Lightweight Deepfake Detection Based on Multi-Feature Fusion

Abstract

Deepfake technology utilizes deep learning based face manipulation techniques to seamlessly replace faces in videos creating highly realistic but artificially generated content. Although this technology has beneficial applications in media and entertainment misuse of its capabilities may lead to serious risks including identity theft cyberbullying and false information. The integration of DL with visual cognition has resulted in important technological improvements particularly in addressing privacy risks caused by artificially generated deepfake images on digital media platforms. In this study we propose an efficient and lightweight method for detecting deepfake images and videos making it suitable for devices with limited computational resources. In order to reduce the computational burden usually associated with DL models our method integrates machine learning classifiers in combination with keyframing approaches and texture analysis. Moreover the features extracted with a histogram of oriented gradients (HOG) local binary pattern (LBP) and KAZE bands were integrated to evaluate using random forest extreme gradient boosting extra trees and support vector classifier algorithms. Our findings show a feature-level fusion of HOG LBP and KAZE features improves accuracy to 92% and 96% on FaceForensics++ and Celeb-DFv2 respectively.

Full text

Academic Editor:
Received: 20 December 2024
Revised: 10 February 2025
Accepted: 11 February 2025
Published:
Citation: Yasir, S.M.; Kim, H.
Lightweight Deepfake Detection
Based on Multi-Feature Fusion. Appl.
Sci. 2024,1, 0. https://doi.org/
Copyright: © 2024 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license
(https://creativecommons.org/
licenses/by/4.0/).
Article
Lightweight Deepfake Detection Based on Multi-Feature Fusion
Siddiqui Muhammad Yasir and Hyun Kim *
Department of Electrical and Information Engineering, Research Center for Electrical and Information Technology,
Seoul National University of Science and Technology, 232 Gongneung-ro, Nowon-gu,
Seoul 01811, Republic of Korea; siddiqui@seoultech.ac.kr
*Correspondence: hyunkim@seoultech.ac.kr
Abstract: Deepfake technology utilizes deep learning (DL)-based face manipulation tech-
niques to seamlessly replace faces in videos, creating highly realistic but artificially gen-
erated content. Although this technology has beneficial applications in media and en-
tertainment, misuse of its capabilities may lead to serious risks, including identity theft,
cyberbullying, and false information. The integration of DL with visual cognition has
resulted in important technological improvements, particularly in addressing privacy risks
caused by artificially generated “deepfake” images on digital media platforms. In this
study, we propose an efficient and lightweight method for detecting deepfake images and
videos, making it suitable for devices with limited computational resources. In order to
reduce the computational burden usually associated with DL models, our method inte-
grates machine learning classifiers in combination with keyframing approaches and texture
analysis. Moreover, the features extracted with a histogram of oriented gradients (HOG),
local binary pattern (LBP), and KAZE bands were integrated to evaluate using random
forest, extreme gradient boosting, extra trees, and support vector classifier algorithms. Our
findings show a feature-level fusion of HOG, LBP, and KAZE features improves accuracy
to 92% and 96% on FaceForensics++ and Celeb-DF(v2), respectively.
Keywords: deepfake detection; feature fusion; Histogram of Oriented Gradients (HOG);
Local Binary Pattern (LBP); KAZE descriptors
1. Introduction
The development of deepfake technologies presents significant challenges for visual
cognition in deep learning (DL) and raises serious concerns for visual information risks,
as these synthetic media convincingly manipulate visual content, spreading misinforma-
tion
[1–3]
. The manipulated content may raise concerns about the potential abuse of
technology and its consequences on politics, finance, and personal privacy [
2
,
4
]. Although
convolutional neural networks (CNNs) have achieved considerable success in computer vi-
sion tasks [
5
–
8
] including deepfake detection, alternative approaches that combine machine
learning and hand-made features are gaining popularity. In computer vision, technologies
such as auto-encoders and generative adversarial networks (GANs) facilitate the genera-
tion of manipulated visuals. Deepfake images are often categorized into facial synthesis,
attribute manipulation, identity swapping, and expression swapping, with CNN models
commonly employed for video detection [
4
,
9
]. To identify highly precise synthetic visual
data as deepfake, an effective deepfake detection method is required [
2
,
10
]. DL reduces
human effort in feature engineering but increases complexity and interpretability due to
high nonlinearity and input interactions. Traditional machine learning (ML) methods often
sacrifice accuracy for interpretability due to their complexity and large data volumes. DL
Appl. Sci. 2024,1, 0 https://doi.org/10.3390/app1010000
arXiv:2502.11763v1 [cs.CV] 17 Feb 2025

Appl. Sci. 2024,1, 0 2 of 17
methods are challenging to train and require computing resources, while ML methods are
easier to evaluate and understand [
11
,
12
]. To address these constraints, we encouraged to
experiment and evaluate traditional machine learning techniques to detect deepfakes.
Machine learning (ML) classifiers that use features such as local binary pattern
(LBP) [
13
] and KAZE descriptor [
14
] offer promising alternatives to CNN-based methods
for deepfake detection. LBP encodes texture information through local spatial patterns [
13
],
while KAZE provides robust, noise-invariant descriptors [
14
] and key points, allowing
the detection of minor deepfake artifacts that CNNs may overlook [
12
–
14
]. Traditional
models can be improved in accuracy and resilience by integrating machine learning feature
extraction methods [
15
]. Smaller datasets can benefit greatly from the use of LBP and KAZE
features, which increase detection process transparency [
16
,
17
]. To address the growing
difficulties presented by deepfake technology, our hybrid approach—which integrates
feature-based classifiers—offers a potential path ahead in differentiating real from fake
visual data [18].
Deepfake detection solutions are limited for social media analysis due to heavy com-
pression [
19
]. A compact model, based on ML classifiers, is needed for memory-constrained
devices like smartphones [
20
,
21
]. The proposed model, focusing on auto-encoder-based
generated videos and keyframe identification, achieves high accuracy with minimal com-
putational demands, making it suitable for memory-constrained devices and enhancing
deepfake detection capabilities [22,23].
This study aims to reduce computational costs in deepfake detection without substan-
tial accuracy loss. Our proposed model targets compressed social media videos, focusing
on the existing multi-feature fusion approach using multiple feature types (HOG, LBP, and
KAZE) for a more comprehensive representation of visual data. By analyzing variations
in visual artifacts, we achieve significant data reduction while preserving accuracy. The
method integrates well-known ML feature extraction methods with diverse texture features,
allowing effective training on limited datasets. The top three classifiers are presented in
Table 2. We evaluated the proposed fusion model using the Face Forensic++ [
24
] and Celeb-
DF [
25
] dataset, which replicates scenarios commonly found on social media platforms.
The use of FaceForensics++ and Celeb-DF datasets aligns with the common practice in the
field, allowing for a direct comparison with existing methods. The main contributions of
this work are summarized as follows:
•
The proposed fusion model introduces a novel approach to deepfake detection on
platforms with limited memory and processing capabilities, effectively managing
compressed video data;
•
Using existing classification techniques for artifact analysis, the method achieves
substantial data reduction while preserving detection accuracy;
•
The methodology combines forty established ML classifiers (using HOG, LBP, and
KAZE features) with diverse texture-based features, demonstrating reliable perfor-
mance even with limited datasets;
•
The evaluation primarily uses the Face Forensic++ dataset, which reflects real-world
scenarios and emphasizes minimizing computational overhead.
The remainder of this paper is organized as follows. Section 2examines the impact
of deepfake technology across different platforms. Section 3reviews lightweight feature
detection methods and details the architecture of the proposed fusion model. Section 4de-
scribes the dataset, experimental setup, and evaluation metrics used to assess the proposed
fusion model, as well as its limitations. Finally, Section 5summarizes the study findings
and outlines potential future directions.

