[Link].

com/scientificreports

OPEN Deep convolutional neural

network-based enhanced crowd
density monitoring for intelligent
urban planning on smart cities
Wahida Mansouri1, Manal Abdullah Alohali2, Hamed Alqahtani3, Nuha Alruwais4,
Menwa Alshammeri5 & Ahmed Mahmud6
The concept of a smart city has spread as a solution ensuring wider availability of data and services
to citizens, apart from as a means to lower the environmental footprint of cities. Crowd density
monitoring is a cutting-edge technology that enables smart cities to monitor and effectively manage
crowd movements in real time. By utilizing advanced artificial intelligence and video analytics,
valuable insights are accumulated from crowd behaviour, assisting cities in improving operational
efficiency, improving public safety, and urban planning. This technology also significantly contributes
to resource allocation and emergency response, contributing to smarter, safer urban environments.
Crowd density classification in smart cities using deep learning (DL) employs cutting-edge NN
models to interpret and analyze information from sensors such as IoT devices and CCTV cameras.
This technique trains DL models on large datasets to accurately count people in a region, assisting
traffic management, safety, and urban planning. By utilizing recurrent neural networks (RNNs) for
time-series data and convolutional neural networks (CNNs) for image processing, the model adapts to
varying crowd scenarios, lighting, and angles. This manuscript presents a Deep Convolutional Neural
Network-based Crowd Density Monitoring for Intelligent Urban Planning (DCNNCDM-IUP) technique
on smart cities. The proposed DCNNCDM-IUP technique utilizes DL methods to detect crowd
densities, which can significantly assist in urban planning for smart cities. Initially, the DCNNCDM-IUP
technique performs image preprocessing using Gaussian filtering (GF). The DCNNCDM-IUP technique
utilizes the SE-DenseNet approach, which effectually learns complex feature patterns for feature
extraction. Moreover, the hyperparameter selection of the SE-DenseNet approach is accomplished
by using the red fox optimization (RFO) methodology. Finally, the convolutional long short-term
memory (ConvLSTM) methodology recognizes varied crowd densities. A comprehensive simulation
analysis is conducted to demonstrate the improved performance of the DCNNCDM-IUP technique.
The experimental validation of the DCNNCDM-IUP technique portrayed a superior accuracy value of
98.40% compared to existing DL models.

Keywords Crowd density monitoring, Red fox optimization, Urban planning, Deep learning, Gaussian
filtering

Crowd management is the capability of monitoring and directing a set of people to guarantee their safety1.
Similar supporting technology is used to help move people to their destinations effectively and design a new
service based on their behaviour. Crowd management is to take care of people on foot. It applies to people on
public transport and in cars2. Every day, the entire solution permits the city to notice the activity of their citizens,

1Department of Computer Science and Information Technology, Faculty of Sciences and Arts, Turaif, Northern
Border University, 91431 Arar, Saudi Arabia. 2Department of Information Systems, College of Computer and
Information Sciences, Princess Nourah Bint Abdulrahman University, P.O. Box 84428, 11671 Riyadh, Kingdom of
Saudi Arabia. 3Unit of Cybersecurity, Department of Information Systems, Center of Artificial Intelligence, College
of Computer Science, King Khalid University, Abha, Kingdom of Saudi Arabia. 4Department of Computer Science
and Engineering, College of Applied Studies and Community Services, King Saud University, [Link] 22459,
11495 Riyadh, Kingdom of Saudi Arabia. 5Department of Computer Science, College of Computer and Information
Sciences, Jouf University, Sakakah, Kingdom of Saudi Arabia. 6Research Center, Future University in Egypt, New
Cairo 11835, Egypt. email: maalohaly@[Link]

Scientific Reports | (2025) 15:5759 | [Link] 1

[Link]/scientificreports/

allowing them to improve the urban schedule and resulting in improved satisfaction between city people and
visitors. Through the fast development of smart securities, smart communities, and smart cities, abnormal
behaviour studies have become a new point in the crowd-even investigation3. The influence of urbanization
puts immense pressure on city planning and management. Miniaturization-sensing technologies and the advent
of IoT devices have inspired the emergence of smart cities4. Crowd analysis is paramount in various smart
city applications, viz., crowd counting, crowd tracking, and drone-based crowd analysis. In crowd examination,
crowd counting is an essential everyday mission, such as gathering many walkers in an image or video sequence5.
The crowd-counting work applies to many real-life situations, likely urban planning, huge gatherings, and
pedestrian surveillance. Intelligent monitoring has increased tremendously in public places like hospitals,
stations, shopping malls, and campuses. Among them, crowd point is the fundamental problem that disturbs
the investigation of the unusual manner of the crowd6. Owing to its extensive coverage and affluent information,
video-based sensing is a powerful sensing technology for crowd monitoring7. Therefore, there is a need for
dedicated video processing techniques to extract crowd density parameters from raw images. Recently, it
has achieved remarkable growth in video-based crowd density estimation algorithms, and video processing
algorithms have shifted from classical computer vision to DL-based techniques, i.e., CNN8. Generally, DL-based
techniques improved estimation accuracy significantly, but the extreme computing difficulty of DL models
intensified the model’s performance, causing poor real-time performance9. Hence, enhancing the effectiveness
of the crowd density estimation algorithm without any performance loss is crucial for improving public safety10.
Figure 1 represents the structure of smart cities.
This manuscript presents a Deep Convolutional Neural Network-based Crowd Density Monitoring for
Intelligent Urban Planning (DCNNCDM-IUP) technique on smart cities. The proposed DCNNCDM-IUP
technique utilizes DL methods to detect crowd densities, which can significantly assist in urban planning for
smart cities. Initially, the DCNNCDM-IUP technique performs image preprocessing using Gaussian filtering
(GF). The DCNNCDM-IUP technique utilizes the SE-DenseNet approach, which effectually learns complex
feature patterns for feature extraction. Moreover, the hyperparameter selection of the SE-DenseNet approach
is accomplished by using red fox optimization (RFO) methodology. Finally, the convolutional long short-term
memory (ConvLSTM) methodology recognizes varied crowd densities. A comprehensive simulation analysis is
conducted to demonstrate the improved performance of the DCNNCDM-IUP technique. The key contribution
of the DCNNCDM-IUP technique is listed below.

• The DCNNCDM-IUP model utilizes DL techniques to precisely detect crowd densities, giving valuable in-
sights for urban planning in smart cities. Advanced methods for real-time crowd analysis assist in better
decision-making and optimization of urban spaces. Furthermore, crowd monitoring aims to improve safety,
resource allocation, and overall city management.
• The GF approach is employed as an image preprocessing step to improve the quality of input data, mitigating
noise and improving clarity. This preprocessing technique confirms that the subsequent DL models receive
cleaner, more accurate data. Refining the image quality significantly enhances the model’s performance in
detecting crowd densities.
• The SE-DenseNet-based methodology extracts key features from images, improving the model’s capability to
detect crowd densities with high accuracy. This technique employs the power of dense connections and SE
blocks to capture complex patterns. As a result, it significantly improves the precision and efficiency of crowd
density detection in urban environments.
• The RFO methodology is utilized to choose optimal hyperparameters for the SE-DenseNet model, improving
its performance. By effectively searching the parameter space, RFO confirms that the model achieves the opti-
mum possible accuracy for crowd density detection. This optimization process crucially improves the overall
efficiency and robustness of the DL model.
• The integration of ConvLSTM for recognizing varying crowd densities presents a novel methodology that
integrates both spatial and temporal data. This method allows the model to capture dynamic changes in
crowd behaviour over time, enhancing accuracy in real-world scenarios. By using ConvLSTM, the system can
predict crowd patterns in complex urban environments with greater precision. This innovative integration
improves the capability of the model to monitor and manage crowd dynamics in smart cities.

