Criminal Identification Using Raspberry Pi: Vivek Kalaskar, Amit Vajrashetti, Swanand Zarekar, Ramij Shaikh
Criminal Identification Using Raspberry Pi: Vivek Kalaskar, Amit Vajrashetti, Swanand Zarekar, Ramij Shaikh
Criminal Identification Using Raspberry Pi: Vivek Kalaskar, Amit Vajrashetti, Swanand Zarekar, Ramij Shaikh
Abstract –This paper presents a real time face recognition camera, robotics etc. This technology is used for
using an automated surveillance camera. The proposed authentication, validation, authorization, and identification. In
system consists of 4 steps, including (1) training of real time developed countries, the government creates a datasets which
images (2) face detection using Haar-classifier (3) is helpful for recognize the human face which compares the
comparison of trained real time images with images from the suspicious act with trained dataset and information stored in
surveillance camera (4) result based on the comparison. An database.
important application of interest is automated surveillance, Face identification [1] is defined in three steps (1) face
where the objective is to recognize people who are on a detection (2) feature extraction (3) face recognition. Camera
watch list. The aspiration of this paper is to compare an configuration is very important to track moving persons and
image with several images which has been already trained. recognize [2] them precisely. Facial feature points encode
In this paper, this system represents a methodology for face critical information about face shape. Precise location and
detection robustly in real time environment. Haar cascading facial feature points tracing are important. Each feature point
is one of the algorithms for face detection. Here system uses is usually detected and traced by performing a local search for
Haar like classifiers to track faces on OpenCV platform. The the better matching position [3]. There are very less researches
accuracy of the face recognition is very high. The system on face recognition using edge-based detection [4]. The edges
can successfully recognize more than one face which is are not only carrying valuable data about face but are also
useful for quickly searching suspected persons as the simple to process. The Viola Jones method builds a classifier
computation time is very low. In India, we have a system for by selecting a few significant features using AdaBoost. Viola
recognizing citizen called Aadhaar. If system uses this as a jones method successfully merges more composite classifiers
citizenship database, it can differentiate between citizen and in cascade structure [5] which exponentially increases speed
foreigner and further investigate whether the identified of detector by focusing on the favourable features of the face.
person is criminal or not.
II. METHODOLOGY
Keywords: Raspberry Pi, Surveillance Camera, Face
Recognition, Image Masking, OpenCV. This paper is going to develop face detection and recognition
system that is capable of processing images very fast while
I. INTRODUCTION acquiring very high true positive face detection rates. Object
recognition frameworks have been tried crosswise over
Face recognition which is a combination of machine learning different standard face databases, with and without noise and
and the biometric techniques which holds the qualities of not other obscuring effects. The outcome of Object recognition
only high precision but also the reliability. For automatically frameworks uncover that well utilized face recognition even
detecting the human’s face from the databases this system can from low quality pictures and shows astounding execution
be used. In recent years open computer vision has been widely productivity, monitor threats, and avoid/examine criminal
used in different kinds of applications such as surveillance
1
National conference on ____________________________________________________________________________
Paper Title:-Criminal Identification using Raspberry Pi
Criminal Identification Using Raspberry Pi is mainly about This section pays special attention to the main consequences
maintaining a secure environment. OpenCV (open source of face recognition for criminal identification. This deals with
computer vision) is the major software that is being used in outline development and testing the Face recognition. Here,
this project. For detecting faces, system using various software’s out there that resemble the same services as the
algorithms like Haar cascade, linear SVM, deep neural face recognition system, however, this also has its own merits
network, etc. The main method that is proposed in this work that can contribute to the society. Due to its intricacy and
is, if any person comes in front of the pi camera, first it will revelation building, the recognition of the automated face
look for potential matches that we have already stored in this identification tool might be burdensome. Moreover, there is
system. If the module finds a match then it captures the person also possibility for giving the interface as user – friendly
and sends an alert to the Headquarter. which wants the minimal amount of interaction among the
users. With this system, which automatically processes the
image, users need only requires to offer input a real time
image & the system will do the rest.
Figure 1: System Architecture the thumb print. Criminal photo captured through a video
source is fed the identification of the system. This system has
In this paper, the video obtained from the surveillance camera the capacity for automatically doing the process like
will be converted into frames. When a face is detected in a recognize, detect and extract the features of the images and
frame, it is pre-processed where noise and redundancies are identifies the actions.
ISSN:-2349-3585 |
3
www.ijrdt.org
Paper Title:-Criminal Identification using Raspberry Pi
ISSN:-2349-3585 |
3
www.ijrdt.org