FACULTAD DE INGENIERÍA
Escuela Académico Profesional de Ingeniería Empresarial
Tesis
Approach to the Luxand Face Facial Recognition
System Aimed at the Detection of People in the
Criminalistics Unit of the PNP In Huancayo City, Peru
Michelle Haydee Benavides Canchari
Mariam Corayma Salcedo Gonazales
Estefanny Yessica Caceres Aldana
Wilson Anthony Lazo Tapia
Diana Paola Chipana Gago
Para optar el Título Profesional de
Ingeniero Empresarial
Huancayo, 2024
Esta obra está bajo una Licencia "Creative Commons Atribución 4.0 Internacional" .
Tesis
INFORME DE CONFORMIDAD DE ORIGINALIDAD DE TESIS: EN FORMATO
ARTÍCULO CIENTÍFICO
A : Dr. Felipe Néstor Gutarra Meza
Decano de la Facultad de Ingeniería
DE : Ing. Diana Paola Chipana Gago
Asesor de tesis en formato artículo científico
ASUNTO : Remito resultado de evaluación de originalidad de tesis en formato
artículo científico
FECHA : 19 de Febrero de 2024
Con sumo agrado me dirijo a vuestro despacho para saludarlo y en vista de
haber sido designado asesor de la tesis en formato artículo científico titulada:
“Approach to the Luxand Face Facial Recognition System Aimed at the Detection of
People in the Criminalistics Unit of the PNP in Huancayo City, Peru”, perteneciente
al/la/los/las estudiante(s) Michelle Haydee Benavides Canchari, de la E.A.P. de
Ingeniería Empresarial; Mariam Corayma Salcedo Gonzalez, de la E.A.P. de Ingeniería
Empresarial; Estefanny Yessica Caceres Aldana, de la E.A.P. de Ingeniería Empresarial;
se procedió con la carga del documento a la plataforma “Turnitin” y se realizó la
verificación completa de las coincidencias resaltadas por el software dando por
resultado 8 % de similitud (informe adjunto) sin encontrarse hallazgos relacionados a
plagio. Se utilizaron los siguientes filtros:
Filtro de exclusión de bibliografía SI X NO
Filtro de exclusión de grupos de palabras menores SI NO X
(Nº de palabras excluidas: )
Exclusión de fuente por trabajo anterior del mismo estudiante SI X NO
En consecuencia, se determina que la tesis en formato artículo científico constituye un
documento original al presentar similitud de otros autores (citas) por debajo del
porcentaje establecido por la Universidad.
Recae toda responsabilidad del contenido la tesis en formato artículo científico sobre el
autor y asesor, en concordancia a los principios de legalidad, presunción de veracidad
y simplicidad, expresados en el Reglamento del Registro Nacional de Trabajos de
Investigación para optar grados académicos y títulos profesionales – RENATI y en la
Directiva 003-2016-R/UC.
Esperando la atención a la presente, me despido sin otro particular y sea propicia la
ocasión para renovar las muestras de mi especial consideración.
Atentamente,
La firma del asesor obra en el archivo original
(No se muestra en este documento por estar expuesto a publicación)
DECLARACIÓN JURADA DE AUTORÍA
El presente documento tiene por finalidad declarar adecuada y explícitamente el
aporte de cada estudiante en la elaboración del trabajo de investigación a ser utilizado
para la sustentación de tesis: formato de artículo científico.
Yo: Michelle Haydee Benavides Canchari, con Documento nacional de identidad (DNI)
N° 73990056; teléfono 996968928; estudiante de la Escuela Académico Profesional de
Ingeniería Empresarial.
Yo: Mariam Corayma Salcedo Gonzalez, con Documento nacional de identidad (DNI)
N° 72723010; teléfono 982308940; estudiante de la Escuela Académico Profesional de
Ingeniería Empresarial.
Yo: Estefanny Yessica Caceres Aldana, con Documento nacional de identidad (DNI) N°
77282396; teléfono 963820316; estudiante de la Escuela Académico Profesional de
Ingeniería Empresarial.
Yo: Wilson Anthony Lazo Tapia, con Documento nacional de identidad (DNI) N°
70237739; teléfono 947198655.
Yo: Diana Paola Chipana Gago, con Documento nacional de identidad (DNI) N°
40974844; teléfono 937601612.