The remaining sections of the article are arranged as follows: “Literature review” section offers the literature
review, and “Methods and techniques” represents the proposed method. Then, “Experimental analysis and
results” section elaborates on the results evaluation, and “Conclusion” section completes the work.

Literature review
Gazis and Katsiri11 presented a new middleware that uses less power computing and cost-effective devices
like Raspberry Pi to analyze data from WSN. It is mainly planned for internal settings such as past museums
and buildings. It helps as a removal support by observing possession and determining the reputation of exact
subjects, art exhibitions, or areas. The middleware uses a simple method, the MapReduce technique, to collect
and allocate data on WSN through obtainable computer nodes. Guo et al.12 developed a lightweight system
called the Ghost Attention Pyramid Networks (GAPNets) method. A lightweight GhostNet was assumed to
support encrypting lower-level features. Furthermore, an effective pyramid fusion unit with a 4-branch structure
was constructed to get a multiscale hierarchy representation while decreasing the parameter. Lastly, a decoder
produces the estimation by using a sequence of transferred convolutional blocks. The authors13 presented Aquila
Optimizer with TL-based Crowd Density Analysis for Sustainable Smart Cities (AOTL-CDA3S). To attain
this, the developed method first uses a weighted average filter (WAF) model to enhance the excellence of the
input surrounds. Then, the proposed system uses an AO technique with the SqueezeNet method for feature

Scientific Reports | (2025) 15:5759 | [Link] 2

[Link]/scientificreports/

Fig. 1. Structure of smart cities.

extractor purposes. Lastly, an XGBoost technique has been utilized to categorize crowd densities. In14, the scale
attentive aggregation network (SA2Net) has been projected. In particular, the SA2Net has dual fundamental
modules, i.e. background noise suppressor (BNS) and multiscale feature aggregator (MFA). The MFA is intended
in a 4-pathway infrastructure and gathers the multiscale feature to simplify the correlation among dissimilar
measures. The BNS module uses the background data among the matrices of the input key and self-attention
to overpower the contextual noise. Zhai et al.15 introduced a new attentive hierarchy ConvNet (AHNet). The
AHNet removes the hierarchy feature by intended discriminative feature extraction and extracts the semantic
feature using a coarse-to-fine method using a hierarchical mixture approach. A module of recalibrated attention
(RA) was constructed in numerous stages to overwhelm the effect of contextual interference, and a feature
enhancement model was made to detect head areas using innumerable measures.
Alrowais et al.16 presented a Meta-heuristics with Deep TL-Enabled Intelligent CDDC (MDTL-ICDDC)
technique for the video surveillance methods. The presented model influences an SSA with NASNetLarge
architecture as a feature extractor, so the hyperparameter selection method was implemented using SSA. Also,
the WELM model was used to determine crowd density and identify the population. Lastly, the krill swarm
algorithm (KSA) was employed for the parameter optimizer method. Lin and Hu17 presented a multi-branch
method, Repmobilenet, for a multiscale spatial extraction of features. Repmobilenet trains with lightweight

Scientific Reports | (2025) 15:5759 | [Link] 3

[Link]/scientificreports/

depthwise separable convolution block (DSBlock) that can efficiently remove features of the dense crowd for
managing with scale disparity. Here, the multi-branch framework is changed into a one-branch feedforward
method over mechanical reparameterization. In this method, Repmobilenet utilizes a multi-branch over-
parameterized layout. Also, the technique inserted dilated convolution in the back end to enlarge the receptive
area of the method. Sharma et al.18 introduced a united solution to achieve the crowd density valuation and
behaviour diagnosis. The density of crowd valuation was implemented by emerging a novel CNN-based
structure that reflects scale data and overwhelms the problems of scale disparities. To address this concern of
scale disparity, the paper presented a module of scale-aware attention. In contrast, manifold branches of the self-
attention model were connected to recover the gap of scale recognition. Selvarajan et al.19 analyze blockchain
(BC) technology for improving the security of Internet of Things (IoT) systems in smart city applications.
The model also integrates a BC technique with a neuro-fuzzy approach to improve data processing security,
maximize monitoring unit confidence, and ensure energy-efficient task completion with enhanced security. In20,
parametric design models incorporating point-to-point protocol and mode control optimization are proposed
to enhance a smart transportation system. Applying multi-objective min-max functions minimizes waiting
times for users. Additionally, it presents a line-following mechanism with improved connectivity, independent
detection, and revitalization functions. Selvarajan and Manoharan21 integrate wireless technologies and digital
twins for monitoring smart cities, improving economic progress and decision-making. The system mitigates
packet exchange and retransmission using IoT and Constrained Application Protocol (CoAP). The authors
in22 explore the integration of sensors and IoT for smart city monitoring, utilizing a node control process. It
integrates K-means clustering and C4.5 classification approaches to secure data and ensure proper responses.
Shitharth et al.23 developed a Cyborg intelligence security model for smart city systems, utilizing QIDI for
data preprocessing, CSOM for feature optimization, and RMML for intrusion classification. The framework
improves attack detection performance by enhancing feature selection, learning speed, and classification
accuracy, ensuring efficient security for smart city networks. Thirumalaisamy et al.24 propose the Averaged One-
Dependence Estimators (AODE) classifier and the SELECT Applicable Only to Parallel Server (SELECT-APSL
ASA) method for data separation. The AODE classifier, incorporated with hybrid data obfuscation, effectively
separates smart city data. Khadidos et al.25 developed a new security model for smart city networks, utilizing
the Collaborative Mutual Authentication (CMA) mechanism for user identity validation and the Meta-heuristic
Genetic Algorithm – Random Forest (MGA-RF) technique for attack detection, ensuring enhanced network
security. Xie et al.26 propose the Bidirectional Gated and Dynamic Fusion Network (BGDFNet) model for
RGB-T crowd counting. The model comprises a novel Bidirectional Gated Module (BGM) method to bridge
modality gaps and a Multiscale Dynamic Fusion (MSDF) module for effectual feature fusion. A crowd density
map is generated through the decoder and fully connected layer. John Pradeep et al.27 utilize a Multi-CNN
(MCNN) based on the VGG-19 architecture for crowd density estimation. The deeper VGG-19 model captures
complex features, improving crowd counting accuracy, while the MCNN handles varying image sizes and
viewpoint effects on head sizes. Kamra et al.28 explore ML models such as Mobilenet SSD, Yolov4, and Mask
RCNN for crowd density estimation. It examines integrating surveillance data, lidar, and thermal sensors in
real-time systems. It emphasizes big data analytics and cloud computing, focusing on future multimodal data
fusion and privacy preservation developments. Hussain et al.29 propose a novel DL model by integrating Graph
Convolutional and Spatio-temporal Gated Recurrent Units (GCST-GRU) to predict the next traffic state. The
model captures spatial dependencies using graph convolution and temporal dependencies through the GRU
neural network. Qaraqe et al.30 present PublicVision using a Swin Transformer-based DL model. A novel dataset
is used to identify crowd size and violence levels, with security ensured through a Dynamic Multipoint Virtual
Private Network (DMVPN), IPSec, and Firewall for data confidentiality and integrity. Sonthi et al.31 This research
reviews CNN-based approaches for gender and age identification, analyzing 30 publications to predict future
trends in CNN methods and transportation systems. It also evaluates two DL techniques, CNN and AlexNet, for
gender classification from facial images.
Vu et al.32 employ three CNN models, namely W-Net, UASD-Net (a fusion of U-Net and Adaptive Scenario
Discovery), and Congested Scene Recognition Network (CSR-Net) to assess traffic density from images captured
in Vietnam. A novel methodology is also introduced to reallocate label points to generate more accurate density
maps. Gündüz and Işık33 propose a real-time method for monitoring physical distancing in indoor spaces using
computer vision and DL. The technique employs the YOLO object detection model pre-trained on the Microsoft
COCO dataset. Alotaibi et al.34 propose the Osprey Optimization Algorithm with Deep Learning Assisted Crowd
Density Detection and Classification (OOADL-CDDC) technique for smart video surveillance to detect and
classify crowd densities using SE-DenseNet and attention bidirectional gated recurrent unit (ABiGRU) models.
Tripathy et al.35 propose a secure, reliable service delivery strategy for smart cities by integrating fog and mist
servers with an intrusion detection system. Alasiry and Qayyum36 integrate UOAL for video anomaly detection,
TempoNet for motion forecasting, and BDSO for resource optimization, improving accuracy, responsiveness,
and efficiency in surveillance. Lakhan et al.37 present a BC-based, energy-efficient IoE smart grid architecture
for smart homes and electric vehicles, integrating a BC-energy-efficient IoE sensors data scheduling (BEDS)
approach for sensor data scheduling and integrating fibre optics for data offloading. Gao et al.38 introduce a
Utility-driven Destination-aware Spatial Task Assignment (UDSTA) problem and solve it by utilizing a dual-
embedding-based deep Q-Network (DE-DQN) for efficient task-worker matching and route planning. An
enhanced DE-Rainbow version optimizes further using Rainbow DQN. Mohammed et al.39 introduce the
Pattern-Proof Malware Validation (PoPMV) approach for BC in ICPS, using LSTM and reinforcement learning
to improve security, processing speed, and attack detection. Ji et al.40 propose the Cloud-Edge Large-Tiny
Model Collaborative (CELTC) architecture, using the deep reinforcement learning (DRL) model to optimize
surveillance data processing on offshore drilling platforms by deploying large models on the cloud and tiny
models on edge devices. Jayasingh et al.41 integrate IoT devices and the YOLO model for real-time, accurate