Appl. Sci. 2024,1, 0 3 of 17
2. Related Works
Deepfakes have emerged as a critical challenge, prompting extensive research into
detection techniques. The DL-based methods have shown the most advancement, leading to
efficient detection systems. While various approaches have been proposed, they primarily
rely on similar underlying principles [
26
,
27
]. Most of the detection methods use CNN-
based models to classify images as fake or real, but state-of-the-art deepfake detectors (e.g.,
N. Bonettini [
28
]) still rely on complex neural networks, struggle with generalization to
unseen deepfake techniques, and lack robustness under real-world distortions [28,29].
Several deepfake detection approaches depend on various modalities and feature
fusion to improve accuracy. Prior research has shown that integrating spatial and frequency
domain features, as well as combining spatial, temporal, and spatiotemporal features,
significantly improves detection accuracy compared to single-modality approaches [
30
–
32
].
For instance, Almestekawy et al. [
33
] fused Facial Region Feature Descriptor (FFR-FD)
with random forest classifier and texture features (standard deviation, gradient domain,
and GLCM) fed into an SVM classifier. Raza et al. [
31
] proposed a three-stream network
utilizing temporal, spatial, and spatiotemporal features for deepfake detection. Moreover,
security techniques for deepfake detection on untrusted servers were introduced by Chen
B. et al. [
34
]; their method enables distant servers to detect deepfake videos without
understanding the content.
Proper methods are essential for extracting valuable information from large unpro-
cessed visual data, with feature-based techniques like LBP and KAZE offering com-
putational efficiency as an alternative to resource-intensive CNNs [
12
]. Recent studies
have suggested that combining extracted features with advanced ML classifiers can de-
velop hybrid models for deepfake detection while maintaining robustness across diverse
datasets [12,15,35].
Alternatively, texture can be encoded by comparing each pixel with its neighbors,
creating a binary pattern that serves as a robust feature descriptor across various lighting
conditions. Feature extraction techniques are divided into global and local descriptor
approaches [
36
]. Global methods analyze the entire image to generate a feature vector
and are considered fast processing but have some limitations, such as Principle Compo-
nent Analysis (PCA) [
37
], Linear Discriminant Analysis (LDA) [
38
], and Global Gabor
generic features [
39
]. Local descriptors, like LBP [
40
] and Histogram of Oriented Gradients
(HOG) [
41
], provide a more effective representation of images. LBP is widely used in
face recognition [
42
], while HOG is used for human detection by dividing the image into
fixed-size blocks and computing HOG features for each block. Likewise, the selection of
custom features (Local Binary Patterns (LBP) based on texture and a customized High-
Resolution Network (HRNet)) proposed by Khalil et al. [
43
] and fed to the SVM classifier.
This efficiency makes LBP a popular choice in tasks where texture details are important,
such as facial recognition and expression analysis, while also reducing processing time
and computational costs [
44
]. Deepfake artifacts regularly change gradient orientations
and edge patterns, which are essential for lightweight detection on resource-constrained
devices. Compared to CNN-based approaches, it is less successful in detecting higher-level
semantic discrepancies [
45
]. KAZE, on the other hand, can detect unique key points that
are invariant to noise and transformations, which is essential for applications requiring
high-fidelity feature matching under variable conditions. By detecting and characterizing
two-dimensional features in nonlinear scale space, the KAZE features [
14
] resist Gaussian
blurring. KAZE’s reliance on nonlinear diffusion allows it to capture image structures
that are often missed by traditional linear approaches, enhancing performance in complex
environments [46].

Appl. Sci. 2024,1, 0 4 of 17
More recent deepfake methods, particularly diffusion models, have introduced high-
quality synthetic images that closely resemble natural visuals, evading common detection
markers such as GAN-related grid artifacts [
47
]. Chen Y et al. and Yuan et al. [
48
,
49
]
developed a model that uses a reference image and text prompt to generate deepfake
images as human identity. These developments initiate a shift in detection strategies, where
integrating extracted features with classifiers holds significant potential for improving
accuracy and reducing computational load [10,27,33,50,51].
As a result, detecting deepfake images/videos contributes to the struggle against
spreading false information and encourages preserving the validity of visual content and
privacy. Our methodology differs from previous approaches in numerous important ways,
including the use of multi-feature-level fusion (HOG, LBP, and KAZE features) prior to
classification. Focus on characteristics that are computationally efficient. For validation,
supervised ML classifiers (such as support vector machines (SVMs), random forest (RF),
and gradient boosting classifiers) were used, and their performance in deepfake detection
has been evaluated.
3. Proposed Fusion Model
In machine learning, a feature refers to a specific, measurable attribute of an image
that helps in distinguishing patterns. This study focuses on the integration of two types of
features (local descriptors) obtained from the LBP and HOG with KAZE features before
the classification. To reduce computing costs and meaningful results, the detection of
important frames and the elimination of insufficient frames are necessary. In the first step,
the keyframes are extracted from videos within an interval of 0.5 s. The first and last 10
frames are overlooked (only if necessary) because they usually have information about the
introduction or credits that are not directly related to deepfake detection preprocessing.
Additionally, to identify the important keyframes, a similarity check between frames is used
as the criterion. Various threshold values are used to determine different approaches. This
part of the algorithm results in a pool of distinctive images of frames, which are available
for feature extraction. The images were resized to a 28 × 28 single-channel format, ensuring
a standardized input for processing. The research explores the use of frames extracted
from video footage or standalone images, treating keyframes as textured representations.
In the second step of feature extraction, the LBP, HOG, and KASE techniques are applied
separately, which typically generates histograms. Furthermore, the fusion of LBP and
KAZE and that of HOG and KAZE are used to input futures for the Extra Trees, Random
Forest, Support Vector, and XGB classifiers. The proposed fusion model aims to improve
detection accuracy while maintaining efficiency.
Figure 1illustrates a general abstract of the proposed fusion model with important key
concepts involved as data preprocessing (extracting keyframes), extracting features from
keyframes using LBP and HOG feature extraction methods, feature-level fusion with KAZE
features, classifier selection of the combined features, and subsequent classification on the
basis of the chosen classifiers. The keyframes are converted to a logarithmic scale and
divided into multiple bands to capture localized information about the texture patterns. We
analyze each band of these frame divisions by calculating the normalized histogram of HOG
or LBP features to pinpoint characteristics using our classifiers. After these histograms are
combined to form an LBP feature vector, they are combined with KAZE features to create
supporting feature sets. To handle the dimensionality caused by merging features at the
level of characteristics, various classifiers are utilized to identify features, with importance
scores that are then fed into the ultimate classifier.
The importance and understanding of feature extraction, along with its configurations,
are explained in the following section.

