement learning,
imitation learning, transfer learning. Sensors 21 (4):1278. doi:10.3390/s21041278. Article 4
Imai, S., S. Kawai, and H. Nobuhara. 2020. Stepwise PathNet: A layer-by-layer
knowledge-selection-based transfer learning algorithm. Scientific Reports 10 (1):8132.
doi:10.1038/s41598-020-64165-3.
Isaac-Medina, B. K. S., M. Poyser, D. Organisciak, C. G. Willcocks, T. P. Breckon, and
H. P. H. Shum. 2021. Unmanned aerial vehicle visual detection and tracking using deep
neural networks: A performance benchmark. arXiv 2103.13933 [Cs]. http://arxiv.org/abs/
2103.13933
Kangunde, V., R. S. Jamisola, and E. K. Theophilus. 2021. A review on drones controlled in
real-time. International Journal of Dynamics and Control 9 (4):1832–46. doi:10.1007/s40435-
020-00737-5.
Kannojia, S. P., and G. Jaiswal. 2018. Effects of varying resolution on performance of CNN
based image classification an experimental study. International Journal of Computer Sciences
& Engineering 6 (9):451–56. doi:10.26438/ijcse/v6i9.451456.
Kolamunna, H., T. Dahanayaka, J. Li, S. Seneviratne, K. Thilakaratne, A. Y. Zomaya, and
A. Seneviratne. 2021. DronePrint: Acoustic signatures for open-set drone detection and
 APPLIED ARTIFICIAL INTELLIGENCE e2318672-27

identification with online data. Proceedings of the ACM on Interactive, Mobile, Wearable and
Ubiquitous Technologies 5 (1):1–31. doi:10.1145/3448115.
Krizhevsky, A., I. Sutskever, and G. E. Hinton. 2017. ImageNet classification with deep
convolutional neural networks. Communications of the ACM 60 (6):84–90. doi:10.1145/
3065386.
Kumar, A., S. Bhatia, K. Kaushik, S. M. Gandhi, S. G. Devi, J. De, D. A. Pacheco, and A. Mashat.
2021. Survey of promising technologies for quantum drones and networks. Institute of
Electrical and Electronics Engineers Access 9:125868–911. doi:10.1109/ACCESS.2021.
3109816.
Kurt, A., N. Saputro, K. Akkaya, and A. S. Uluagac. 2021. Distributed connectivity main­
tenance in swarm of drones during post-disaster transportation applications. IEEE
Transactions on Intelligent Transportation Systems 22 (9):6061–73. doi:10.1109/TITS.
2021.3066843.
Lai, J., L. Mejias, and J. J. Ford. 2011. Airborne vision-based collision-detection system. Journal
of Field Robotics 28 (2):137–57. doi:10.1002/rob.20359.
Lee, Z. W., W. H. Chin, and H. W. Ho. 2023. Air-to-air micro air vehicle interceptor with an
embedded mechanism and deep learning. Aerospace Science and Technology 135:108192.
doi:10.1016/j.ast.2023.108192.
Ling, S., F. Zhu, and X. Li. 2015. Transfer learning for visual categorization: A survey. IEEE
Transactions on Neural Networks and Learning Systems 26 (5):1019–34. doi:10.1109/TNNLS.
2014.2330900.
Lin Tan, L. K., B. C. Lim, G. Park, K. H. Low, and V. C. Seng Yeo. 2021. Public acceptance of
drone applications in a highly urbanized environment. Technology in Society 64:101462.
doi:10.1016/j.techsoc.2020.101462.
Mahdavi, F., and R. Rajabi. 2020. Drone detection using convolutional neural networks. 2020
6th Iranian Conference on Signal Processing and Intelligent Systems (ICSPIS), 1–5. doi:10.
1109/ICSPIS51611.2020.9349620
Oh, H. M., H. Lee, and M. Y. Kim. 2019. Comparing Convolutional Neural Network
(CNN) models for machine learning-based drone and bird classification of anti-drone
system. In 2019 19th International Conference on Control, Automation and Systems
(ICCAS) (pp. 87-90). IEEE.