Scientific Reports | (2025) 15:5759 | [Link] 4

[Link]/scientificreports/

crowd counting in public spaces. Sidharta et al.42 distinguish between group and non-group pedestrians using
semantic segmentation with directional-oriented density features and a shallow U-network architecture. Zhou
et al.43 propose the MJPNet-S* for drone-based RGB-T/D crowd density estimation, using a trimodal fusion
module and knowledge distillation to enhance accuracy and efficiency and reduce resource consumption.
While significant advancements have been made in integrating DL, IoT, and sensor technologies for smart
city applications, various limitations and research gaps remain. Many existing approaches, such as those using
CNNs and other ML models for crowd density estimation, mostly face challenges related to occlusions, varying
environmental conditions, and the requirement for large, annotated datasets. Moreover, despite efforts to
improve model accuracy, issues like real-time processing efficiency and system scalability for large-scale urban
environments still pose significant challenges. Additionally, while several techniques integrate multimodal
data fusion, attaining seamless integration of diverse data types (e.g., thermal, LiDAR, and video) without
compromising privacy remains a concern. The energy effectiveness of smart city systems, particularly those using
IoT, has not been sufficiently addressed, specifically in minimizing energy consumption during continuous data
transmission and analysis. Furthermore, existing models often lack robustness against dynamic changes in traffic
patterns or crowd behaviours, resulting in inaccurate predictions. Therefore, there is a requirement for more
adaptable and resilient models that can handle real-world complexities such as rapidly changing environments,
heterogeneous data, and privacy concerns.

Methods and techniques

This paper presents a novel DCNNCDM-IUP methodology for smart cities. The proposed method utilizes the
DL approach for detecting crowd densities, which helps to aid urban planning in smart cities. Figure 2 illustrates
the entire flow of the DCNNCDM-IUP technique.

Dataset
The DCNNCDM-IUP technique’s experimental evaluation is tested using a crowd-density dataset16. Table 1
depicts 1000 samples with four classes.

Image preprocessing
At the initial stage, the DCNNCDM-IUP technique undergoes image preprocessing and is performed using
the GF approach44. The key advantage of choosing the GF model over other methods is its ability to minimize
high-frequency noise without significantly blurring edges. Unlike median or mean filters, GFs apply a weighted

Fig. 2. Overall flow of DCNNCDM-IUP technique.

Scientific Reports | (2025) 15:5759 | [Link] 5

[Link]/scientificreports/

Labels Class names No. of instances

Class0 “Dense Crowd” 250
Class1 “Medium Dense Crowd” 250
Class2 “Sparse Crowd” 250
Class3 “No Crowd” 250
Total number of instances 1000

Table 1. Details on database.

Fig. 3. Working flow of the GF technique.

Fig. 4. Overall flow of SE-DenseNet approach.

average where nearby pixels have a higher influence, improving edge preservation. This technique particularly
benefits object detection and feature extraction applications, where clear boundaries are crucial. Moreover,
GF is computationally effectual, making it appropriate for real-time applications. Its adaptability in handling
various image noise levels makes it a robust choice for preprocessing in complex visual tasks. Figure 3 depicts
the workflow of the GF model.
GF is a basic image preprocessing method to mitigate noise and detail by image smoothing. This also employs
a GF, considered by its bell-shaped curve, to compute the weighted average of a pixel value within a defined
neighbourhood. The result is a blurred image where the variance of the Gaussian function controls the degree of
blurring. By allocating weights to the pixel closer to the centre of the neighbourhood, GF is utilized to preserve
edges better than other linear smoothing approaches, which makes it incredibly efficient for applications
requiring noise reduction while retaining structural data.

DL model: SE-DenseNet model

The DCNNCDM-IUP technique utilizes SE-DenseNet, which effectively learns complex feature patterns by
leveraging the power of dense connectivity and the SE mechanism45. DenseNet improves feature propagation
and reuse by connecting each layer to every other layer in a feedforward fashion, which assists in better gradient
flow and reduces the vanishing gradient problem. The SE module improves the model’s capability to selectively
emphasize crucial features by adaptively recalibrating channel-wise feature responses. This integration allows
SE-DenseNet to attain superior performance in tasks comprising high-level feature extraction, such as image
classification and object recognition. Compared to conventional CNNs or standalone DenseNet, SE-DenseNet’s
ability to adaptively focus on essential features confirms higher accuracy and robustness, particularly in complex
datasets. The model’s efficient use of parameters makes it computationally efficient without compromising
accuracy, presenting a balanced tradeoff for real-time applications. Figure 4 demonstrates the working flow of
the SE-DenseNet model.