Appl. Sci. 2024,1, 0 5 of 17
Figure 1. General abstract of the proposed feature-level fusion method.
3.1. LBP Features
The LBP is an advanced technique for extracting features from images for texture
study due to its efficient handling of value variations and straightforward computational
process [
52
]. LBP sets thresholds for neighboring pixels, enabling accurate spatial pattern
extraction from images and transforming textual information into binary data for classifica-
tion and detection [
53
]. LBP analyzes every pixel in an image by evaluating the relationship
between each pixel and its surrounding pixels within a specified radius
R
distance away
from it. If the neighboring pixel value exceeds that of the pixel level, a binary bit is assigned
as 1; otherwise, it is marked as 0 [54].
Given a grayscale image
I
of size
M×N
, the LBP feature for each pixel
(x
,
y)
is
computed via the following formula:
LBP(xp,yp) = 


1 if I(xp,yp)≥I(x,y)
0otherwise (1)
For a pixel
(x
,
y)
, we compare its intensity
I(x
,
y)
with the intensities of its
P
neighbor-
ing pixels on a circle of radius
R
. Let the intensities of these neighbors be
{I(xp
,
yp)}PP=1
. A
binary value is assigned to each neighboring pixel. These binary values are concatenated to
form a binary number, which is then converted to a decimal value. Compute the histogram
of these LBP values over the entire image.
HLBP(k) =
M
∑
x=1
N
∑
y=1
δ(LBP(x,y),k),k∈ {0, 1, . . . , 2P−1}
where
(a
,
b)
is the Kronecker delta function, which is 1 if
a=b
and 0 otherwise. Finally, the
histogram is normalized with a small constant to prevent division by zero.
The calculation used in this study is explained as follows: for a given pixel
(xp
,
yp)
,
the intensity
I(pi)
in the center of the
(
3
×
3
)
block is computed by comparing
xp
to its 8
neighboring pixels. The texture classification process relies on illumination, translation,
and rotational variance, while keyframes lack control over these attributes, focusing instead
on uniform pattern representation. A uniform pattern,
LBP(xp
,
yp)
, is more suitable.
Preliminary experiments showed
P=
12 and
R=
2 provide the best performance for the
feature descriptor. With regard to categorizing textures on the basis of their patterns in
images or videos, the way the pattern is perceived is influenced by factors such as lighting
changes, shifting positions, and orientations of the texture details. However, when we focus
on moments in a sequence, the patterns are not affected by rotations or translations. Instead,
they are determined by how colors and contrasts spread out over different frequencies
and points in time, creating a consistent pattern overall. Therefore, using patterns such
as
LBP(xp
,
yp)
is more suitable for detecting deepfake keyframes. To evaluate this, we
conducted some experiments where we tried different values for the radius
R
as well as
the neighboring pixel
P
. After conducting our analysis, we determined that the values of
P=12 and R=2 are suitable for the feature descriptor [54].

Appl. Sci. 2024,1, 0 6 of 17
These comparisons provide a binary vector that represents the connection between the
intensity of one pixel and its neighbors. LBPs are widely utilized intensity-based features
in various domains, including human detection [
55
,
56
], facial recognition [
55
], background
subtraction, and textured surface recognition [
57
]. The LBP operator is particularly appeal-
ing because of its computational simplicity [
58
]. However, a notable limitation of the LBP
method is the extensive number of histogram bins needed, which reduces its efficiency for
localized image patches. Despite this drawback, LBP remains effective for global image
representations, making it a valuable tool for image classification tasks.
3.2. HOG Features
The HOG is a feature descriptor widely used for texture analysis and object detec-
tion [
59
]. This method is particularly suitable for tasks requiring robust edge and gradient-
based analysis, such as detecting structural inconsistencies in deepfake images. HOG
divides an image into smaller spatial regions, known as cells, and computes a histogram
of gradient directions within each cell. This process can be summarized into three steps:
gradient calculation, cell histogram generation, and feature vector construction:
•
Gradient Calculation: For each pixel in the image, the gradients along the
x
- and
y-axes are calculated using Sobel filters:
Gx=I(x+1, y)−I(x−1, y),Gy=I(x,y+1)−I(x,y−1)
The magnitude Mand direction θof the gradient are computed as:
M=qG2
x+G2
y,θ=arctanGy
Gx
•
Cell Histogram Generation: The gradient magnitudes
M
are binned into orientation
histograms, where the direction
θ
is quantized into a fixed number of bins (e.g., 9 bins
for 0°–180° or 18 bins for 0°–360°). To improve invariance to illumination and contrast
changes, the histograms are normalized within overlapping spatial blocks. Given a
block B, normalization can be performed as:
HOGnorm(B) = HOG(B)
p∥HOG(B)∥2+ϵ
where ϵis a small constant to prevent division by zero.
•
Feature Vector Construction: The normalized histograms obtained from all the blocks
are concatenated to form a single feature vector representing the image. HOG captures
fine-grained details about edge orientations and their distribution, making it suitable
for identifying subtle spatial distortions caused by deepfake manipulations.
In this study, the following HOG parameters were used:
• Cell Size: 8 ×8 pixels;
• Block Size: 2 ×2 cells;
• Number of Orientation Bins: 9 (0°–180°);
• Step Size: 50% overlap between blocks.
HOG is an effective choice for resource-constrained deepfake detection due to these
parameters, which maintain a balance between descriptive strength and processing effi-
ciency. HOG can be used in combination with other robust feature descriptors, such as
KAZE, to improve its sensitivity to high-level semantic adjustments to improve deepfake
detection.

Appl. Sci. 2024,1, 0 7 of 17
3.3. KAZE Features
The KAZE features are computed to capture the multi-scale and nonlinear structure
of the keyframes. The KAZE algorithm involves detecting keypoints and computing
descriptors. This process is involved by applying nonlinear diffusion filtering to the
keyframe
I
to create a nonlinear scale space. The keypoints are detected with the KAZE
detector using the formula:
{I(xi
,
yi)}K(i=
1
)
. For each keypoint
(xi
,
yi)
, compute a
descriptor vector
di
that represents the local image patch around the keypoint. Concatenate
the descriptor vectors into a single feature vector
D
. If the total number of features exceeds
a predefined length, the feature vector is truncated or padded as follows:
D= [d1,d2, ..., dk](1:m)
where
m
is the designed length of the KAZE feature vector. Finally, computed descriptors
are used for each key point. KAZE descriptors are computed by sampling the responses of
the nonlinear scale space at keypoint locations using orientation and scale information. The
extracted features are later used to classify the video as either fake or real. This classification
is accomplished via ML classifiers with KAZE robustness in extracting image features to
detect deepfakes precisely.
3.4. Proposed Feature Fusion and Classification
The process of combining LBPs and KAZE features for image classification involves
extracting two distinct sets of features from a single image, merging these feature sets into
a unified feature vector, and then using this combined vector to train a classifier. This
procedure improves classification performance, particularly for detecting deepfake content
by utilizing the included strengths of KAZE (keypoint detection and description) and LBP
(texture analysis).
The proposed fusion model is a comprehensive method for image classification that
integrates LBP or HOG and KAZE features, followed by the selected classifier. Initially, the
algorithm extracts LBP or HOG features, which capture texture features with the distribu-
tion of binary patterns in the neighborhood of each pixel. Mathematical representations
(Equations 2and 3) by the LBP histogram are used to normalize and achieve uniform
feature scaling. Concurrently, KAZE features are extracted by detecting key points and
computing their descriptors, effectively capturing local invariant features. These descrip-
tors are integrated into a single vector, which is then truncated or padded to maintain
a consistent feature length. The LBP and KAZE feature vectors are combined to create
a single feature representation for each image, utilizing the improved strengths of both
feature extraction methods.
The process of merging the LBP and KAZE features is explained in these steps. First,
we extract the LBP feature vector
FLBP
from the histogram
HLBP
by employing the follow-
ing formula:
FLBP =HLBP(0),HLBP(1), . . . , HLBP (2P−1)(2)
Second, the KAZE feature vector
FK AZE
is extracted from the concatenated descriptor
vectors Dby FK AZE =D.
Finally, the FLBP and FKAZE feature vectors are concatenated as follows:
Fcombined = [FLBP,FKAZE ](3)
The integration of LBP and KAZE features improves the algorithm’s robustness and
accuracy in deepfake classification tasks, especially when detecting false or real images.
LBP features improve in detecting texture patterns, which are important in distinguishing

