See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/353295394

Criminal Identification System using Facial Recognition

Conference Paper · July 2021

CITATIONS READS
4 9,343

4 authors:

Nagnath Aherwadi Deep Chokshi

Lovely Professional University Lovely Professional University
4 PUBLICATIONS 29 CITATIONS 2 PUBLICATIONS 8 CITATIONS

SEE PROFILE SEE PROFILE

Sagar Dhanraj Pande Aditya Khamparia

VIT-AP University Babasaheb Bhimrao Ambedkar University
92 PUBLICATIONS 870 CITATIONS 155 PUBLICATIONS 3,766 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Deep Chokshi on 16 July 2021.

The user has requested enhancement of the downloaded file.

Criminal Identification System using Facial Recognition

Nagnath B. Aherwadi1, Deep Chokshi2, Sagar Pande3, *, Aditya Khamparia4

[email protected], Lovely Professional University, Punjab, India1.
[email protected], Lovely Professional University, Punjab, India2.
[email protected], Lovely Professional University, Punjab, India3.
[email protected], Babasaheb Bhimrao Ambedkar University, India4

Abstract. We all know that our Face is a unique and crucial part of the human body structure that identifies
a person. Therefore, we can use it to trace the identity of a criminal person. With the advancement in
technology, we are placed CCTV at many public places to capture the criminal’s crime. Using the previously
captured faces and criminal’s images that are available in the police station, the criminal face recognition
system of can be implemented. In this paper, we propose an automatic criminal identification system for
Police Department to enhance and upgrade the criminal distinguishing into a more effective and efficient
approach. Using technology, this idea will add plus point in the current system while bringing criminals
spotting to a whole new level by automating tasks. Technology working behind it will be face recognition,
from the footage captured by the CCTV cameras; our system will detect the face and recognize the criminal
who is coming to that public place. The captured images of the person coming to that public place get
compared with the criminal data we have in our database. If any person’s face from public place matches, the
system will display their image on the system screen and will give the message with their name that the
criminal is found and present in this public place. This system matching more than 80% of the captured
images with database images.
Keywords: Face detection, Face recognition, Open-CV, Image Processing.

1 Introduction involved in this system is automated.

Criminal identification is the most important Identification System

task for the Police who are finding the
criminals, but it is the difficult and most
time-consuming task as they have to find it
everywhere. It will be more difficult in cities
or public places with high people density. In Manual Identification Automated
System Identification System
some cases, manual type of identification (MIS) (AIS)
gives chance for getting more information
related to criminals. Hence this paper Figure1. Types of
proposes an automatic criminal identification Identification system
system by detecting the face of criminals.
This will help Police to identify and catch the Automated Criminal identification monitoring
criminals in public places. system’s some important things shown below:
1. Criminal Enrolment :
Criminal identification can be done Criminal images with their name to
in two ways, which is shown in figure 1. In photos are added to the criminal
Manual Identification System (MIS), database so that we can compare the
identification is done by the Police officers captured images with database one.
searching them at public places. It takes a lot
of time to give the proper attention and it 2. CCTV Connectivity :
also has the chances of skipping criminals as CCTV Cameras should be connected to
they will be alerted by seeing cops easily gets the system on which we are having a
escape from there. Since the MIS is in the criminal database and program where
process of taking more time and we will not we are running.
properly focus on everyone. But when it
comes to an automated identification system 3. Criminal Confirmation :
(AIS) there is no need for observation going If a person is found from a public place
in a public place. Here all the process by using this system, then check who

1
Electronic copy available at: https://ssrn.com/abstract=3884827
 was the criminal using a special folder
available on the desktop.