Scientific Reports | (2025) 15:5759 | [Link] 6

[Link]/scientificreports/

The basic concept of the attention module is to train the model for autonomously learning the most crucial
channels within specific tasks. As a result, this technique considerably improves the attention model on the
feature, eventually producing precise and more efficient predictive or classification outcomes. The channel
attention mechanism (CAM), in particular, may help to focus further on salient channels during feature
processing, thus improving the model’s generalization capability and overall performance. This model holds
hidden value and immense application potential in many areas, namely natural language processing, image
processing, and voice recognition.
Generally, the typical CAM pattern is the squeeze-and‐excitation network (SENet). The primary function
is to decrease global spatial information. Consequently, the feature is learned across channels to ascertain each
channel’s significance. Finally, weight is allocated to each channel through the excitation phase. h and w are the
feature matrix dimensions, and c is the channel number. Lastly, the normalized weight is multiplied by the input
feature mapping to generate the weighted feature maps.
The SENet automatically learns feature weight according to loss through the FC network. The SENet optimizes
the channel weight rather than categorizing it based on the mathematical values of the feature channel. Then,
criticality is used to assign weight to the feature by obtaining the degree of importance for all the channels in the
feature mapping. Consequently, the NN focuses on specific feature channels through the SENet.
DenseNet is a CNN framework where all the layers take the output of each prior layer as an input. It helps the
backpropagation of the gradient during training and allows the DCNN model to be trained. Feature multiplexing
is a kind of DenseNet obtained by connecting features on the channel. DenseNet has an i(i + 1)/2 connection
for a NN with i layers. It exploits less computation to obtain superior outcomes. Each prior layer is connected.
The output at ith layers is given below:
Xi = Hi ([X0 , X1 , . . . , Xi−1 ])(1)

In Eq. (1), the quantity of network layers is i. The outcome corresponding to ith layer is Xi . Hi is a set of
nonlinear transformations, such as activation, batch normalization, pooling, and convolution.
In DenseNet, a small growth rate, Dense Block, and Transition Layer are utilized. It reduces the calculation
and narrows the network, efficiently suppressing overfitting problems. Classical CNN relies on pooling and
convolution to minimize the feature map dimensionality. In DenseNet, the layer is a dense connection,
necessitating the feature map retention of a similar dimension for forward propagation. Therefore, the Transition
Layer and Dense Block better solve these problems.
The feature vector is first zero-padded through the Zero‐Padding layer to control the feature vector lengths
and later passed through the convolutional layers. Decreased feature map dimensions enhance computation
efficacy. Moreover, the Transition Layer decreases the feature map dimension while reducing the feature number.

DL model optimization
Moreover, the hyperparameter selection of the SE-DenseNet approach is accomplished by employing the RFO
technique46. This model presents various advantages over conventional methods. RFO employs the power
of decision trees to handle the complex, nonlinear relationships between hyperparameters, making it highly
effective for optimizing models such as SE-DenseNet. It is robust against overfitting and can give more accurate
hyperparameter configurations by computing diverse feature interactions. RFO is more computationally efficient
and adaptive than grid or random search, as it concentrates on the most promising regions of the hyperparameter
space. Furthermore, RFO automatically handles high-dimensional optimization problems, making it ideal for
DL models with many hyperparameters like SE-DenseNet. This approach ensures the optimum configuration
for improved performance without excessive computational cost. Figure 5 illustrates the structure of the RFO
method.
The red fox is a predator that hunts farm and wild animals, and it is hunted by the humans who interrupt
it in this attack. Since the better individual survives to pursue the various territories, the relations between
individuals, predators, and prey in a similar pack are mathematically mimicked. Foxes have a hunting strategy
that is considered for both local and global searches. Along with these two stages, a population control mechanism
has been further suggested, which simulates the group leaving for a different place or being killed by the hunter.
These two better individuals are chosen to replace and reproduce the weaker component in all the iterations.
This allows us to achieve the global extrema, reducing the risk of being trapped in local extrema. The steps of the
proposed algorithm are given below:
Step 1—Initial Population: The population is generated by the ( number
)t of foxes selected, which is constant
in all the iterations. All the components are characterized as xji , where i is the individual identification
in the population, j is the coordinate in size of the solution space, and t denotes the existing iteration. The
individual that offers the better solution is recorded, creating the population of the initial iteration according to
the assessment of the population produced by the objective function (OF).
Step 2—Global Search: The pack members move towards the remote location, looking for food, and later,
the gathered data is given to other members to ensure their development and survival. Therefore, the optimum
individuals are recorded, and the Euclidean distance is evaluated as follows:
( ) √
d xi t , xbest t = || xi t − xbest t || (2)

Individual moves to the best red foxes using:

( )
xi t = xi t + α sign xbest t − xi t (3)

Scientific Reports | (2025) 15:5759 | [Link] 7

[Link]/scientificreports/

Fig. 5. RFO architecture.

Where
[ ( α is a random )] integer for each individual in all the iterations, which belongs to the range
0, d (xi )t , (xbest )t , the novel population is now assessed by the OF. If the optimum value is better, the
individual takes the newest position or returns to the initial location. This is simulated by the path the other
group members should follow once the information is transferred by the components that have searched the
surrounding territories, where they keep finding feasible locations. Also, the model is used to predict the
reproduction of optimum individuals or the death of the worst individual from the population.
Step 3—Local Search: Once the fox spots the target, they try to approach it slowly in a circular movement
until they are closer enough to catch it. This behaviour is modelled through the µ parameter. During the
approach movement, this is randomly produced within [0,1] and simulates the fox probability being realized by
the target. If µ > 0.75, the fox methods the target, or it leaves. The observation and approach movement of all
the individuals to the prey has been estimated by the modified Cochleoid equation, and φ 0 model the angle of
vision of all the individuals and is generated randomly within [0,2π ] for each individual.
{
a sin(ϕ0 )
rv = , ϕ0 =
̸ 0
ϕ0 (4)
θ, ϕ0 = 0

In Eq. (4), θ denotes the random integer within [0,1] and characterizes the impact of atmospheric conditions.
The movement is modelled by Eq. (14) for the spatial coordinates, where all the angular values φ n−1 are
randomly produced within [0,2π ] .

Scientific Reports | (2025) 15:5759 | [Link] 8

[Link]/scientificreports/

 0
 xnew = arccos (φ 1 ) + x0actual

 x1 = arcsin (φ 1 ) + arccos (φ 2 ) + x1actual

 new
2
xnew = arcsin (φ 1 ) + arcsin (φ 2 ) + arccos (φ 3 ) + x2actual
... (5)

 ∑ n−2 ( )


 xn−2
new = arc k=1
sin (φ k ) + arccos φ n−1 + xn−2
(
actual )
new = arcsin (φ 1 ) + arcsin (φ 2 ) + · · · + arcsin φ n−1 + xactual
n−l
xn−1

Step 4—Reproduction and Pack Abandonment: the worst individual in the population is chosen based on the
OF for the dynamic control of the population to introduce slight modifications to the pack. Since these are the
worst individuals, it is considered that they have been killed by the hunter or moved territory, but the worst
individual is replaced by a newer one to make sure that the population is constant. At all the iterations, the two
( )t ( )t
(1) (2)
better individuals (the alpha couple) are chosen, X and X , and the centre of diameter and habitat
are evaluated:
( )t ( )t
(1) (2)
X + X
(habitatcenire )t =
2

 t  t 

(habitat ) =  X̄ (1) − X̄ (2) 
diameter t
(6)

A random variable κ is produced in [0,1] that determines the individual’s replacement for all the iterations. If
κ ≥ 0.45, the new individual is presented into the population, or the alpha couple reproduces. Initially, new
member moves towards other areas, being capable of reproducing. Next, a new individual is the result of the
alpha couple reproduction as
( )t ( )t
(1) (2)
X + X (7)
(x(reproduced) )t = κ
2

Step 5—Evaluate all the individuals based on the OF— the best value xbest is recorded.
Algorithm 1 illustrates the RFO model.

Scientific Reports | (2025) 15:5759 | [Link] 9

[Link]/scientificreports/

Algorithm 1. RFO algorithm.

