PAAVAI COLLEGE 0F ENGINEERING

Pachal,Namakkal-637018

AN INTELLIGENT PERSON IDENTIFICATION FOR CRIMINAL DETECTION

GUIDED BY: PRESENTED BY:

MRS.M.MENAGA M.E (AP) KANISHKARAN E-622020243018
DEPARTMENT OF AI&DS SAKTHIVEL S-622020243032
PAAVAI COLLEGE OF ENGINEERING SANJAI R-622020243033
NAMAKKAL-637018 SIVASANKAR K-622020243036
 Abstract
• Face recognition is an interesting and challenging problem and impacts
important applications in authentication and personal identification among
others.
• Here implement a application for criminal detection, It helps forensic
department for the accurate identification of criminal using his face image.
• Criminal images are collected and stored by forensic department.
• During capturing process, face image will be classified with the criminal image
database.
• If match is found then we can get to know who is criminal and can arrest that
criminal easily.
 Objective

• To provide efficient facial recognition system for criminal database using Haar
& cascade classifier.
• Send immediate alert to the forensic department, when the criminal was
detected.
 Introduction
• Face recognition is a rapidly growing field in machine learning, and it has a wide range of

applications in various industries.

• From security and surveillance to entertainment and social media, face recognition

technology can revolutionize how we interact with technology.

• Face recognition is an identification method that uses the individual's face's distinctive

features to identify them.

• A machine learning model can be trained to recognize faces by being fed a large dataset of

images of faces.

• Automatic face recognition is all about extracting those meaningful features from an

image, putting them into a useful representation and performing some classifications on

them.
 Existing System
• Implement a video surveillance system based on a composable deep face recognition
method.
• A face detection method is proposed for processing video frames in real-time through
down-sampling.
• A precise facial landmark or segmented mask must be acquired, as a face boundary
needs to be estimated to perform face alignment.
• Deformable convolutional network (DCN) to construct a general deep CNN as the
feature extractor.
• Euclidean distance calculation between feature vectors can be vectorized for closely
matching feature identification.
• This system detects the faces of criminals in real time from videos captured by a
surveillance camera and notifies relevant institutions of the appearance of criminals.
 Disadvantages
• When the resolution increases, the depth of the feature extractor and

latency also increases.

• In face identification process, the number of targets has little impact on the

latency.

• Extremely high labor cost is required for data labeling

• Unable to define the answer label when multiple persons are caught in the

sampled individual video.

 Proposed System
• Develop a system that can be used by police or investigation department to recognize

criminal from their faces.

• The face detection algorithm looks for specific Haar features.

• When one of these features is found, the algorithm allows the face candidate to pass to the

next stage of detection.

• The feature are classified with dataset using Haar Cascade Feature Classification.

• It is further compared with the images stored in database to identify if the person is

criminal/suspect.

• If he is criminal/suspect the time for which he was under the surveillance of the camera is

noted and provide notification.

• This application send SMS and Email notification to the authority and also generate alarm

sound when the criminal was detected.

 Advantages
• The method of face recognition used is fast, robust, reasonably simple and
accurate with a relatively simple and easy to understand algorithms and
technique.

• This method works very efficiently because face image of a person will not
change.

• The proposed system can successfully recognize more than one face which
is useful for quickly searching suspected persons as the computation time is
very low.
 Hardware Requirements

• Processor : Intel Pentium 4

• RAM : 4 GB
• Hard disk : 160 GB
• Compact Disk : 650 Mb
• Keyboard : Standard keyboard
• Monitor : 15 inch color monitor
 Software Requirements

• Operating system : Windows OS

• Front End : HTML, CSS
• Back End : PYTHON, MySQL Server
• Tool : PyCharm
Architecture Diagram
 Modules List

• Criminal Image Storage

• Face Image Acquisition
• Feature Extraction
• Criminal Detection
• Alert System
 Modules Description

• Criminal Image Storage

• This is the process of collect and store criminal image for the detection of
criminals easily.

• The database will contain the images and details of the criminals and
suspects who involved in crimes.

• Then it undergoes feature extraction using Haar cascade algorithm.

• It is further compared with the images captured in real time to identify if

the person is criminal/suspect.

• Admin can update the system with new criminal images.

 Modules Description

• Face Image Acquisition

• With the advent of security technology, cameras especially CCTV have
been installed in many public and private areas to provide surveillance
activities.
• The footage of the CCTV can be used to identify suspects on scene.
• In this module, we capture the face image or upload the datasets.
• The face image which is captured by web camera.
 Module Description

• Feature Extraction
• The face detection algorithm looks for specific Haar features and not pixels
of a human face.
• When one of these features is found, the algorithm allows the face candidate
to pass to the next stage of detection.
• A face candidate is a rectangular section of the original image which is
called as a sub-window.
• A simple rectangular Haar-like feature can be defined as the difference of the
sum of pixels of areas inside the rectangle, which can be at any position and
scale within the original image.
 Modules Description

• Criminal Detection
• The cascade classifier contains a list of stages, where each stage consists of a
list of weak learners.
• The system detects the required object by moving a window over the image.
• Each stage of the classifier labels the specific region defined by the current
location of the window as either positive or negative where positive means
that an object was found and negative means that the specified object was not
found in the image.
 Module Description

• Alert System
• The automatic detection of abnormal activities can be used to alert the
related authority of potential criminal or dangerous behaviors, such as
automatic reporting of a person.
• In proposed system unknown event alert send to the predefined contact
numbers regarding particular officers.
• Here also implement image sharing for easy identification of criminals.
SCREENSHOTS
 Conclusion

• To detect and recognize faces of the criminals in an image and in a video

stream obtained from a camera in real time.
• Providing alert to the corresponding authority to track the criminals
inefficient manner.
 Future Enhancement

• As the future scope of this system can be extended further by adding

additional IoT features to detect the people face if they wore the mask on
his/her face.
• And also implement various algorithms to provide still to video face
matching with improved accuracy rate.
 References
• [1] Kwan-Loo, Kevin B., José C. Ortíz-Bayliss, Santiago E. Conant-Pablos, Hugo
Terashima-Marín, and P. Rad. "Detection of violent behavior using neural networks and
pose estimation." IEEE Access 10 (2022): 86339-86352.
• [2] Adil, Muhammad, Saqib Mamoon, Ali Zakir, Muhammad Arslan Manzoor, and
Zhichao Lian. "Multi scale-adaptive super-resolution person re-identification using
GAN." Ieee Access 8 (2020): 177351-177362.
• [3] Alansari, Mohamad, Oussama Abdul Hay, Sajid Javed, Abdulhaid Shoufan, Yahya
Zweiri, and Naoufel Werghi. "GhostFaceNets: Lightweight Face Recognition Model
From Cheap Operations." IEEE Access (2023).
• [4] Deng, Jiankang, Jia Guo, Evangelos Ververas, Irene Kotsia, and Stefanos Zafeiriou.
"Retinaface: Single-shot multi-level face localisation in the wild." In Proceedings of the
IEEE/CVF conference on computer vision and pattern recognition, pp. 5203-5212. 2020.
• [5] Deng, Jiankang, Jia Guo, Niannan Xue, and Stefanos Zafeiriou. "Arcface: Additive
angular margin loss for deep face recognition." In Proceedings of the IEEE/CVF
conference on computer vision and pattern recognition, pp. 4690-4699. 2019.
THANK YOU