Appl. Sci. 2024,1, 0 8 of 17
between real and fake images because deepfakes often possess irregular or inconsistent
textures. In contrast, KAZE features retrieve fine-grained keypoints, which is necessary
for detecting minor modifications that may not greatly alter texture but have an impact on
the structural integrity of the image. The proposed fusion model improves the classifier’s
capability to identify deepfakes by integrating these two feature sets to provide a deeper
and more discriminative feature space. In experiments utilizing fake images, the improved
detection capabilities and decreased false positives demonstrate how the integration of
texture-based and keypoint-based features results in the improvement of classification
accuracy. The proposed fusion model improves performance on deepfake detection tasks
by utilizing the advantages of KAZE (robust keypoint descriptors) and LBP (sensitive to
textures). Furthermore, Algorithm 1is expressed to provide a more detailed explanation
from the perspective of implementation.
Algorithm 1 Algorithm for merging LBP/HOG and KAZE features and classification
Require: Set of images {I1,I2, . . . , IN}, corresponding labels {y1,y2, . . . , yN}.
Ensure: Classification accuracy.
1: function EXTRACTLBPFEATURES(I,R=3, P=24)
2: Convert image Ito LBP and HOG using radius Rand Ppoints.
3: return ˆ
HLBP/HOG
4: end function
5: function EXTRACTKAZEFEATURES(I,M=64)
6: Detect keypoints in Iusing KAZE.
7: Compute descriptors for each keypoint and Concatenate descriptors into D.
8: return D
9: end function
10: function COMBI NE FEATURES(FLBP/HOG,FK AZE)
11: return [FLBP/HOG ,FKAZE ]
12: end function
13: function PREPAREDATAS ET({P1,P2, . . . , PN},{y1,y2, . . . , yN})
14: for each path Piin {P1,P2, . . . , PN}do
15: Load image Ifrom path Pand convert it to grayscale.
16: F(i)
LBP/HOG ←EXTRACTLBP/HOGFEATURES(Ii)
17: F(i)
KA ZE ←EXTRACTKAZEFEATURES(Ii)
18: F(i)
combined ←COMB IN EFEATURES(F(i)
LBP/HOG ,F(i)
KA ZE )
19: Append F(i)
combined to features list.
20: end for
21: return feature matrix Xand label vector y
22: end function
23: function TRAINANDEVALUATE(X,y)
24:
Split & Train classifier on the training set. Predict, Compute accuracy of predictions.
25: return accuracy
26: end function
27: Main Program
28: X,y←PREPAREDATAS ET({P1,P2, . . . , PN},{y1,y2, . . . , yN})
29: accuracy ←TRAINANDEVALUATE(X,y)

Appl. Sci. 2024,1, 0 9 of 17
4. Implementation
4.1. Experimental Design
This section summarizes the experimental details and results of deepfake datasets
for FaceForensic++ [
24
] and Celeb-DF [
25
]. The robustness of the fusion (LBP, HOG, and
KAZE) features under various classifiers is evaluated. We applied some preprocessing on
the raw data prior to experimentation with the datasets. The complete video sequence is
not taken into account. Instead, as described in Section 3, some keyframes are extracted.
Both fake and real videos are included in the video dataset. After being extracted, the
frames were placed in a folder with the appropriate name. The NVIDIA 3090 GPU was
used for feature extraction and the development of ML algorithms.
FaceForensics++ [
24
] is the popular publicly available forensic dataset that includes
1000 original video sequences that have been manipulated using four distinct face ma-
nipulation techniques: deepfake, Face2Face, FaceSwap, and NeuralTextures. The dataset
consists of 977 YouTube videos with 48,431 face counts and a data size of 575 mb, all of
which feature a clearly visible face, allowing automated tampering methods to produce
highly accurate forgeries (an example of the dataset is depicted in Figure 2). We targeted
deepfake videos and their original equivalents for our experiment. The dataset was created
by extracting keyframes from several videos. Following preprocessing, there were 2946 fake
images and 2930 real images in the training set. There were 198 real and 197 fake images in
the validation.
Figure 2. An example of fake faces from the FaceForensics++ dataset. The pristine image is in
the first column, whereas the forged images produced by DeepFakes, Face2Face, FaceSwap, and
NeuralTextures are in the second through fifth columns [24].
The Celeb-DF [
25
] dataset is divided into real and fake video/frames, where the real
videos are 590 original YouTube videos with people of all ages and the fake videos are
5639 deepfakes. Keyframes from these videos were extracted to create the dataset, making
sure that the images mostly featured the faces of different celebrities. The dataset consists
of 100 genuine images, where 900 deepfake images make up the validation set and 1130
real images and 8022 deepfake images make up the training set. A balanced evaluation
framework for deepfake detection models is provided by the test set, which consists of 340
deepfake images and 178 real images.
There are two classes in the Celeb-DF dataset: real and fake. The percentage of fake
classes is higher than that of real classes. The FaceForensics++ dataset has five classes,
including 1000 videos in each class. The original video is the first class, and the other
four classes are videos that have been altered/are fake. These five classes were reduced to
binary classes in equal parts for the purposes of this study. To remove the dataset imbalance
problem, a similar process [
60
] is adopted, where 800 videos were finalized from each

Appl. Sci. 2024,1, 0 10 of 17
dataset, 400 of which were chosen for each class. Following the average number of frames
prepared for the experiment, the detailed dataset is displayed in Table 1.
Table 1. FaceForensics++ and Celeb-DF dataset compositions.
Dataset Real Images Fake Images
Celeb-DF 382 346
FaceForensics++ 496 458
4.2. Evaluation Criteria
Empirical benchmarking is a popular way to accurately analyze feature extraction
and training times. This involves directly quantifying the time spent throughout experi-
mental runs, resulting in exact and dependable data. This method is especially useful for
machine learning tasks, where computational complexity varies depending on dataset size,
hardware capabilities, and specific implementation choices. In this paper, we describe the
methodological approach used to calculate feature extraction, training, and inference times
for ML classifiers, providing a thorough assessment of computing efficiency.
The time required for feature extraction can be computed as follows: for each feature
extraction method, the extraction process for all the data points in the dataset is applied.
The start and end times
Tf eatu re =Tend −Tstart
are recorded for each feature extraction
run [
61
]. To mitigate variability owing to hardware or background processes, the feature
extraction process is repeated multiple times, and the average time is computed as follows:
AverageFeatureExtractionTime =∑n
i=1Tfeature,i
n(4)
where
T f eature,i
denotes the feature extraction time for the
i−th
run and
n
is the number
of runs. For reporting purposes, the time per data instance (e.g., per frame in video
processing) can also be computed as follows:
Tf eatu re,instance =Tf eatu re
N(5)
where
N
is the total number of data instances. Training time refers to the duration required
to train an ML model on a specified dataset. For each classifier (e.g., RF, SVM, and CNN),
the training process is initiated and calculated by recording the start and end times as
follows:
Ttrain =Tend −Tstart
[
62
]. Similar to feature extraction, it is often beneficial to
perform multiple runs and compute the average through the following equation to obtain
a reliable estimate:
AverageTrainingTime =∑n
i=1Ttrain,i
n(6)
For larger datasets, the training time may also be approximated on the basis of model
complexity. For example, the training time complexity of RF with
N
trees is generally
O(NlogN)
, whereas the support vector classifier may exhibit
O(N2)
complexity. The
inference time is the duration required to classify a new instance after training. The total in-
ference time over a dataset
Tin f erence
can be approximated by
Tin f erence =Tin f erenc e,instancex N
,
where
N
is the total number of instances in the test dataset. Multiple test runs were car-
ried out to establish temporal consistency among approaches, with start and end times
carefully documented. Measuring the time necessary for each instance provides for more
detailed comparisons and brings out performance differences more clearly. This established
approach makes sure that all important time calculations for feature extraction, training,
and inference are directly comparable, resulting in a rigorous and repeatable experimental
framework.