Ante Usted, con el debido respeto me presento y expongo:
Declaramos que hemos participado en la ideación del problema, recolección de datos,
elaboración y aprobación final del artículo científico.
La firma del autor y del asesor obra en el archivo original
(No se muestra en este documento por estar expuesto a publicación)
Approach to the Luxand Face Facial Recognition System Aimed at the Detection
of People in the Criminalistics Unit of the PNP in Huancayo City, Peru
Michelle H. Benavides Canchari
Professional Academic School of Management Engineering, Faculty of Engineering, Universidad Continental,
73990056@continental.edu.pe
Mariam C. Salcedo Gonzalez
Professional Academic School of Management Engineering, Faculty of Engineering, Universidad Continental,
72723010@continental.edu.pe
Estefanny Y. Cáceres Aldana
Professional Academic School of Management Engineering, Faculty of Engineering, Universidad Continental,
77282396@continental.edu.pe
Wilson A. Lazo Tapia*
Faculty of Engineering, Universidad Continental, 70237739@continental.edu.pe
Diana P. Chipana Gago
Faculty of Engineering, Universidad Continental, dchipana@continental.edu.pe
Throughout Latin America, citizen security has been negatively affected, due to the increase in criminal acts and the
lack of optimal management of the security system in the different countries, in addition to the deficiency of
technological equipment that covers all needs of the police forces to fulfill their obligations. In Peru, one of the aspects
that most harms citizen security is the lack of a video surveillance system that provides people detection functions. That
is why the objective of this article is to efficiently influence the improvement of the process of detecting people in the
criminalistics unit of the PNP (Peruvian National Police) in Huancayo city, Peru. For this research, the Engineering -
Cascade Model methodology was used, which confers four stages: requirements analysis, design, construction, and
testing. Regarding the results, the Degree of similarity indicator reached a value of 89% and the Police relationship
with the community indicator obtained a result of 93%. When making the comparison with a high-level investigation,
similar results were observed, which shows the reliability of our article. With this, we can conclude that the
implementation approach of the Luxand Face facial recognition system in the criminal unit really had a significant
effect on the detection of people, being considered an advanced technology project that meets the needs of citizen
security.
CCS CONCEPTS • Security and privacy~Biometrics
Additional Keywords and Phrases: Degree of similarity, Detection, Facial recognition, Video surveillance
cameras
ACM Reference Format:
First Author’s Name, Initials, and Last Name, Second Author’s Name, Initials, and Last Name, and Third Author’s
Name, Initials, and Last Name. 2018. The Title of the Paper: ACM Conference Proceedings Manuscript Submission
Template: This is the subtitle of the paper, this document both explains and embodies the submission format for authors
using Word. In Woodstock ’18: ACM Symposium on Neural Gaze Detection, June 03–05, 2018, Woodstock, NY.
ACM, New York, NY, USA, 10 pages. NOTE: This block will be automatically generated when manuscripts are
processed after acceptance.
1 Introduction
Throughout Latin America, citizen security has been negatively affected, due to the increase in criminal acts and the
lack of optimal management of the security system in the different countries, in addition to the deficiency of
technological equipment that covers all needs of the police forces to fulfill their obligations [1]. Specifically in Peru,
one of the aspects that most harms citizen security is the lack of a video surveillance system that provides people
detection functions [2].
One of the problems is the insufficient research on facial recognition issues through police images with the use of
video surveillance cameras [3]. In Peru, citizen insecurity is quite serious, which is why a system based on facial
recognition increases the satisfaction of society in relation to security tools [4]. Biometric technology, which makes
use of the unique properties of each person, is a valuable resource for security in several fields, which contributes to a
more effective and rapid identification, since it reduces the need for patrolling by the authorities because the system
can operate 24 hours a day [5]. Specifically, facial recognition is easy and simple to access because it can be found in
various programs, identifying and recognizing facial points that are registered through a database [6].
* Corresponding author
The automatic recognition of facial points is widely used, thus automating security activities with face
identification and verification [7]. Facial recognition-based security systems are efficient for video surveillance of
streets and crowded areas, adapting to variations in face poses and angles [8]. In addition, the algorithms of facial
recognition systems have been improved to allow the identification of people despite the fact that they have various
accessories on their faces [9]. Therefore, this technology is used by security authorities to locate lost people or
identify people who have committed crimes and are wanted by the authorities [10]. In citizen security, it is
recommended to use facial recognition since it is the most reliable and viable technology for the identification of
people [11].