2 Ease of Use

2.1 Open-CV

Open-CV is Open-Source Computer Vision

Fig.2. Facial mapping on a captured image
Library. The library contains 2500+
algorithms that are optimized which include
a comprehensive set of both classic and state- 3 Literature Survey
of-the-art computer vision and machine
[1] In this paper, the authors are taking help
learning techniques. Also, it has C++,
of the CCTV footage and comparing the
PYTHON, JAVA, and MATLAB interfaces
images from the footage with criminal
which support Windows, Linux, Android,
database if they didn’t find any fingerprint
and Mac-OS. For commercial and non-
from the crime scene. This system consists of
commercial, Open-CV is free for use. Open-
five stages where the first stage is planning in
CV is used for capturing the images and
which the why and how the system is made
videos in public-place.
are discussed. The second stage of
2.2 Face Detection Requirement analysis discussed the
requirement to design the system. Design, the
The primary function of this step is to third stage where they defined system design
capture the faces of the people who are and its workflow. The fourth ultimate
available in front of the camera. The outputs important stage is Implementation and
from this step are patches that contain each testing, system is implemented using
face in the input image. To design a perfect Principal Component Analysis (PCA)
and preferable face recognition system. Face Technique and tested. The last stage is
alignment is performed to rationalize the maintenance; this phase hadn’t undertaken
scales and orientation of these patches. due to this system was developed in a
Further Next step after the face detection step controlled environment. For criminal
is human face patches are extracted. identification, authors had used PCA
Technique for finding similar features of
2.3 Face Recognition images available in the database with
captured images of footage. The machine
Face recognition is a method of identifying will use a database that contains the person's
or verifying the identity of an individual personal information so that if FRCI
using their face. The step after the identifies a face, it can display the person's
representation of faces is to identify them. In information. The system interface is
this comparison of the detected face image implemented using Visual Studio Code and
with the images, we have in our database database and coding using MATLAB
based on face encodings. R2013b. They achieved 80% accuracy using
A facial recognition system maps the proposed model.
facial expressions from an image or video [2] This paper consists of four steps, the first
using biometrics. To find known faces match one is real-time image training and the
from the database, it compares the details to second one is Harr-classifier using for face
a database. Facial recognition may aid in the detection. The third step is the comparison of
identification of personal identity, but it also Surveillance camera captured images with
introduces privacy concerns. Commercial real-time images and last, is the result part
applications use facial recognition as well as based on the comparison. The authors are
it is used for a variety of purposes ranging using the Haar-classifier on Open-CV for
from security to promotions. face detection; Haar-cascading is one of the
algorithms for face detection. On the open-
CV platform, face tracking is taken with help
of Harr-like classifiers. More than one person
is identified in this system and it can be used
2
Electronic copy available at: https://ssrn.com/abstract=3884827
to find the suspects whom we are finding. using machine learning technology with an
The accuracy of the proposed system is very SVM classifier for name detection and
high as compared to the previous model. gradient-oriented Histogram for face
They also told us that we use or Adhar detection. They are using open-CV for image
database we can easily identify the Indians detection & recognition, Tkinter for GUI
and foreigners and further can investigate application creation, and Numpy to work
whether a person is a criminal or not. We can with arrays as those are libraries of python.
use this system by taking the citizenship To develop and test the application using the
database which is already available. Xampp server, as it is a free open source
server. There proposed model has achieved
[3] In this paper Using the Passport database,
an accuracy of 99.38%. Using Cloud
they are identifying whether the traveler is an
feasibility of the system can be increased.
authorized passport holder or not. In this,
they are using image processing techniques [6] In this paper, the authors had discussed
as well as LBPH mathematical model. This that an attendance monitoring system is very
method consists of six steps for airport important in the teaching and learning
security purpose that are: a) Capture image process. The student who is entering in
using webcam b) Captured image is sent to classroom his/her image is captured. Pre-
the Django server c) Using LBPH feature set processing and Face region extraction take
is taken from image d) Image is compared place using that captured image for further
with database image by applying classifier e) process. They are using a face recognition
If matching is done user details are fetched algorithm for marking present if the student
from database f) The predicted details of the came to school or absent if the student is not
user are sent to the admin via mail. They are coming to school. They are capturing the