Scientific Reports | (2025) 15:5759 | [Link] 10

[Link]/scientificreports/

The RFO approach derives an FF to achieve better classifier outcomes. It describes a positive integer to
characterize the more significant outcomes of the solution candidate. Now, reducing the classifier error rate is
regarded as an FF.
f itness (xi ) = Classif ierErrorRate (xi )
N o. of misclassif ied samples
= × 100(8)
T otal N o. of samples

Classification method: ConvLSTM

Finally, the recognition of varied crowd densities is accomplished by the ConvLSTM47. This approach integrates
the merits of CNN and LSTM networks. This hybrid architecture allows ConvLSTM to capture spatial patterns
through CNN layers and effectually model temporal dependencies in crowd dynamics via LSTM units. Unlike
conventional CNNs, which concentrate solely on spatial features, ConvLSTM is specifically appropriate for
crowd density recognition as it handles sequential data and temporal changes, such as crowd movement over
time. This makes it more accurate in dynamic environments where crowd density fluctuates. Additionally,
ConvLSTM needs fewer parameters than entirely separate CNN and LSTM models, presenting computational
efficiency while maintaining high accuracy. Its capability to process video frames or time-series data makes it
superior for real-time crowd monitoring and analysis compared to other methods that may face difficulty with
temporal information. Figure 6 specifies the structure of the ConvLSTM technique.
As a traditional DL model, the DFN comprises an input, hidden layer (HL), and output layer. This is suitable
to find output values based on the input value. However, it is not applicable to predict the time series data. As
a result, an RNN is introduced to train the time series data. An RNN is a DL method where hidden nodes are
linked to a directed edge to form the cyclic structure known as a directed cycle. It helps predict data series, viz.,
text and voice. In the next step, the state of the prior time step is utilized, and the resultant values are affected
by the previous state. The LSTM among different RNN models is applied to prevent the vanishing gradients.
In addition, the LSTM is better than the DFN for understanding the features. Many DL models have been
introduced for training data time series. LSTM is more prevalent among them.
The cell state is a conveyor belt. Thus, the gradient of input values is relatively propagated well.
( )
ft = σ Wxf xt + Whf ht−1 + bf (9)

it = σ (Wxi xt + Whi ht−1 + bi )(10)

ot = σ (Wxo xt + Who ht−1 + bo )(11)
( )
gt = tanh Wxg xt + Whg ht−1 + bg (12)

ct = ft ⊙ ct−1 + it ⊙ gt (13)
ht = ot ⊙ tanh (ct )(14)

In the equations, σ and tanh are the sigmoid and hyperbolic tangent functions, W denotes the weight for all
the layers, xt indicates the input at the t time step , and b shows the bias. In Eqs. (13) and (14), ⊙ indicates the
Hadamard product operators. The forget gate ( ft ) forgets the prior data. The values are attained by taking the
sigmoid once it receives ht−1 , and the value xt denotes the forget gate transmits. The sigmoid output function
ranges within [0,1]. The data from the prior state is overlooked if the value is 0, and the data from the preceding
state is retained if it is 1. The input gate it ⊙ gt stores the existing data. It takes ht−1 and xt and exploits the

Fig. 6. Structure of ConvLSTM.

Scientific Reports | (2025) 15:5759 | [Link] 11

[Link]/scientificreports/

sigmoid function. Then, the values that take the Hadamard product and tanh functions are passed through the
input gate. Each represents the direction and intensity of storing existing data since it ranges from 0 to 1, and
the gt ranges from −1 to 1.
The 2D images (data) must be utilized as input to consider spatial connection in learning. 2D data have the
feature that increases the size; the data dimension is considerably augmented during training, viz., evaluating the
model weight. Thus, one of the DL models, a CNN considering the image features, may be utilized in the LSTM
model. The widely adopted strategy includes an LRCN, where the 2D feature vectors are extracted first through
the CNN and used as an LSTM input. Next, the input dataset is an image (2D). However, while 2D images from
various channels are used as input datasets, a 3D tensor is utilized as input, and the feature vector extracted by
the CNN undergoes a similar process within the LSTM. Thus, it is unfit for long-term data training. On the other
hand, the ConvLSTM exploits various approaches. The computation amount is reduced dramatically by carrying
out the internal operation of LSTM as a convolutional function.
This is mathematically modelled as follows:
( )
ft = σ WXf Xt + WHf Ht−1 + Wcf ⊙ Ct−1 + bf (15)

it = σ (WXi Xt + WHi Ht−1 + WCi ⊙ Ct−1 + bi )(16)

( )
gt = tanh WXg Xt + WHg Ht−1 + bg (17)

Ct = ft ⊙ Ct−1 + it ⊙ gt (18)
ot = σ (WXo Xt + WHo Ht−1 + WCo ⊙ Ct + bo )(19)
Ht = ot ⊙ tanh (Ct )(20)

When compared Eqs. (9)–(14) and Eqs. (15)–(20) are quite similar. But there are two important distinctions.
Firstly, the 3D tensor includes forget gate f , input gate i, output gate o, cell output C , cell input X , and cell
state H . Unlike the original LSTM, each element is a 1D vector. Next, matrix multiplication was implemented,
each replaced by a convolution operation. This implies that the weight existing in W might be lesser than
LSTM. This is similar to the FC layer replaced by the convolution layer; it reduces the number of weights and is
appropriate for effectively training an enormous quantity of data.

Experimental analysis and results

This section examines the experimental result analysis of the DCNNCDM-IUP technique. Figure 7 represents
the sample images.
Figure 8 reports a set of confusion matrices produced by the DCNNCDM-IUP method on several epochs.
On 500 epochs, the DCNNCDM-IUP method detected 246 samples in class 0, 213 in class 1, 234 in class 2,
and 240 in class 3. Likewise, on 1000 epochs, the DCNNCDM-IUP method detected 249 samples in class 0,
232 in class 1, 241 in class 2, and 241 in class 3. In addition, on 1500 epochs, the DCNNCDM-IUP method has
detected 250 samples in class 0, 232 in class 1, 244 in class 2, and 242 in class 3. Additionally, on 2000 epochs,
the DCNNCDM-IUP method detected 249 samples in class 0, 216 in class 1, 240 in class 2, and 244 in class 3.
Finally, on 3000 epochs, the DCNNCDM-IUP method detected 248 samples in class 0, 224 in class 1, 239 in
class 2, and 240 in class 3.
Table 2; Fig. 9 indicate the overall crowd classification outcomes of the DCNNCDM-IUP technique. The
results exemplify that the DCNNCDM-IUP technique correctly detected varied classes. With 500 epochs, the
DCNNCDM-IUP technique offers average accuy , precn , recal , Fscore , and Gmeasure of 96.65%, 93.38%,
93.30%, 93.26%, and 93.30%, correspondingly. Also, with 1000 epochs, the DCNNCDM-IUP model exhibits
average accuy , precn , recal , Fscore , and Gmeasure of 98.15%, 96.32%, 96.30%, 96.29%, and 96.30%,
correspondingly. Besides, with 1500 epochs, the DCNNCDM-IUP model exhibits average accuy , precn , recal
, Fscore , and Gmeasure of 98.40%, 96.82%, 96.80%, 96.79%, and 96.80%, correspondingly. Moreover, with 2000
epochs, the DCNNCDM-IUP model exhibits average accuy , precn , recal , Fscore , and Gmeasure of 97.45%,
95.03%, 94.90%, 94.85%, and 94.91%, correspondingly. Finally, with 3000 epochs, the DCNNCDM-IUP model
exhibits average accuy , precn , recal , Fscore , and Gmeasure of 97.55%, 95.12%, 95.10%, 95.08%, and 95.09%,
correspondingly.
Figure 10 illustrates the DCNNCDM-IUP method’s training and validation accuracy outcomes. The accuracy
values are calculated in the range of 0-3000 epochs. The figure emphasizes that the training and validation
accuracy values exhibit a rising tendency, which indicates the method’s ability to enhance performance over
various iterations. Furthermore, the training and validation accuracy remain closer over the epochs, exhibiting
improved performance and indicating low minimal overfitting of the DCNNCDM-IUP method, guaranteeing
consistent prediction on unseen samples.
Figure 11 shows the training and validation loss graph of the DCNNCDM-IUP technique. The loss values
are calculated in the range of 0-3000 epochs. It is represented that the training and validation accuracy values
validate a decreasing tendency, which notified the capability of the DCNNCDM-IUP method in balancing a
tradeoff between data fitting and generalization. The progressive reduction in loss values additionally guarantees
the enhanced performance of the DCNNCDM-IUP method and tunes the prediction results over time.
In Fig. 12, the precision-recall (PR) investigation of the DCNNCDM-IUP method under several epochs offers
an interpretation of its performance by plotting Precision against Recall for all the classes. The figure exhibits that
the DCNNCDM-IUP method constantly achieves enhanced PR values across various classes, representing its
ability to maintain a significant portion of true positive predictions amongst each positive prediction (precision)