Osco, L. P., J. Marcato Junior, A. P. Marques Ramos, L. A. de Castro Jorge, S. N. Fatholahi, J. de
Andrade Silva, E. T. Matsubara, H. Pistori, W. N. Gonçalves, and J. Li. 2021. A review on
deep learning in UAV remote sensing. International Journal of Applied Earth Observation
and Geoinformation 102:102456. doi:10.1016/j.jag.2021.102456.
Park, S., H. T. Kim, S. Lee, H. Joo, and H. Kim. 2021. Survey on anti-drone systems:
Components, designs, and challenges. Institute of Electrical and Electronics Engineers
Access 9:42635–59. doi:10.1109/ACCESS.2021.3065926.
Patel, J. S., F. Fioranelli, and D. Anderson. 2018. Review of radar classification and RCS
characterisation techniques for small UAVs or drones. IET Radar, Sonar & Navigation
12 (9):911–19. doi:10.1049/iet-rsn.2018.0020.
Pawełczyk, M. Ł., and M. Wojtyra. 2020. Real world object detection dataset for quadcopter
unmanned aerial vehicle detection. Institute of Electrical and Electronics Engineers Access
8:174394–409. doi:10.1109/ACCESS.2020.3026192.
Prechelt, L. (n.d.). Early stopping | but when? 15.
Rozantsev, A., V. Lepetit, and P. Fua. 2016. Detecting flying objects using a single moving
camera IEEE transactions on pattern analysis and machine intelligence. 39(5): 879–892.
Samadzadegan, F., F. Dadrass Javan, F. Ashtari Mahini, and M. Gholamshahi. 2022. Detection
and recognition of drones based on a deep convolutional neural network using visible
imagery. Aerospace 9 (1):31. doi:10.3390/aerospace9010031.
e2318672-28 Y. GHAZLANE ET AL.

Schindler, A., T. Lidy, and A. Rauber. n.d. Comparing shallow versus deep neural network
architectures for automatic music genre classification, vol. 5.
Seidaliyeva, U., D. Akhmetov, L. Ilipbayeva, and E. T. Matson. 2020. Real-time and accurate
drone detection in a video with a static background. Sensors 20 (14):3856. doi:10.3390/
s20143856.
Serrano-Hernandez, A., A. Ballano, and J. Faulin. 2021. Selecting freight transportation modes
in last-mile urban distribution in Pamplona (Spain): An option for drone delivery in smart
cities. Energies 14 (16):4748. doi:10.3390/en14164748.
Shi, X., C. Yang, W. Xie, C. Liang, Z. Shi, and J. Chen. 2018. Anti-drone system with multiple
surveillance technologies: Architecture, implementation, and challenges. IEEE
Communications Magazine 56 (4):68–74. doi:10.1109/MCOM.2018.1700430.
Spanaki, K., E. Karafili, U. Sivarajah, S. Despoudi, and Z. Irani. 2021. Artificial intelli­
gence and food security: Swarm intelligence of AgriTech drones for smart AgriFood
operations. Production Planning & Control 33 (16):1498–516. doi:10.1080/09537287.
2021.1882688 .
Svaigen, A. R., L. M. S. Bine, G. L. Pappa, L. B. Ruiz, and A. A. F. Loureiro (2021). Automatic
drone identification through rhythm-based features for the internet of drones. 2021 IEEE
33rd International Conference on Tools with Artificial Intelligence (ICTAI), 1417–21. doi:10.
1109/ICTAI52525.2021.00225
Taha, B., and A. Shoufan. 2019. Machine learning-based drone detection and classification:
State-of-the-art in research. Institute of Electrical and Electronics Engineers Access
7:138669–82. doi:10.1109/ACCESS.2019.2942944.
Torvik, B., K. E. Olsen, and H. Griffiths. 2016. Classification of birds and UAVs based on radar
polarimetry. IEEE Geoscience and Remote Sensing Letters 13 (9):1305–09. doi:10.1109/LGRS.