using webcam images for LBPH processing student’s image using a camera and after
and then applying the classifiers comparing preprocessing comparing with their student
them with database images. This will also database and marking attendance.
help to catch the criminals who travel from
[7] In this paper, the authors have presented a
one country to another and also detect if the
face identification system that uses the fast
traveler having a loan from the bank then
algorithm. This model uses two datasets: 1)
traveler's detailed information will be sent
Olivetti Research Laboratory (ORL), 2)
for verification to the police station.
Unconstrained Facial Images (UFI). ORL
[4] In this research Portraits of Soldiers from contains 400 images of 92X92 pixels where
the Civil War of America in 1861-65 are 9 images are used for training purposes and 1
identified by creating a web-based platform image for testing of each person. UFI
i.e., Photo Sleuthing. They told that this contains 401 images of 128X128 pixels
identification system is taken as finding a where 7 images are used for training
needle in the haystack. Where it contains 1) purposes and 1 image for testing of each
haystack building 2) the haystack narrowed person. Captured image converted to HSV
down and 3) needle is finding in the system and after that force field features is
haystack. Work is by combining automated extracted from that image. Classification is
face recognition and crowd-sourced human done by using three distance methods that
expertise. When this method was launched it are: Manhattan, Euclidean, and Cosine. By
helped to identify the unknown portraits and comparing these methods they got the best
authors discussed implications for person resolution and achieved an accuracy of
identification pipelines. They show how 99.9% for the datasets ORL and UFI.
Photo Sleuth's pipeline has helped to identify
thousands of unknown images and also 4 Proposed Work
encouraging long-term volunteer
contribution. In this paper, we are Using CCTV cameras
which are continuously working in a public
[5] The authors of this paper are presenting
place. In the Implementation of the system,
an automatic face recognition system for
we already saved criminal's images data with
attendance monitoring. They are capturing
their names on photographs in the database.
faces by using a camera and the captured
We are processing those images and
image is compared with images that are
extracting features from them and in feature
already present in their database. They are
3
Electronic copy available at: https://ssrn.com/abstract=3884827
extraction; we are taking the face encodings
of the present images and saving them into
one file using Pickle. Using open-CV while
capturing the footage in CCTV and captured
images face encodings taken placed and
comparing with our saved face encodings of
the criminal database if any match is found
then automatically on screen it will display
an image of that criminal whose face
matches and display the message with his Fig.4. 68 Landmarks on Every Face
name that criminal found, and his captured
image will be saved into special folder police CCTV camera captured footage or image is
will go and catch him from that public place stored and its features i.e., encodings are
even if he once captured in the CCTV extracted after that it will be compared with
footage. As we saved the person's image in the image encodings which are available in
the special folder from there police will come the database. The process of matching the
to know whose image was matched with face will take place in the database; the name
captured once. and the criminal found message will be
displayed with the criminal image on the
screen present in CCTV Room.
Capture from CCTV Working: We are using CCTV Camera to
capture the images of the Public so that we can
identify the proper person who is having a
criminal record in the database to catch.
Feature Extraction 1. First, we will find out the face
encodings of the criminal database
images and save those into one list, and
DAT Image Comparison splitting the name which is saved with
A
criminal image saving into another list.
2. Now we are using CCTV to capture
Public images so that we can identify
Get details & Save the criminal who is present in a public
place and easily can catch
3. Extracting the features from captured
images i.e., taking the face-encodings
Display Criminal of the captured images.
Details 4. Comparing Captured image encoding
values with our database image
Fig.3. Execution Flow of System encoding values.
5. If the encoding values matching with
captured image encoding values, then
Criminal enrolment is the initial step in criminal image, name, and criminal
which we are storing the criminal found message will be displayed on the
information using the criminal database. screen.
Here criminal names with their image will 6. Image of that person will be saved
get stored in the database. By using this into a special folder on the desktop, so
image and information, face recognition and that police can easily identify the
identification will take place. After the face criminal who differs from other
detection process i.e., using face encodings peoples present in a public place.
of images will take place. Criminal Enrolling
is the database process, but the main process 5 Applications
starts from face detection. Face detection is
done considering 68 landmarks available on We can use this system Anywhere for
the face as shown in figure 4. Security Purpose for ex. in the Jewellery shop,
Banks, Hotels, Airports, Restricted areas,