Likewise, in the article "Face Recognition for Criminal Identification: An Alert System in Suspects Scenarios", it
is mentioned that in Peru there is no adequate technology to optimize the process of detecting people, it also shows
the importance of a facial recognition system. to be able to identify people in the country in real time and allows to
know the percentage of optimization in the time of detection through an analysis of their facial features and a
comparison with the police database [12]. The article "Acceptance of facial recognition as a measure of surveillance
and security" manages to demonstrate that the fact of recognizing faces can be perceived as a measure in relation to
security for society [13]. In addition, the book "Trends: Keys to understand the future and build the present" explains
the case of the implementation of a surveillance system with facial recognition in another country, in addition to
explaining how this system has been developed and the results it has achieved., with the aim of contributing to citizen
security [14]. In the context of a pandemic in which various investigations related to facial recognition systems were
carried out; one of the most outstanding indicated that the performance of the programs used for the detection system
is related to the characteristics of the face and the use of a mask [15]. However, this was not an obstacle for the
technology because there are new methodologies that focus on the upper parts of the face (eye, eyebrow, and
forehead) with 93% accuracy in face detection [16].
The research problem is the development of the implementation of the facial recognition system Luxand Face
oriented to the detection of people, presenting as an objective to efficiently influence the improvement of the process
of detection of people in the criminalistics unit of the PNP (National Police of Peru) in Huancayo city, Peru, taking
into account that two variables were studied, the detection system based on facial recognition and the people
detection process. The rest of the document is organized as follows, section 2 contains the methodology, section 3
contains the results, section 4 contains the discussion and section 5 contains the conclusions.
2 METHODOLOGY
A detection system based on facial recognition is necessary, since it allows the criminology unit of the National
Police of Peru to locate wanted people, it also has the security necessary to avoid violating the personal data of the
population [17]. Figure 1 shows the independent variable, which is the facial recognition based detection system, and
the dependent variable, which is the people detection process. The dependent variable is affected by the independent
variable since it goes from being a poor people detection process to being an efficient people detection process, as it
is shown in Figure 2.
Figure 1: Table of research variables
Figure 2: Systematic model of the research problem
According to the book Investigation Methodology written by Hernández, R.; Fernández, C. and Baptista, M.,
mention “The quantitative approach (which symbolizes, as we explained, a set of processes) is sequential and
demonstrative. Each stage precedes the next and we cannot "jump" or avoid steps.” [18]
For the development of this research work, the Engineering - Waterfall Model methodology was used, as it is
shown in Figure 3, because it presents an approach aimed at systems and software projects, and is based on an
automatic learning technique used to solve problems of classification. It is based on the idea of dividing the problem
into several stages or "levels" and applying a model to each of them. Each level uses the output of the previous level
as input, and in turn provides an output for the next level [19].
Figure 3: Waterfall Model methodology
2.1 Requirements analysis
To start with the prototype of the system, it is essential to compile the requirements focused on it, since the person in
charge of the project must obtain the necessary information to evaluate the requirements, based on said information,
the functionality and performance of the system, as well as the objectives set. Table 1 presents the requirements
related to the business, the stakeholders, and the solution.
Table 1: Requirements analysis
Need
Determine the development of the implementation of the facial recognition system Luxand Face oriented to the
detection of people in the criminalistics unit of the PNP in Huancayo city, Peru.
Business requirements
ID Description
NB1 Optimize the degree of similarity
NB2 Strengthen the police relationship with the community
Stakeholder requirements
ID Description Owner Relationship
SR1 Generate compliance from police personnel
Head of the
criminology unit of
the PNP
NB1
SR2 Increase community trust in police forces police personnel NB2
Solution requirements
ID Description
Functional requirements
FR1 The system must detect the faces of the people sought System manager SR1, SR2
FR2
The system should notify in case of similarity of facial
features
System manager SR1
Non-functional requirements
NFR1
Police personnel must have the necessary technological
equipment
Head of the
criminology unit of
the PNP
FR1, FR2
NFR2
The system must have a comprehensive database for
photographs
Programador del
sistema
FR1
Transition requirements
TR1 Train police personnel from the criminology unit
SOLUTIO
N
Approach to the Luxand Face facial recognition system aimed at the detection of people in the criminalistics
unit of the PNP in Huancayo city, Peru.