Appl. Sci. 2024,1, 0 11 of 17
4.3. Results and Discussion
The results are further provided and analyzed in depth. This study proposed a feature-
level fusion of LBP, HOG, and KAZE features for classification via the FaceForeensics++
and Celeb-DF datasets. The evaluation of the results on the basis of the provided validation
is presented below: Table 2presents the classification accuracies obtained using various
classifiers with various feature sets, including LBP alone, KAZE alone, and the fusion of
LBP, HOG, and KAZE features. The experiment was conducted on the FaceForensics++ and
Celeb-DF datasets to evaluate the effectiveness of these features in distinguishing between
genuine and manipulated visual content/deepfake.
Table 2. Classification accuracy of fusion of LBP, HOG, and KAZE features with FaceForensic++.
Fusion of Features with Classifiers Accuracy
LBP Features Extra Trees Classifier 71.22%
RF Classifier 70.76%
KAZE Feature
Extra Trees Classifier 85%
Support Vector Classifier 86.12%
RF Classifier 75.70%
XGB Classifier 65.29%
HOG + KAZE Feature
Extra Trees Classifier 91%
Support Vector Classifier 92.12%
RF Classifier 85.70%
XGB Classifier 83.19%
LBP + KAZE Features
Extra Trees Classifier 82.61%
Support Vector Classifier 86.22%
RF Classifier 85.54%
XGB Classifier 88.56%
Analyzing the results shows that both LBP and KAZE features individually perform
well across different classifiers when tested on the FaceForensics++ dataset. LBP features
demonstrated their effectiveness in texture-based analysis, achieving an accuracy of 71.22%
with Extra Trees and 70.76% with Random Forest classifiers. Similarly, KAZE features,
which focus on detecting structural alterations and keypoint variations in images, pro-
duced accuracies ranging from 75.70% with Random Forest to 86.12% with a Support
Vector Classifier.
HOG features also showed strong performance, with accuracies between 85.76% and
92.12% using a Support Vector Classifier. This result is close to the benchmark accuracy of
94.44% for the FaceForensics++ dataset, which was achieved by EfficientNet [
28
]. These
findings highlight the potential of integrating feature extraction techniques to improve
deepfake detection performance.
However, the most notable results were achieved by the fusion of HOG and KAZE
features, demonstrating a clear advantage over individual feature sets. This fused approach
showed superior performance across all classifiers, with accuracy rates of 91.12% using Ex-
tra Trees, 92.12% with the Support Vector Classifier, and an impressive 94.44% when tested
with the state-of-the-art EfficientNet. These results indicate that integrating texture-based
(i.e., HOG) and keypoint-based (i.e., KAZE) features significantly improved the model’s
ability to detect deepfake content. These features allow for more effective detection of
deepfakes compared to using either method alone (see Figure 3for a detailed visualization).

Appl. Sci. 2024,1, 0 12 of 17
Future research could focus on strengthening the fusion technique of HOG and KAZE
features to improve feature selection and reduce dimensionality. Another promising di-
rection is the exploration of DL architectures, such as CNNs, to gain deeper insights into
hierarchical feature representations and further boost classification accuracy. Additionally,
evaluating the proposed fusion model on larger and more diverse datasets other than
FaceForensics++ would help assess its robustness in real-world scenarios. These advance-
ments would play an important role in strengthening deepfake detection, particularly in
addressing the growing challenges in digital manipulation and deepfake technologies.
Figure 3. Outcome of the proposed and ML algorithms in terms of accuracy (i.e., RF, extra trees,
and SVC).
The results presented in Table 3highlight the classification accuracy of different
feature extraction techniques when applied to deepfake detection using the Celeb-DF
dataset. The findings demonstrate that combining multiple feature descriptors improves
detection performance compared to using individual features alone. Among the single-
feature approaches, LBP features with a Support Vector Classifier (SVC) achieved the
highest accuracy (72%), which yielded a lower accuracy of 68%. This suggests that LBP
is more effective in capturing local texture variations relevant to distinguishing real from
fake images. When feature fusion was applied, the combination of HOG and KAZE
features achieved the highest accuracy (78%), showing a significant improvement over
individual features. This indicates that integrating gradient-based descriptors (HOG) with
keypoint-based features (KAZE) provides a more comprehensive representation of image
characteristics, improving classification reliability. Similarly, LBP combined with KAZE
achieved an accuracy of 75%, further confirming that KAZE features contribute positively
to deepfake detection by enhancing feature diversity.

Appl. Sci. 2024,1, 0 13 of 17
Table 3. Classification accuracy of fusion of LBP, HOG, and KAZE features with Celeb-DF.
Fusion of Features with Classifier Accuracy
LBP Features Support Vector Classifier 72%
HOG Features Support Vector Classifier 68%
HOG + KAZE Features Support Vector Classifier 78%
LBP + KAZE Features Support Vector Classifier 75%
Overall, these results demonstrate that fusion strategies, particularly HOG + KAZE,
are more effective than single-feature approaches in deepfake detection on the Celeb-DF
dataset. The findings suggest that feature extraction techniques can improve the robustness
of detection models, making them more resilient to sophisticated deepfake manipulations.
Future research might investigate refining feature selection and classifier tweaking to
improve overall performance.
Table 4presents a comparison of execution times for feature extraction and classifi-
cation, measured in milliseconds on both GPU and CPU. This comparison evaluates the
inference time and classification accuracy of different methods, as outlined in the evaluation
criteria (Section 4.1), using the FaceForensics++ dataset.
Table 4. Execution time comparison for feature extraction and classification.
Methods Feature Extraction Training Inference GPU CPU
Random forest 0.5 s 30 m 15 ms 92 ms
Extra Trees Classifier 0.5 s 25 m 15 ms 95 ms
Support Vector Classifier 0.5 s 60 m 13 ms 63 ms
XGB Classifier 0.5 s 45 m 12 ms 75 ms
Support Vector Machine 0.5 s 120 m 25 ms 85 ms
XceptionNet 0.5 s 210 m 20 ms 2 s
Convolutional Neural Network 0.5 s 180 m 10 ms 1.5 s
HOG + KAZE (Proposed) 1.0 s 45 m 16 ms 67 ms
LBP + KAZE (Proposed) 0.5 s 30 m 09 ms 56 ms
Feature extraction time reflects the duration required to extract LBP, HOG, and KAZE
features, in contrast with the extraction times of traditional machine learning models.
Training time refers to the time duration needed to train the classifier after feature extraction,
while inference time measures the time taken to classify a single instance once the model has
been trained. This analysis provides insights into the computational efficiency of various
approaches, helping to assess their suitability for real-time deepfake detection.
These tables provide a clear comparison of the efficiency and effectiveness of the
proposed fusion model using KAZE and HOG features against traditional ML approaches.
Table 4demonstrates that while feature extraction with KAZE and HOG may take slightly
longer (per frame) than traditional ML models on GPU, it is significantly faster than
training DL models. Table 2highlights that the proposed fusion model achieves competitive
accuracy while maintaining comparative training and inference time, making it suitable for
real-time applications where speed is important (see Table 4for GPU and CPU inference).
These comparisons underscore the practical advantages of the proposed fusion model in
terms of computational efficiency without compromising classification performance. The
proposed fusion-based methods perform better in terms of training time, despite CNN’s
marginally better performance and accuracy, but are lacking in inference time, especially
when the CPU is used. Although CNN’s inference is somewhat superior when the GPU
is used compared to the proposed fusion model, the methods are reliable and perfect for
resource-constrained situations, as they performed better when the CPU is used. The