Scientific Reports | (2025) 15:5759 | [Link] 12

[Link]/scientificreports/

Fig. 7. Sample images.

while capturing a large proportion of actual positives (recall). The continual increase in PR outcomes among all
classes depicts the efficiency of the DCNNCDM-IUP method in the classifier process.
In Fig. 13, the ROC curve of the DCNNCDM-IUP technique under various epochs is studied. The results
imply that the DCNNCDM-IUP method obtains superior ROC outcomes over all the classes, representing the
significant ability to discriminate the classes. This reliable trend of enhanced ROC values over different classes
signifies the proficient performance of the DCNNCDM-IUP technique in predicting classes, highlighting the
drastic nature of the classifier process.
To demonstrate the proficiency of the DCNNCDM-IUP method, a detailed comparison study is made in
Table 348.
Figure 14 provides comparative precn and recal results of the DCNNCDM-IUP method. The results
indicate that the BoW-SRP, BoW-LBP, and GLCM-SVM techniques have shown worse values of precn and
recal . Simultaneously, the GoogleNet and VGGNet models have attained slightly improved precn and recal .
Meanwhile, the MDTL-ICDDC and AICDA-SSC models have demonstrated closer values of precn and recal
. Nevertheless, the DCNNCDM-IUP technique results in enhanced performance with precn and recal of
96.82% and 96.80%, correspondingly.
Figure 15 provides a comparative accuy and Fscore results of the DCNNCDM-IUP model. The outcomes
specify that the BoW-SRP, BoW-LBP, and GLCM-SVM techniques have demonstrated worse values of accuy
and Fscore . Simultaneously, the GoogleNet and VGGNet models have obtained slightly improved accuy and
Fscore . Meanwhile, the MDTL-ICDDC and AICDA-SSC models have established closer values of accuy and
Fscore . Nevertheless, the DCNNCDM-IUP method leads to superior performance with accuy and Fscore of
98.40% and 96.79%, respectively.
Therefore, the DCNNCDM-IUP technique is applied to enhance the detection of crowd densities.
Table 4; Fig. 16 demonstrate the computational time (CT) analysis of the DCNNCDM-IUP approach with
existing methodologies. Among the listed methods, the DCNNCDM-IUP model demonstrates the fastest
detection time with a CT of 9.80 s, outperforming others significantly. In comparison, methods like BoW-

Scientific Reports | (2025) 15:5759 | [Link] 13

[Link]/scientificreports/

Fig. 8. Confusion matrices of DCNNCDM-IUP method (a−f) Epochs 500–3000.

SRP, Bow-LBP, and GLCM-SVM show higher detection times of 15.40s, 16.28s, and 17.50s, respectively. Other
advanced models, including GoogleNet, VGGNet, MDTL-ICDDC, and AICDA-SSC, take even longer, with
detection times ranging from 18.15s to 21.15s. This demonstrates the efficiency and speed of the DCNNCDM-
IUP model in cyberthreat detection compared to its counterparts.

Conclusion
In this paper, a novel DCNNCDM-IUP method was presented for smart cities. The presented DCNNCDM-IUP
method aimed to utilize the DL approach for detecting crowd densities, which, in turn, helps aid urban planning
in smart cities. Initially, the DCNNCDM-IUP technique performed image preprocessing using the GF approach.
The DCNNCDM-IUP technique utilized the SE-DenseNet method, which effectually learns complex feature
patterns for feature extraction. Moreover, the RFO system selected the hyperparameter of the SE-DenseNet
approach. Finally, the recognition of varied crowd densities was accomplished by the design of ConvLSTM.
A widespread simulation analysis was conducted to illustrate the enhanced performance of the DCNNCDM-
IUP technique. The experimental validation of the DCNNCDM-IUP technique portrayed a superior accuracy
value of 98.40% compared to existing DL models. The limitations of the DCNNCDM-IUP technique comprise
its reliance on a fixed population size and a static learning rate, which may not be optimal for dynamic or
complex problem spaces. Moreover, the model’s performance can degrade in high-dimensional problems, as it
may face difficulty efficiently exploring the entire solution space. The study does not address the computational
cost and scalability issues when applied to massive datasets. Future work may incorporate adaptive population
sizes, dynamic learning rates, and hybrid models to improve flexibility. Furthermore, exploring parallelization
or distributed computing strategies could improve scalability and reduce computation time. Additionally, more
robust evaluation metrics across various real-world applications provide deeper insights into the generalization
capabilities of the model.

Scientific Reports | (2025) 15:5759 | [Link] 14

[Link]/scientificreports/

Class Labels Accuy P recn Recal FScore GM easure

Epoch − 500
Class 0 98.60 96.09 98.40 97.23 97.24
Class 1 94.70 93.01 85.20 88.94 89.02
Class 2 97.30 95.51 93.60 94.55 94.55
Class 3 96.00 88.89 96.00 92.31 92.38
Average 96.65 93.38 93.30 93.26 93.30
Epoch − 1000
Class 0 99.10 96.89 99.60 98.22 98.23
Class 1 97.20 95.87 92.80 94.31 94.32
Class 2 98.70 98.37 96.40 97.37 97.38
Class 3 97.60 94.14 96.40 95.26 95.26
Average 98.15 96.32 96.30 96.29 96.30
Epoch − 1500
Class 0 99.30 97.28 100.00 98.62 98.63
Class 1 97.60 97.48 92.80 95.08 95.11
Class 2 98.90 97.99 97.60 97.80 97.80
Class 3 97.80 94.53 96.80 95.65 95.66
Average 98.40 96.82 96.80 96.79 96.80
Epoch − 2000
Class 0 99.00 96.51 99.60 98.03 98.04
Class 1 95.90 96.86 86.40 91.33 91.48
Class 2 98.10 96.39 96.00 96.19 96.19
Class 3 96.80 90.37 97.60 93.85 93.92
Average 97.45 95.03 94.90 94.85 94.91
Epoch − 2500
Class 0 99.30 97.65 99.60 98.61 98.62
Class 1 97.40 95.16 94.40 94.78 94.78
Class 2 98.60 98.36 96.00 97.17 97.17
Class 3 97.90 95.26 96.40 95.83 95.83
Average 98.30 96.61 96.60 96.60 96.60
Epoch − 3000
Class 0 99.00 96.88 99.20 98.02 98.03
Class 1 96.20 94.92 89.60 92.18 92.22
Class 2 98.00 96.37 95.60 95.98 95.98
Class 3 97.00 92.31 96.00 94.12 94.14
Average 97.55 95.12 95.10 95.08 95.09

Table 2. Crowd classification outcome of DCNNCDM-IUP technique under various epochs.