2.2 Design
For the design, the previously analyzed requirements must be met, in addition to developing the programming codes
and algorithms of the system prototype, focusing on the structure attributes, detail of the procedure and the
characterization of the program and interfaces.
Figure 4 shows the first model of the system that starts with the camera to recognize people's faces linked to a
laptop that is connected to the Internet to connect to the Luxand FaceSDK System program that will work cyclically
for optimal work.
Figure 4: Design of the first model of the system
Figure 5 shows the process of detecting people where the person responsible of the detection management
interacts with the head of the criminalistics unit of the PNP.
Figure 5: People detection process modeling
2.3 Building
In this phase, the prototype of the system is put into operation according to what was previously designed to be able
to carry it out, checking its internal and external operation. It will be done mechanically if the design is detailed.
In figure 6, it shows the installation of the prototype and how they interact with each other.
Figure 6: Installation and connection of equipment
A.- Security cameras: Instruments that allow monitoring in real time and lead to investigations on the recognition
of humans in order to guarantee a safe environment [20].
B.- Router: “Routers or routers are the interconnection devices that allow each packet sent to reach its destination
following the most feasible path or route.” [21]
C.- Database: A set of data and the way they are organised The first databases were hierarchical and network
databases, developed during the 1960s. They exposed procedural query interfaces (as opposed to descriptive ones), so
that accessing specific information in one of these systems was similar to navigating to a specific file within a file
system [22]. The database for facial recognition helps to register and compare a greater number of registered facial
points of a person for comparison [23].
D.- Internet: The internet has become part of our daily life, ceasing to be an alternative and optional space to
become a point of obligatory use, it originates needs and in turn creates products that satisfy [24].
E.- Criminalistics Unit: “It is the science that integrates the set of scientific technical knowledge applicable to the
investigation of the crime, in order to establish, through studies and analysis of the indications and/or evidence, the
means of proof, form and circumstances of its perpetration, as well as the identification of the perpetrator(s) and
implicated.” [25]
Figure 7 shows the Luxand FaceSDK system which allows the recognition of people through a camera,
identifying several points of the facial features of the people that are captured by the video surveillance camera
connected to the system, on the other hand the application allows register and store in a database the faces of people
so that they can be identified later.
Figure 7: Luxand Face SDK System
2.4 Testing
After installing the system prototype, the tests will have to be carried out. The prototype tests are carried out
internally and externally, checking its functions to ensure that the results required by the client are produced. Figure 8
shows how the program manages to correctly recognize two people from the sample, since it identifies them from the
database, this is thanks to the fact that both faces were registered in the system, from the front and in profile, and it is
evident that the names of both people in the sample are correctly identified.
Figure 8: Face Recognition System Test
Figure 9 shows the recognition of 70 facial points of the person who is registered in the system, this function of
the system is essential to differentiate each person registered in the database, because the group of facial points of
each person form a unique recognition [26]. This type of program, such as the Kinect sensor, executes recognition of
the entire region of the face in general, for this, a robust recognition is generated with a system that detects specific
facial points [27].
Figure 9: Facial dot recognition system test
3 RESULTS
“According to the book Investigation Methodology written by Hernández, Roberto; Fernández, Carlos and Baptista,
María del Pilar, the pre-test and re-test method consists in that this measurement instrument is applied two or more
times to the same group of people or cases after a certain period.” [18] In the research work, the Pre-test and Re-test
and Post.test methodology will be analyzed and carried out in the criminology unit of the Peruvian National Police in
Huancayo city. That is why the statistical software SPSS Statistics was used to carry out the analysis and discussion
of results.
Indicator: Degree of Similarity
Figure 10 shows the descriptive statistical results of the Degree of Similarity indicator can be observed according
to the results in the statistical software SPSS (Spanish version):
Figure 10: Descriptive analysis - Degree of similarity
For the Degree of Similarity indicator, in the Pre-Test an average value of degree of similarity of 32% per person
was reached. In the same way, after the implementation of the detection system based on facial recognition, an
average value of degree of similarity of 89.49% per person was reached, which indicates that the implementation of
said system generates a noticeable increase in the indicator. On the other hand, in the Pre-Test, the standard deviation
had the following value of 0.11; and in the Post-Test, it had a value of 0.06, which is why it can be stated that the
standard deviation decreased, which means that the dispersion of the degree of similarity has been reduced [28].