Appl. Sci. 2024,1, 0 14 of 17
fusion of HOG and KAZE provides a good balance between computational efficiency and
classification performance.
4.4. Future Work and Implications of Visual Information Security
The fusion of LBP, HOG, and KAZE features has proven effective in detecting deep-
fake content. HOG captures texture patterns, while KAZE detects structural distortions
introduced by deepfake generation techniques. Future research could refine classifiers to im-
prove the model’s ability to distinguish between real and fake content. Exploring advanced
hybrid techniques like ORB, DSIFT, and Wavelet Transform Features could further improve
detection accuracy and computational efficiency. Additionally, dimensionality reduction
methods such as PCA and t-SNE can optimize feature selection, while DL approaches like
hybrid CNN architectures or GANs could bolster the robustness of deepfake detection.
This approach has strong potential for forensic and legal applications, providing
a reliable means to verify the authenticity of digital media in critical legal proceedings.
It could also contribute to real-time authentication systems for digital media platforms,
potentially integrating blockchain or watermarking techniques for added security. The
research highlights the importance of robust feature extraction methods in deepfakes, and
future efforts should focus on refining these techniques and adapting to new manipulation
strategies to ensure continued efficacy in securing digital media.
5. Conclusions
In conclusion, this study introduced an effective approach for detecting deepfake
images utilizing texture-based features through the fusion of HOG/LBP and KAZE within
an ML framework. The computational load is significantly reduced compared to traditional
DL models, making this method ideal for real-time applications with limited processing
resources. The experiments using classifiers such as RF, XGBoost, extra trees, and support
vector classifiers demonstrated the distinct advantages of each method in evaluating feature
importance across HOG feature bands. The feature-level fusion technique further improved
performance on both the FaceForensics++ and Celeb-DF datasets, achieving an accuracy of
92.12% and 78%, respectively. This approach not only improves accuracy and efficiency in
detecting deepfake content but also provides a scalable solution against the potential abuse
of technology and its consequences on politics, finance, and personal privacy. Beyond
deepfake detection, the method holds the potential for authenticating various forms of
digital content, emphasizing its broad applicability in fields that require reliable visual
data verification.
Funding: This study was financially supported by the Seoul National University of Science and
Technology, Seoul, South Korea.
Conflicts of Interest: The authors declare no conflicts of interest.
References
1.
Kharvi, P.L. Understanding the Impact of AI-Generated Deepfakes on Public Opinion, Political Discourse, and Personal Security
in Social Media. IEEE Secur. Priv. 2024,22, 115–122. https://doi.org/10.1109/MSEC.2024.3405963.
2.
Domenteanu, A.; T˘ataru, G.C.; Cr˘aciun, L.; Mol˘anescu, A.G.; Cotfas, L.A.; Delcea, C. Living in theAge of Deepfakes: A Bibliometric
Exploration of Trends, Challenges, and Detection Approaches. Information 2024,15, 525. https://doi.org/10.3390/info15090525.
3.
Bale, D.; Ochei, L.; Ugwu, C. Deepfake Detection and Classification of Images from Video: A Review of Features, Techniques, and
Challenges. Int. J. Intell. Inf. Syst. 2024,13, 20–28. https://doi.org/10.11648/j.ijiis.20241302.11.
4.
Vijaya, J.; Kazi, A.A.; Mishra, K.G.; Praveen, A. Generation Furthermore, Detection of Deepfakes using Generative Adversarial
Networks (GANs) and Affine Transformation. In Proceedings of the 2023 14th International Conference on Computing Commu-
nication and Networking Technologies (ICCCNT), Delhi, India, 6–8 July 2023; pp. 1–6. https://doi.org/10.1109/ICCCNT56998.2
023.10307811.