2016.2582538.
Unlu, E., E. Zenou, and N. Rivière. 2018. Using shape descriptors for UAV detection. Electronic
Imaging 2017 (9):128–5. https://hal.archives-ouvertes.fr/hal-01740282.
Upadhyay, M., S. K. Murthy, and A. A. B. Raj. 2021. Intelligent system for real time detection
and classification of aerial targets using CNN. 2021 5th International Conference on
Intelligent Computing and Control Systems (ICICCS), 1676–81. doi:10.1109/ICICCS51141.
2021.9432136
U.S.C. Title 49—TRANSPORTATION. n.d.-c. Accessed March 15, 2021. https://www.govinfo.
gov/content/pkg/USCODE-2018-title49/html/USCODE-2018-title49-subtitleVII-partA-
subpartiii-chap448-sec44801.htm
Wang, P., E. Fan, and P. Wang. 2021. Comparative analysis of image classification algorithms
based on traditional machine learning and deep learning. Pattern Recognition Letters
141:61–67. doi:10.1016/j.patrec.2020.07.042.
Wang, Z., L. Qi, Y. Tie, Y. Ding, and Y. Bai. 2018. Drone detection based on FD-HOG
descriptor. 2018 International Conference on Cyber-Enabled Distributed Computing and
Knowledge Discovery (CyberC), 433–33. doi:10.1109/CyberC.2018.00084
Wu, Y., Y. Sui, and G. Wang. 2017. Vision-based real-time aerial object localization and
tracking for UAV sensing system. Institute of Electrical and Electronics Engineers Access
5:23969–78. doi:10.1109/ACCESS.2017.2764419.
Yang, X., Y. Zhang, W. Lv, and D. Wang. 2021. Image recognition of wind turbine blade
damage based on a deep learning model with transfer learning and an ensemble learning
classifier. Renewable Energy 163:386–97. doi:10.1016/j.renene.2020.08.125.
Yasmine, G., G. Maha, and M. Hicham. 2022. Survey on current anti-drone systems: Process,
technologies, and algorithms. International Journal of System of Systems Engineering
12 (3):235–70. doi:10.1504/IJSSE.2022.125947.
 APPLIED ARTIFICIAL INTELLIGENCE e2318672-29

Yasmine, G., G. Maha, and M. Hicham. 2023. Anti-drone systems: An attention based
improved YOLOv7 model for a real-time detection and identification of
multi-airborne target. Intelligent Systems with Applications 20:200296. doi:10.1016/j.
iswa.2023.200296 .
Zeng, Y., Q. Duan, X. Chen, D. Peng, Y. Mao, and K. Yang. 2021. UAVData: A dataset for
unmanned aerial vehicle detection. Soft Computing 25 (7):5385–93. doi:10.1007/s00500-020-
05537-9.
Zhao, J., J. Zhang, D. Li, and D. Wang. 2022. Vision-Based Anti-UAV Detection and Tracking
(arXiv:2205.10851). arXiv. http://arxiv.org/abs/2205.10851
Zheng, Y., Z. Chen, D. Lv, Z. Li, Z. Lan, and S. Zhao. 2021. Air-to-air visual detection of
micro-UAVs: An experimental evaluation of deep learning. IEEE Robotics and Automation
Letters 6 (2):1020–27. doi:10.1109/LRA.2021.3056059.
Zhong, G., X. Ling, and L. Wang. 2019. From shallow feature learning to deep learning:
Benefits from the width and depth of deep architectures. WIREs Data Mining and
Knowledge Discovery 9 (1). doi:10.1002/widm.1255.
Zhu, W., B. Braun, L. H. Chiang, and J. A. Romagnoli. 2021. Investigation of transfer learning
for image classification and impact on training sample size. Chemometrics and Intelligent
Laboratory Systems 211:104269. doi:10.1016/j.chemolab.2021.104269.