Indicator: Police relationship with the community
Figures 11 and figure 12 show the police relationship with the community since "it turns out to be a fundamental
informant when measuring police performance.” [29]
Figure 11: Police relationship with the community - Pre-test
Figure 12: Police relationship with the community - Post-test
The following figure 13 shows the descriptive statistical results of the Police Relations with the Community
indicator according to the results in the statistical software SPSS (Spanish version):
Figure 13: Descriptive analysis - Police relationship with the community
For the Police relationship with the community indicator, in the Pre-Test an average value of police relationship
with the community of 44% was reached. In the same way, after the implementation of the detection system based on
facial recognition, an average value of police relationship with the community of 93.33% was reached, which
indicates that the implementation of said system generates a relevant increase in the indicator. On the other hand, in
the Pre-Test, the standard deviation had a value of 0.13; and in the Post-Test, this had a value of 0.09, which is why it
can be concluded that the standard deviation decreased, which means that the dispersion of the degree of similarity
has decreased [29].
The table 2 shows the comparison of the results that were taken before carrying out the simulation and after the
simulation. The data shared by the PNP of Peru was used as a pre-test and the data that was collected within the
simulation was used as a post-test, having a non-probabilistic sample due to the pandemic [30], considering a total of
15 adults.
Table 2: Comparative table of pre-test and post-test
N° Degree of similarity Police relationship with the community
Pre-test Post-test Pre-test Post-test
1 40% 87,88% 40% 80%
2 35% 95,45% 60% 100%
3 35% 89,39% 40% 100%
4 30% 96,97% 40% 100%
5 25% 93,94% 60% 100%
6 20% 90,91% 60% 80%
7 45% 90,91% 40% 80%
8 25% 96,97% 20% 100%
9 35% 89,39% 40% 100%
10 20% 87,88% 60% 100%
11 35% 89,39% 40% 100%
12 55% 86,36% 20% 80%
13 45% 71,21% 60% 100%
14 20% 89,39% 40% 80%
15 15% 86,36% 40% 100%
The approach to the Luxand Face facial recognition system aimed at the detection of people in the criminalistics
unit of the PNP in Huancayo city has the following analysis:
It increases the average degree of similarity, thus improving the recognition of people more quickly, with the
detection of facial points [31]. Apart from this, the police relationship with the average community increases, which
shows a greater trust of people with the police regarding the process of detecting people with the use of the system.
4 discussion
For the discussion, we took the article "Development of a Security System Based on Facial Recognition Oriented to
the Management and Diversion of Criminal Attacks" as a reference, to compare the methodology and results [4]. The
methodology that was used in our article is the waterfall model engineering methodology, which has 4 phases that are
requirement analysis, design, construction, and testing. On the other hand, in the other article it is the extreme
programming methodology which also has 4 phases of planning, design, coding and testing. Both methodologies are
similar in that they allow you to emphasize the adaptability of a project; the main difference that is made in these two
methodologies is that the other article in phase 4 is focused on the programming of the facial recognition system and
in our article, it is based on the installation of an already established program that meets all the requirements.
requested by users to then test and validate its operation. Regarding the results, in our article we measure the
indicators of the degree of similarity and of the police relationship with the community, unlike the other article that
only measures the degree of satisfaction of the security software. Although the results of police relations with the
community and satisfaction with the security software were obtained using a similar tool, and the results in both cases
were 93%, it is necessary to highlight that the comparative article did not use an indicator to measure the
development of the system already installed, to have accurate data on the functionality of the system. On the other
hand, our article shows a pre- and post-installation evaluation of the system. According to everything detailed, our
article makes use of an accurate methodology and shows the calculations of the indicators to measure the
implementation of the system, for which it is concluded that its implementation in the criminalistics unit of the PNP
in Huancayo, Peru is favourable.