4
Electronic copy available at: https://ssrn.com/abstract=3884827
Religious Places and Famous places where
maximum crowd use to gather.

6 Experimental Results

We have proposed a promising Criminal

Detection system for Face Videos. CCTV
Cameras are used for continuous capturing of
the video and images; we will get the
information on our main screen that which
image from the database is matching.
When the database image matches with CCTV
captured image then on the main screen the
name of the criminal with the criminal found Fig.6. Database images which are saved in a special
message will be displayed as shown below in folder when criminal found
figure 5;
7 Conclusion and Future Work

This upgraded version of the criminal

detecting system not only provides a huge
convenience to the Police in the
identification of criminals but also saves time
for them as processes are automated in the
system. The novelty of this Research Paper is
face detection done by using Face
Encodings.
For future work, we can add the Alarms
to the criminal detection system. It will range
only when matches are found so that if
anyone is not there to keep watch in the
CCTV room, they will come to know that
someone is found from the database in that
public place. [21]This paper presents a
Fig.5. Camera images with name and Criminal found
Message surveillance system that will give us alerts
when any controversy, fight, or intruder is
When a criminal is identified using a detected by using CCTV footage.
CCTV image, then which criminal found from
the database that criminal image will be saved
in Special folder on desktop as shown in figure 8 Acknowledgement
6. Police will come to know that who is found
We are very thankful to Prof. Sagar Pande
in a public place because he will not be there in
sir, CSE Dept. for his guidance throughout
front of a CCTV camera for a long time. So
the project process. We are thankful to
this approach more helpful when criminals
everyone who has helped us directly or
come for few minutes in front of a CCTV
indirectly in this project work.
camera that one will be identified.

References
[1] Nurul Azma Abdullah, Md. Jamri Saidi, Nurul
Hidayah Ab Rahman, Chuah Chai Wen, and Isredza
Rahmi A. Hamid’s “FACE RECOGNITION FOR
CRIMINAL IDENTIFICATION: AN
IMPLEMENTATION OF PRINCIPAL
COMPONENT ANALYSIS FOR FACE
RECOGNITION”, AIP Conference Proceedings
1891, 020002

5
Electronic copy available at: https://ssrn.com/abstract=3884827
 (2017); https://doi.org/10.1063/1.5005335
[2] Apoorva. P, Impana. H.C, Siri. S.L, Varshitha.
M.R and Prof. Ramesh. B’s “AUTOMATED
CRIMINAL IDENTIFICATION BY FACE
RECOGNITION USING OPEN COMPUTER
VISION CLASSIFIERS”, Third International
Conference on Computing Methodologies and
Communication (ICCMC 2019), DOI:
10.1109/ICCMC.2019.8819850
[3] Prof. Rupali T. Umbare, Ms. Janhavi S.
Takalgavankar, Ms. Harshada S. Yadav, and
Ms.Pruthvi A. Tilekar's “AIRPORT SECURITY
USING FACE-RECOGNITION”, International
Journal of Future Generation Communication and
Networking, vol. 13, No. 3, (2020).
[4] Vikram Mohanty, David Thames, Sneha Mehta,
and Kurt Luther, “Photo Sleuth: Combining Human
Expertise and Face Recognition to Identify
Historical Portraits”, Conference: the 24th
International Conference, March
2019, https://doi.org/10.1145/3301275.3302301
[5] Dr. Asif Ali, Radhika Mandhanya, Shraddha
Birla, Ujjwal Mandloi and Vipul Jain’s “Automatic
Face Recognition Attendance System using Python
and OpenCV”, GRD Journals- Global Research and
Development Journal for Engineering | Volume 6 |
Issue 4 | March 2021
[6] P. Kowsalya, J. Pavithra, G. Sowmiya and C.K.
Shankar’s “ATTENDANCE MONITORING
SYSTEM USING FACE DETECTION & FACE
RECOGNITION”, International Research Journal of
Engineering and Technology (IRJET), Volume: 06
Issue: 03 | Mar 2019.
[7] Kian Raheem Qasim and Sara Salman Qasim’s,
“Force Field Feature Extraction Using Fast
Algorithm for Face Recognition Performance”, Iraqi
Academics Syndicate International Conference for
Pure and Applied
Sciences, https://doi.org/10.1088/1742-
6596/1818/1/012195

6
Electronic copy available at: https://ssrn.com/abstract=3884827
View publication stats