Appl. Sci. 2024,1, 0 15 of 17
5.
Lee, J.; Jang, J.; Lee, J.; Chun, D.; Kim, H. CNN-Based Mask-Pose Fusion for Detecting Specific Persons on Heterogeneous
Embedded Systems. IEEE Access 2021,9, 120358–120366.
6.
Lee, S.I.; Kim, H. GaussianMask: Uncertainty-aware Instance Segmentation based on Gaussian Modeling. In Proceedings of the
2022 26th International Conference on Pattern Recognition (ICPR), Montreal, QC, Canada, 21–25 August 2022; pp. 3851–3857.
7.
Chun, D.; Lee, S.; Kim, H. USD: Uncertainty-Based One-Phase Learning to Enhance Pseudo-Label Reliability for Semi-Supervised
Object Detection. IEEE Trans. Multimed. 2024,26, 6336–6347.
8.
Lee, J.J.; Kim, H. Multi-Step Training Framework Using Sparsity Training for Efficient Utilization of Accumulated New Data in
Convolutional Neural Networks. IEEE Access 2023,11, 129613–129622.
9.
Abbas, F.; Taeihagh, A. Unmasking deepfakes: A systematic review of deepfake detection and generation techniques using
artificial intelligence. Expert Syst. Appl. 2024,252, 124260. https://doi.org/https://doi.org/10.1016/j.eswa.2024.124260.
10.
Naskar, G.; Mohiuddin, S.; Malakar, S.; Cuevas, E.; Sarkar, R. Deepfake detection using deep feature stacking and meta-learning.
Heliyon 2024,10, e25933. https://doi.org/https://doi.org/10.1016/j.heliyon.2024.e25933.
11.
Rana, M.S.; Murali, B.; Sung, A.H. Deepfake Detection Using Machine Learning Algorithms. In Proceedings of the 2021
10th International Congress on Advanced Applied Informatics (IIAI-AAI), Niigata, Japan, 11–16 July 2021; pp. 458–463.
https://doi.org/10.1109/IIAI-AAI53430.2021.00079.
12.
Tsalera, E.; Papadakis, A.; Samarakou, M.; Voyiatzis, I. Feature Extraction with Handcrafted Methods and Convolutional Neural
Networks for Facial Emotion Recognition. Appl. Sci. 2022,12, 8455. https://doi.org/10.3390/app12178455.
13.
Moore, S.; Bowden, R. Local binary patterns for multi-view facial expression recognition. Comput. Vis. Image Underst. 2011,
115, 541–558. https://doi.org/10.1016/j.cviu.2010.12.001.
14.
Alcantarilla, P.F.; Bartoli, A.; Davison, A.J. KAZE Features. In Proceedings of the European Conference on Computer Vision—
ECCV 2012, Florence, Italy, 7–13 October 2012; Fitzgibbon, A.; Lazebnik, S.; Perona, P.; Sato, Y.; Schmid, C., Eds.; Springer:
Berlin/Heidelberg, Germany, 2012; pp. 214–227.
15.
Huda, N.u.; Javed, A.; Maswadi, K.; Alhazmi, A.; Ashraf, R. Fake-checker: A fusion of texture features and deep learning for
deepfakes detection. Multimed. Tools Appl. 2024,83, 49013–49037. https://doi.org/10.1007/s11042-023- 17586-x.
16.
A. Khalid, N.A.; Imran Ahmad, M.; Shie Chow, T.; H. Mandeel, T.; Majid Mohammed, I.; Kadhim Alsaeedi, M.A. Palmprint
recognition system using VR-LBP and KAZE features for better recognition accuracy. Bull. Electr. Eng. Inform. 2024,13, 1060–1068.
https://doi.org/10.11591/eei.v13i2.4739.
17.
Ghosh, B.; Malioutov, D.; Meel, K.S. Efficient Learning of Interpretable Classification Rules. J. Artif. Intell. Res. 2022,74, 1823–1863.
https://doi.org/10.1613/jair.1.13482.
18.
Patel, Y.; Tanwar, S.; Gupta, R.; Bhattacharya, P.; Davidson, I.E.; Nyameko, R.; Aluvala, S.; Vimal, V. Deepfake Generation and
Detection: Case Study and Challenges. IEEE Access 2023,11, 143296–143323. https://doi.org/10.1109/ACCESS.2023.3342107.
19.
Chen, P.; Xu, M.; Wang, X. Detecting Compressed Deepfake Images Using Two-Branch Convolutional Networks with Similarity
and Classifier. Symmetry 2022,14, 2691. https://doi.org/10.3390/sym14122691.
20.
Hong, H.; Choi, D.; Kim, N.; Kim, H. Mobile-X: Dedicated FPGA Implementation of the MobileNet Accelerator Optimizing
Depthwise Separable Convolution. IEEE Trans. Circuits Syst. II Express Briefs 2024,71, 4668–4672.
21.
Ki, S.; Park, J.; Kim, H. Dedicated FPGA Implementation of the Gaussian TinyYOLOv3 Accelerator. IEEE Trans. Circuits Syst. II
Express Briefs 2023,70, 3882–3886.
22.
Du, M.; Pentyala, S.; Li, Y.; Hu, X. Towards Generalizable Deepfake Detection with Locality-aware AutoEncoder. In Proceedings
of the 29th ACM International Conference on Information & Knowledge Management (CIKM’20), Virtual, 19–23 October 2020.
https://doi.org/10.1145/3340531.3411892.
23.
Lanzino, R.; Fontana, F.; Diko, A.; Marini, M.R.; Cinque, L. Faster Than Lies: Real-time Deepfake Detection using Binary Neural
Networks. In Proceedings of the 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW),
Seattle, WA, USA, 17–18 June 2024; pp. 3771–3780. https://doi.org/10.1109/CVPRW63382.2024.00381.
24.
Rössler, A.; Cozzolino, D.; Verdoliva, L.; Riess, C.; Thies, J.; Nießner, M. FaceForensics++: Learning to Detect Manipulated Facial
Images. In Proceedings of the International Conference on Computer Vision (ICCV), Seoul, Republic of Korea, 27 October–2
November 2019.
25.
Li, Y.; Yang, X.; Sun, P.; Qi, H.; Lyu, S. Celeb-DF: A Large-Scale Challenging Dataset for DeepFake Forensics. In Proceedings of the
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA, 13–19 June 2020.
26.
Hong, H.; Choi, D.; Kim, N.; Lee, H.; Kang, B.; Kang, H.; Kim, H. Survey of convolutional neural network accelerators on
field-programmable gate array platforms: Architectures and optimization techniques. J. Real-Time Image Process. 2024,21, 64.
27.
Heidari, A.; Jafari Navimipour, N.; Dag, H.; Unal, M. Deepfake detection using deep learning methods: A systematic and
comprehensive review. WIREs Data Min. Knowl. Discov. 2024,14, e1520. https://doi.org/https://doi.org/10.1002/widm.1520.
28.
Bonettini, N.; Cannas, E.D.; Mandelli, S.; Bondi, L.; Bestagini, P.; Tubaro, S. Video Face Manipulation Detection Through Ensemble
of CNNs. In Proceedings of the 2020 25th International Conference on Pattern Recognition (ICPR), Milan, Italy, 10–15 January
2021; pp. 5012–5019. https://doi.org/10.1109/ICPR48806.2021.9412711.