Scientific Reports | (2025) 15:5759 | [Link] 15

[Link]/scientificreports/

Fig. 9. Average of DCNNCDM-IUP technique under various epochs.

Scientific Reports | (2025) 15:5759 | [Link] 16

[Link]/scientificreports/

Fig. 10. Accuy curve of DCNNCDM-IUP technique (a–f) Epochs 500–3000

Scientific Reports | (2025) 15:5759 | [Link] 17

[Link]/scientificreports/

Fig. 11. Loss curve of DCNNCDM-IUP technique (a–f) Epochs 500–3000.

Scientific Reports | (2025) 15:5759 | [Link] 18

[Link]/scientificreports/

Fig. 12. PR curve of DCNNCDM-IUP method (a–f) Epochs 500–3000.

Scientific Reports | (2025) 15:5759 | [Link] 19

[Link]/scientificreports/

Fig. 13. ROC curve of DCNNCDM-IUP technique (a–f) Epochs 500–3000.

Methods P recn Recal Accuy FScore

BoW-SRP 68.33 67.85 80.40 67.88
Bow-LBP 74.68 74.15 84.04 74.35
GLCM-SVM 75.47 73.52 79.78 87.99
GoogleNet model 82.98 85.26 84.40 81.00
VGGNet classifier 86.14 82.78 84.75 84.95
MDTL-ICDDC 92.90 92.90 96.45 92.87
AICDA-SSC 94.93 94.90 97.45 94.88
DCNNCDM-IUP 96.82 96.80 98.40 96.79

Table 3. Comparative analysis of DCNNCDM-IUP technique with recent models48.

Scientific Reports | (2025) 15:5759 | [Link] 20

[Link]/scientificreports/

Fig. 14. P recn and recal analysis of DCNNCDM-IUP technique with recent models

Fig. 15. Accuy and Fscore analysis of DCNNCDM-IUP technique with recent models

Methods CT (sec)
BoW-SRP 15.40
Bow-LBP 16.28
GLCM-SVM 17.50
GoogleNet model 18.15
VGGNet classifier 19.10
MDTL-ICDDC 20.25
AICDA-SSC 21.15
DCNNCDM-IUP 9.80

Table 4. CT evaluation of DCNNCDM-IUP approach with existing models.

Scientific Reports | (2025) 15:5759 | [Link] 21

[Link]/scientificreports/

Fig. 16. CT evaluation of DCNNCDM-IUP approach with existing models.

Data availability
The datasets used and analyzed during the current study available from the corresponding author on reasonable
request.

Received: 12 July 2024; Accepted: 13 February 2025

References
1. Santana, J. R. et al. A privacy-aware crowd management system for smart cities and smart buildings. IEEE Access. 8, 135394–
135405 (2020).
2. Solmaz, G., Baranwal, P. & Cirillo, F. CountMeIn: Adaptive Crowd Estimation with Wi-Fi in Smart Cities. In 2022 IEEE
International Conference on Pervasive Computing and Communications (PerCom), 187–196 (Pisa, Italy, 2022).
3. Ahmed, I., Ahmad, M., Ahmad, A. & Jeon, G. IoT-based crowd monitoring system: Using SSD with transfer learning. Comput.
Electr. Eng. 93, 107226 (2021).
4. Yu, Q. et al. Intelligent visual-IoT-enabled real-time 3D visualization for autonomous crowd management. IEEE Wirel. Commun.,
28(4) 34–41 (2021).
5. Guastella, D. A., Campss, V. & Gleizes, M. P. A cooperative multiagent system for crowd sensing based estimation in smart cities.
IEEE Access. 8, 183051–183070 (2020).
6. Yang, Y., Yu, J., Wang, C. & Wen, J. Risk assessment of crowdgathering in urban open public spaces supported by spatio-temporal
big data. Sustainability 14(10), 6175 (2022).
7. Bai, L., Wu, C., Xie, F. & Wang, Y. Crowd density detection method based on crowd gathering mode and multi-column convolutional
neural network. Image Vis. Comput. 105, 104084 (2021).
8. Fan, Z. et al. A survey of crowd counting and density estimation based on convolutional neural network. Neurocomputing 472,
224–251 (2022).
9. Schweizer, P. Neutrosophy for physiological data compression: In particular by neural nets using deep learning. Int. J. Neutrosophic
Sci. 1(2), 74–80 (2020).
10. Fitwi, A., Chen, Y., Sun, H. & Harrod, R. Estimating interpersonal distance and crowd density with a single-edge camera.
Computers 10, 143 (2021).
11. Gazis, A. & Katsiri, E. Streamline intelligent crowd monitoring with iot cloud computing middleware. Sensors 24(11), 3643 (2024).
12. Guo, X. et al. Crowd counting in smart cities via lightweight ghost attention pyramid network. Future Gener. Comput. Syst. 147,
328–338 (2023).
13. Al Duhayyim, M. et al. I. and M., Aquila optimization with transfer learning based crowd density analysis for sustainable smart
cities. Appl. Sci. 12(21), 11187 (2022).
14. Zhai, W. et al. Scale Attentive Aggregation Network for Crowd Counting and Localization in Smart City (ACM Transactions on
Sensor Networks, 2024).
15. Zhai, W. et al. An attentive hierarchy ConvNet for crowd counting in smart city. Clust. Comput. 26(2), 1099–1111 (2023).
16. Alrowais, F. et al. M., Deep transfer learning enabled intelligent object detection for crowd density analysis on video surveillance
systems. [Link]. 12(13), 6665 (2022).
17. Lin, C. & Hu, X. Efficient crowd density estimation with edge intelligence via structural reparameterization and knowledge
transfer. Appl. Soft Comput. 111366 (2024).
18. Sharma, V. K., Mir, R. N. & Singh, C. Scale-aware CNN for crowd density estimation and crowd behavior analysis. Comput. Electr.
Eng. 106, 108569 (2023).
19. Selvarajan, S. et al. SCBC: Smart city monitoring with blockchain using internet of things for and neuro fuzzy procedures. Math.
Biosci. Eng. 20(12), 20828–20851 (2023).
20. Selvarajan, S., Manoharan, H., Khadidos, A. O., Khadidos, A. O. & Hasanin, T. Directive transportation in smart cities with line
connectivity at distinctive points using mode control algorithm. Sci. Rep. 14(1), 17938 (2024).
21. Selvarajan, S. & Manoharan, H. Digital Twin and IoT for Smart City Monitoring. In Learning Techniques for the Internet of Things
(131–151). Cham: Springer Nature Switzerland. (2023).
22. Selvarajan, S. et al. SCMC: Smart city measurement and control process for data security with data mining algorithms. Meas. Sens.
31, 100980 (2024).
23. Shitharth, S. et al. A conjugate self-organizing migration (CSOM) and reconciliate multi-agent Markov learning (RMML) based
cyborg intelligence mechanism for smart city security. Sci. Rep. 13(1), 15681 (2023).