5 conclusions
The article carried out shows that the development of the implementation of the facial recognition system Luxand
Face, oriented to the detection of people efficiently influences the improvement of the process of detecting people for
the criminalistics unit of the PNP in Huancayo city, Peru. The development of the implementation of the facial
recognition system Luxand Face oriented to the detection of people efficiently influences the degree of similarity,
thus allowing more accurate and reliable results to be obtained, in the same way, this approach positively influences
the police relationship with the community, allowing the population to have greater confidence and security in the
police of their city. We are currently in the process of implementing the Luxand Face facial recognition system in the
criminalistics unit of the PNP in Huancayo city. In this way, we will make a second article that shows the
implementation made.
REFERENCES
[1]Noam Lopes. 2017. The impact of the video surveillance system on the levels of violent crime in zone 1 of
Lima Cercado 2011- 2014. Lima: UNI-Tesis http: hdl.handle.net/20.500.14076/6847
[2]Cristina Domingo. 2021. Use of the facial recognition system to preserve citizen security. s.l.: The digital
criminalist. Criminology papers.ISSN 2340-6046
[3]Paolo Contardo, Paolo Sernani, Selene Tomassini, Paolo Castellini, and Aldo Dragoni. 2023. FRMDB: Face
Recognition Using Multiple Points of View, Sensors 23(4),1939 http://doi.org/10.3390/s23041939
[4]Alejandro Boza-Chua, Karen Gabriel-Gonzales, Enrique Huamaní, and Alexi Delgado. 2022. Development of
a Security System Based on Facial Recognition Oriented to the Management and Diversion of Criminal Attacks. International Journal of Emerging
Technology and Advanced Engineering 12(2), pp. 48-54 http://doi.org/10.46338/ijetae0222_06
[5]Anita Gehlot, Rajesh Singh, and Poonam Negi. 2022. Image Processing Based Security Management System.
International Interdisciplinary Humanitarian Conference for Sustainability, IIHC 2022 - Proceedings, pp. 974-979
http://doi.org/10.1109/IIHC55949.2022.10059901
[6]Indo Intan, Nurdin Nurdin, and Fitriaty Pangerang. 2023. Facial recognition using multi edge detection and
distance measure. IAES International Journal of Artificial Intelligence, 12 (3), pp. 1330-1342 http://doi.org/10.11591/ijai.v12.i3.pp1330-1342
[7]Ho Nguyen Anh Tuan, Nguyen Dao Xuan Hai, and Nguyen Truong Thinh. 2022. The Improved Faster R-
CNN for Detecting Small Facial Landmarks on Vietnamese Human Face Based on Clinical Diagnosis. Journal of Image and Graphics, Vol. 10, No. 2,
pp. 76-81
[8]Gaurav Srivastav and Richa Singh. 2022. Security system in the workplace based on facial recognition that
uses the LBPH algorithm. AIP 2555 Conference Proceedings, 040008 http://doi.org/10.1063/5.0124629
[9]Jay Mehta, Shreya Talati, Shivani Upadhyay, Sharada Valiveti, and Gaurang Raval. 2023. Regenerating vital
facial keypoints for impostor identification from disguised images using CNN. Expert Systems with Applications 219,119669
http://doi.org/10.1016/j.eswa.2023.119669
[10]Artak Melkumyan and Katya Mkrtchyan. 2023. Facial recognition and tracking system in real time using
drones. Proceedings - IEEE Consumer Communications and Networking Conference, CCNC 2023-January, pp. 975-976
http://doi.org/10.1109/CCNC51644.2023.10059664
[11]Bindhu Sri, Prudhvi Raju, K. Vydehi, and B. Kiruthika. 2023. Face detection system for smart security
application. Paper presented at the Advances in Transdisciplinary Engineering, , 32 651-657. http://doi.org/10.3233/ATDE221327
[12]Luiggi Alvarez and Bryan Orellana. 2021. Face Recognition for Criminal Identification: An Alert System in
Suspects Scenarios. Lima : Universidad Peruana de Ciencias Aplicadas (UPC)
[13]Cristian Bravo, Patricio Ramirez, and Jorge Arenas. 2018. Acceptance of Face Recognition as a
Surveillance and Safety Measure: An Empirical Study in Chile. vol.29, n.2, pp.115-122. ISSN 0718-0764. http://doi.org/10.4067/S0718-
07642018000200115
[14]Silvia Leal. 2021. Trends: Keys to understand the future and build the present. Spain: Punto Rojo Libro.