Appl. Sci. 2024,1, 0 16 of 17
29.
Saberi, M.; Sadasivan, V.S.; Rezaei, K.; Kumar, A.; Chegini, A.; Wang, W.; Feizi, S. Robustness of AI-Image Detectors: Fundamental
Limits and Practical Attacks. In Proceedings of the 12th International Conference on Learning Representations, Vienna, Austria,
7–11 May 2024.
30.
Dong, F.; Zou, X.; Wang, J.; Liu, X. Contrastive learning-based general Deepfake detection with multi-scale RGB frequency clues.
J. King Saud Univ.-Comput. Inf. Sci. 2023,35, 90–99.
31.
Raza, M.A.; Malik, K.M.; Haq, I.U. Holisticdfd: Infusing spatiotemporal transformer embeddings for deepfake detection. Inf. Sci.
2023,645, 119352.
32.
Zhu, Y.; Zhang, C.; Gao, J.; Sun, X.; Rui, Z.; Zhou, X. High-compressed deepfake video detection with contrastive spatiotemporal
distillation. Neurocomputing 2024,565, 126872.
33.
Almestekawy, A.; Zayed, H.H.; Taha, A. Deepfake detection: Enhancing performance with spatiotemporal texture and deep
learning feature fusion. Egypt. Inform. J. 2024,27, 100535. https://doi.org/https://doi.org/10.1016/j.eij.2024.100535.
34. Chen, B.; Liu, X.; Xia, Z.; Zhao, G. Privacy-preserving DeepFake face image detection. Digit. Signal Process. 2023,143, 104233.
35.
Tareen, S.A.K.; Saleem, Z. A comparative analysis of SIFT, SURF, KAZE, AKAZE, ORB, and BRISK. In Proceedings of the 2018
International Conference on Computing, Mathematics and Engineering Technologies (iCoMET), Sukkur, Pakistan, 3–4 March
2018; pp. 1–10. https://doi.org/10.1109/ICOMET.2018.8346440.
36.
Yallamandaiah, S.; Purnachand, N. A novel face recognition technique using Convolutional Neural Network, HOG, and histogram
of LBP features. In Proceedings of the 2022 2nd International Conference on Artificial Intelligence and Signal Processing (AISP),
Vijayawada, India, 12–14 February 2022; pp. 1–5. https://doi.org/10.1109/AISP53593.2022.9760679.
37.
Cavalcanti, G.D.; Ren, T.I.; Pereira, J.F. Weighted Modular Image Principal Component Analysis for face recognition. Expert Syst.
Appl. 2013,40, 4971–4977. https://doi.org/https://doi.org/10.1016/j.eswa.2013.03.003.
38. Lu, G.F.; Zou, J.; Wang, Y. Incremental complete LDA for face recognition. Pattern Recognit. 2012,45, 2510–2521.
39.
Fathi, A.; Alirezazadeh, P.; Abdali-Mohammadi, F. A new Global-Gabor-Zernike feature descriptor and its application to face
recognition. J. Vis. Commun. Image Represent. 2016,38, 65–72.
40.
Topi, M.; Timo, O.; Matti, P.; Maricor, S. Robust texture classification by subsets of local binary patterns. In Proceedings of the
15th International Conference on Pattern Recognition (ICPR-2000), Barcelona, Spain, 3–7 September 2000; Volume 3, pp. 935–938.
41.
Dalal, N.; Triggs, B. Histograms of oriented gradients for human detection. In Proceedings of the 2005 IEEE Computer Society
Conference on Computer Vision and Pattern Recognition (CVPR’05), San Diego, CA, USA, 20–25 June 2005; Volume 1, pp.
886–893.
42.
Déniz, O.; Bueno, G.; Salido, J.; De la Torre, F. Face recognition using histograms of oriented gradients. Pattern Recognit. Lett. 2011,
32, 1598–1603.
43.
Khalil, S.S.; Youssef, S.M.; Saleh, S.N. iCaps-Dfake: An integrated capsule-based model for deepfake image and video detection.
Future Internet 2021,13, 93.
44.
Ruano-Ordás, D. Machine Learning-Based Feature Extraction and Selection. Appl. Sci. 2024,14, 6567. https://doi.org/10.3390/
app14156567.
45. Aslan, M.F.; Durdu, A.; Sabanci, K.; Mutluer, M.A. CNN and HOG based comparison study for complete occlusion handling in
human tracking. Measurement 2020,158, 107704. https://doi.org/https://doi.org/10.1016/j.measurement.2020.107704.
46.
Zare, M.R.; Alebiosu, D.O.; Lee, S.L. Comparison of Handcrafted Features and Deep Learning in Classification of Medical X-ray
Images. In Proceedings of the 2018 Fourth International Conference on Information Retrieval and Knowledge Management
(CAMP), Kota Kinabalu, Malaysia, 26–28 March 2018; pp. 1–5. https://doi.org/10.1109/INFRKM.2018.8464688.
47.
Zotova, D.; Pinon, N.; Trombetta, R.; Bouet, R.; Jung, J.; Lartizien, C. Gan-Based Synthetic Fdg Pet Images from T1 Brain MRI Can
Serve to Improve Performance of Deep Unsupervised Anomaly Detection Models. SSRN 2024, 34. https://doi.org/10.2139/ssrn.
4917648.
48.
Chen, Y.; Haldar, N.A.H.; Akhtar, N.; Mian, A. Text-image guided Diffusion Model for generating Deepfake celebrity interactions.
In Proceedings of the 2023 International Conference on Digital Image Computing: Techniques and Applications (DICTA), Port
Macquarie, Australia, 28 November–1 December 2023; pp. 348–355.
49.
Yuan, G.; Cun, X.; Zhang, Y.; Li, M.; Qi, C.; Wang, X.; Shan, Y.; Zheng, H. Inserting anybody in diffusion models via celeb basis.
arXiv 2023, arXiv:2306.00926.
50.
Abhisheka, B.; Biswas, S.K.; Das, S.; Purkayastha, B. Combining Handcrafted and CNN Features for Robust Breast Cancer
Detection Using Ultrasound Images. In Proceedings of the 2023 3rd International Conference on Innovative Sustainable
Computational Technologies (CISCT), Dehradun, India, 8–9 September 2023; p. 1–6. https://doi.org/10.1109/cisct57197.2023.10
351282.
51.
Mohtavipour, S.M.; Saeidi, M.; Arabsorkhi, A. A multi-stream CNN for deep violence detection in video sequences using
handcrafted features. Vis. Comput. 2021,38, 2057–2072. https://doi.org/10.1007/s00371-021-02266-4.
52.
Devi, P.A.R.; Budiarti, R.P.N. Image Classification with Shell Texture Feature Extraction Using Local Binary Pattern (LBP) Method.
Appl. Technol. Comput. Sci. J. 2020,3, 48–57. https://doi.org/10.33086/atcsj.v3i1.1745.

Appl. Sci. 2024,1, 0 17 of 17
53.
Werghi, N.; Berretti, S.; del Bimbo, A. The Mesh-LBP: A Framework for Extracting Local Binary Patterns From Discrete Manifolds.
IEEE Trans. Image Process. 2015,24, 220–235. https://doi.org/10.1109/tip.2014.2370253.
54.
Kumar, D.G. Identical Image Extraction from PDF Document Using LBP (Local Binary Patterns) and RGB (Red, Green and Blue)
Color Features. Int. J. Res. Appl. Sci. Eng. Technol. 2022,10, 3563–3566. https://doi.org/10.22214/ijraset.2022.45811.
55.
Karunarathne, B.A.S.S.; Wickramaarachchi, W.H.C.; De Silva, K.K.K.M.C. Face Detection and Recognition for Security System
using Local Binary Patterns (LBP). J. ICT Des. Eng. Technol. Sci. 2019,3, 15–19. https://doi.org/10.33150/jitdets-3.1.4.
56.
Yang, B.; Chen, S. A comparative study on local binary pattern (LBP) based face recognition: LBP histogram versus LBP image.
Neurocomputing 2013,120, 365–379. https://doi.org/10.1016/j.neucom.2012.10.032.
57.
Huang, Z.R. CN-LBP: Complex Networks-Based Local Binary Patterns for Texture Classification. In Proceedings of the 2021
International Conference on Wavelet Analysis and Pattern Recognition (ICWAPR), Adelaide, Australia, 4–5 December 2021; pp.
1–6. https://doi.org/10.1109/icwapr54887.2021.9736189.
58.
Rahayu, M.I.; Nasihin, A. Design of Face Recognition Detection Using Local Binary Pattern (LBP) Method. J. Teknol. Inf. Dan
Komun. 2020,9, 48–54. https://doi.org/10.58761/jurtikstmikbandung.v9i1.145.
59.
Albiol, A.; Monzo, D.; Martin, A.; Sastre, J.; Albiol, A. Face recognition using HOG–EBGM. Pattern Recognit. Lett. 2008,
29, 1537–1543.
60.
Kusniadi, I.; Setyanto, A. Fake Video Detection using Modified XceptionNet. In Proceedings of the 2021 4th International
Conference on Information and Communications Technology (ICOIACT), Yogyakarta, Indonesia, 30–31 August 2021; pp. 104–107.
https://doi.org/10.1109/ICOIACT53268.2021.9563923.
61.
Hastie, T.; Tibshirani, R.; Friedman, J. The Elements of Statistical Learning: Data Mining, Inference, and Prediction; Springer Series in
Statistics; Springer: New York, NY, USA, 2009; Volume 1.
62.
Fawzi, A.; Fawzi, O.; Gana, M.F. Robustness of classifiers: From the theory to the real world. IEEE Trans. Neural Networks Learn.
Syst. 2016,28, 505–518.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.