Scientific Reports | (2025) 15:5759 | [Link] 22

[Link]/scientificreports/

24. Thirumalaisamy, M. et al. Interaction of secure cloud network and crowd computing for smart city data obfuscation. Sens. 22(19),
7169 (2022).
25. Khadidos, A. O. et al. An intelligent security framework based on collaborative mutual authentication model for smart city
networks. IEEE Access 10, 85289–85304 (2022).
26. Xie, Z. et al. Bgdfnet: bidirectional gated and dynamic fusion network for rgb-t crowd counting in smart city system. IEEE Trans.
Instrum. Meas. (2024).
27. John Pradeep, J., Raja Arumugham, M., Santhi, S. & Kalaiselvi, S. October. Multi-convolutional neural network based crowd
counting using VGG-19 architecture. In International Conference on Robotics, Control, Automation and Artificial Intelligence, 365–
378. (Springer Nature Singapore, Singapore, 2023).
28. Kamra, V., Vaishnav, A., Verma, A., Khan, R. & Singh, S. May. A Novel approach for crowd analysis and density estimation by using
machine learning techniques. In 2024 International Conference on Intelligent Systems for Cybersecurity (ISCS) 1–6. (IEEE, 2024).
29. Hussain, B., Afzal, M. K., Anjum, S., Rao, I. & Kim, B. S. A novel graph convolutional gated recurrent unit framework for network-
based traffic prediction. IEEE Access 11, 130102–130118 (2023).
30. Qaraqe, M. et al. PublicVision: A secure smart surveillance system for crowd behavior recognition. IEEE Access 12, 26474–26491
(2024).
31. Sonthi, V. K. et al. May. A Deep learning technique for smart gender classification system. In 2023 3rd International Conference on
Advance Computing and Innovative Technologies in Engineering (ICACITE) 983–987. (IEEE, 2023).
32. Vu, L. Q. P., Le Bui, B. D., Nguyen, M. T. T. & Pham, N. K. Estimating traffic density using convolutional neural networks based
on crowd counting. In International Conference on Future Data and Security Engineering 282–296. (Springer Nature Singapore,
Singapore, 2024).
33. Gündüz, M. Ş. & Işık, G. A new YOLO-based method for social distancing from real-time videos. Neural Comput. Appl. 35(21),
15261–15271 (2023).
34. Alotaibi, S. R. et al. Integrating explainable artificial intelligence with advanced deep learning model for crowd density estimation
in real-world surveillance systems. IEEE Access (2025).
35. Tripathy, S. S. et al. A secure mist-fog-assisted cooperative offloading framework for sustainable smart city development. Digit.
Commun. Netw. (2024).
36. Alasiry, A. & Qayyum, M. A smart vista-lite system for anomaly detection and motion prediction for video surveillance in vibrant
urban settings. J. Supercomput. 81(1), 1–49 (2025).
37. Lakhan, A. et al. BEDS: Blockchain energy efficient IoE sensors data scheduling for smart home and vehicle applications. Appl.
Energy 369, 123535 (2024).
38. Gao, Y. et al. A Dual-embedding based reinforcement learning scheme for task assignment problem in spatial crowdsourcing.
World Wide Web 28(1), 13 (2025).
39. Mohammed, M. A. et al. Securing healthcare data in industrial cyber-physical systems using combining deep learning and
blockchain technology. Eng. Appl. Artif. Intell. 129,107612. (2024).
40. Ji, X., Gong, F., Wang, N., Xu, J. & Yan, X. Cloud-edge collaborative service architecture with large-tiny models based on deep
reinforcement learning. IEEE Trans. Cloud Comput. (2025).
41. Jayasingh, S. K., Naik, P., Swain, S., Patra, K. J. & Kabat, M. R. Integrated crowd counting system utilizing IoT sensors, OpenCV and
YOLO models for accurate people density estimation in real-time environments. In 2024 1st International Conference on Cognitive,
Green and Ubiquitous Computing (IC-CGU) 1–6. (IEEE, 2024).
42. Sidharta, H. A., Al Kindhi, B., Mulyanto, E. & Purnomo, M. H. Semantic segmentation of pedestrian groups based on directional-
oriented density features with shallow U-net Architecture. Int. J. Intell. Eng. Syst., 18(1). (2025).
43. Zhou, W. et al. MJPNet-S*: Multistyle joint-perception network with knowledge distillation for drone RGB-thermal crowd density
estimation in smart cities. IEEE Internet Things J. (2024).
44. Nyemeesha, V. & Ismail, B. M. Implementation of noise and hair removals from dermoscopy images using hybrid gaussian filter.
Netw. Model. Anal. Health Inf. Bioinf. 10, 1–10 (2021).
45. Tang, J. et al. A SE-DenseNet-LSTM model for locomotion mode recognition in lower limb exoskeleton. PeerJ Comput. Sci. 10,
e1881 (2024).
46. Gaspar, J. S. D., Loja, M. A. R. & Barbosa, J. I. Metaheuristic optimization of functionally graded 2D and 3D discrete structures
using the red fox algorithm. J. Compos. Sci. 8(6), 205 (2024).
47. Kim, K. S., Lee, J. B., Roh, M. I., Han, K. M. & Lee, G. H. Prediction of ocean weather based on denoising autoencoder and
convolutional LSTM. J. Marine Sci. Engi. 8(10), 805 (2020).
48. Alsubai, S. et al. Design of artificial intelligence driven crowd density analysis for sustainable smart cities. IEEE Access (2024).

Acknowledgments
The authors extend their appreciation to the Deanship of Research and Graduate Studies at King Khalid Univer-
sity for funding this work through Large Research Project under grant number RGP2/267/45. Princess Nourah
bint Abdulrahman University Researchers Supporting Project number (PNURSP2025R330), Princess Nourah
bint Abdulrahman University, Riyadh, Saudi Arabia. Researchers Supporting Project number (RSPD2025R608),
King Saud University, Riyadh, Saudi Arabia. The authors extend their appreciation to the Deanship of Scientific
Research at Northern Border University, Arar, KSA for funding this research work through the project number
“NBU FFR-2025-2899-01.

Author contributions
Conceptualization: W.M. Data curation and Formal analysis: H.A. Investigation and Methodology: W.M., A.M.
Project administration and Resources: Supervision; M.A.A., M.A. Validation and Visualization: N.A., A.M.
Writing—original draft, W.M. Writing—review and editing, M.A.A., M.A. All authors have read and agreed to
the published version of the manuscript.

Declarations

Competing interests
The authors declare no competing interests.

Ethical approval
This article contains no studies with human participants performed by any authors.

Scientific Reports | (2025) 15:5759 | [Link] 23

[Link]/scientificreports/

Additional information
Correspondence and requests for materials should be addressed to M.A.A.
Reprints and permissions information is available at [Link]/reprints.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives
4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in
any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide
a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have
permission under this licence to share adapted material derived from this article or parts of it. The images or
other third party material in this article are included in the article’s Creative Commons licence, unless indicated
otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence
and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to
obtain permission directly from the copyright holder. To view a copy of this licence, visit ​h​t​t​p​:​/​/​cr​ ​e​a​t​i​v​e​c​o​mm
​ ​o​
n​s​.​o​r​g​/​li​ ​c​e​n​s​e​s​/​b​y-​ ​n​c​-​n​d​/​4​.​0​/​​​​.​​

© The Author(s) 2025

Scientific Reports | (2025) 15:5759 | [Link] 24