9788418654367. ISBN: 9788418654367
[15]Md Rahman, Mir Jannat, Md Islam, Sathya Bursic, and Md Aktaruzzaman. 2023. Real-time face mask
position recognition system based on the MobileNet model. Smart health. 28, 100382 http://doi.org/10.1016/j.smhl.2023.100382
[16]Yomna Alayary, Nadeen Shoukry, Mohammed El Ghany, and Mohammed Salem. 2022. Face Masked and
Unmasked Humans Detection and Tracking in Video Surveillance.NILES 2022 - 4th Novel Intelligent and Leading Emerging Sciences Conference,
Proceedings, pp. 211-215 http://doi.org/10.1109/NILES56402.2022.9942375
[17]Robert Barreto and David Lizarraga. 2019. Facial recognition system model for the control of human
trafficking. Arequipa : s.n. http://hdl.handle.net/20.500.12867/2063
[18]Roberto Hernandez, Carlos Fernandez, and Maria Baptista, M. 2014. Investigation methodology. Mexico
City: McGrawHill, 2014. 978-1-4562-2396-0
[19]Manuel Ortega and Edgard Camacho. 2019. Use of traditional models and agile methodologies applied in
the Colombian software industry. Cali : Universidad Santiago de Cali
[20]Maryam Bukhari, Sadaf Yasmin, Sheneela Naz, Muazzam Maqsood, Jehyeok Rew, and Seungmin Rho.
2023. Language and vision based person re - identification for surveillance systems using deep learning with LIP layers. Image and Vision Computing,
132, art. no. 104658 http://doi.org/10.1016/j.imavis.2023.104658
[21]Lissette Nuñez. 2013. Interconnection of networks by routers. IDICT Información y gestión tecnológica.
ISSN 1029-3450
[22]Wolfram Wingerath., Norbert Ritter, and Felix Gessert. 2019. Database managment. SpringerBriefs in
Computer Science, pp. 9-19. http://doi.org/10.1007/978-3-030-10555-6_2
[23]Mehmet Korkmaz and Nihat Yilmaz. 2016. Face Recognition by Using Back Propagation Artificial Neural
Network and Windowing Method Journal of Image and Graphics, Vol. 4, No. 1, pp. 15-19. http://doi: 10.18178/joig.4.1.15-19
[24]Alberto Ramírez and Miguel Casillas. 2015. Internet en Educación Superior. Córdoba, Argentina: Editorial
Brujas, 204 p. http://doi.org/10.1016/j.resu.2015.08.001
[25]Alejandro E. Meza, Aida Melgar and Hipólito Aguirre. Guia de procedimientos criminalísticos PNP. EM -
DIRCRI PNP. Lima, Peru.
[26]Waqar Ali, Wenhong Tian, Salah Din, Desire Iradukunda, and Abdullah Khan. 2021. Classical and modern
face recognition approaches: a complete review. Multimedia tools and applications,80 (3), pp. 4825-4880. Cited 38 times.
http://doi.org/10.1007/s11042-020-09850-1
[27]Yoshihiro Sato and Yue Bao. 2018. 3D Face Recognition without Using the Positional Relation of Facial
Elements. Journal of Image and Graphics. Vol. 6. No. 1, pp. 33-38. http://doi: 10.18178/joig.6.1.33-38
[28]Luis Blázquez. 2013. Facial Recognition Based on Characteristic Points of the face in uncontrolled
environments. Madrid : Universidad Autónoma de Madrid
[29]Alejandra Mohor. 2017. Use of indicators to evaluate police performance. Santiago de Chile : Centro de
estudios en seguridad ciudadana. Universidad de Chile
[30]Edison Cabezas, Diego Andrade, and Johana Torres. 2018. Introduction to the Methodology of Scientific
Research. Ecuador : Universidad de las Fuerzas Armadas ESPE
[31]Omer Naser, Sharifa Ahmad, Khairulmizam Samsudin, Siti Shafie, and Nor Zamri. 2023. Facial recognition
for partially occluded faces. Indonesian Journal of Electrical Engineering and Computer Science.30(3), pp. 1846-1855
http://doi.org/10.11591/ijeecs.v30.i3.pp1846-1855
http://doi.org/10.11591/ijai.v12.i3.pp1330-1342