Centralized Smart Security System

CSS-System is basically a desktop-based application project, it aims to provide a clear and focused user environment for authentication of user?s biometric verification of fusion of serval models. Although, its end users are mainly specified as Bank Managers, Server room holders, or any company requ

Project Title

Centralized Smart Security System

Project Area of Specialization

Artificial Intelligence

Project Summary

CSS-System is basically a desktop-based application project, it aims to provide a clear and focused user environment for authentication of user’s biometric verification of fusion of serval models. Although, its end users are mainly specified as Bank Managers, Server room holders, or any company requiring a high-level security system will be able to use this application. Biometrics systems are automated systems that recognize a person based on physical or behavioural characteristics. There are several primary biometric disciplines including fingerprint, facial recognition, voice recognition, iris scan, and retina scan. A multimodal biometric system refers to a combination of the above physiological and behavioural human characteristics. Security is the degree of protection from harm. It applies to any asset, such as a person, dwelling, community, nation, or organization. In today’s age of digital technology and intelligent systems, automation, security system has become one of the fastest developing application-based technologies in the world. The proposed method deals with fingerprint, voice, and face recognition. This method increases the accuracy and reliability of the system and will present to be more secure than the previously available methods. This uses image processing and ensure that it does not authenticate any fake user.

Project Objectives

Objectives of the system are as follows:

• System will allow  regional manager to keep an eye on the data.
• System will allow manager to enter the vault. The authentication of Manager (target user of the system), will be done entirely automatically without the interaction of any other human. 

Project Implementation Method

In this proposed system, the system is instantiated by the web-based application. After it triggers then the system starts processing the image. In first phase, we start our system. We capture an image from a fingerprint scanner which is predominantly checked for certain constraints. An individual’s fingerprints are defined by a complex combination of patterns: lines, arches, loops, and whorls. An image of a fingerprint is acquired either by optical scanning, or capacitance sensing. Biometric templates are generated by matching intricate fingerprint features. After pre-processing data is passed to second phase where in 2nd phase, we capture an image from a camera which is predominantly checked for certain constraints like lightning, density, facial expressions. The captured image is resolute for our requirements. Once it is resolute, we make sure it is either in png or jpeg format else it is converted. Static or video images of a face can be used to facilitate recognition. Once the features are extracted and selected, the next step is to classify the image. Appearance-based face recognition algorithms use a wide variety of classification methods. Sometimes two or more classifiers are combined to achieve better results. After these two flags and an input file from the microphone is passed and system calls the 3rd model. In the voice recognition, the user speaks some specific words into a microphone attached to the system. Software analyses his or her voice and abstracts significant measures on roughly about twenty parameters (pitch, speech, energy density, waveforms, etc.). This live profile is compared against a profile stored on a central database or the trained model. A good match authenticates the user. An advantage of voiceprint techniques over other forms of biometric is the potential to detect coercion through the analysis of stress patterns in the sample voiceprint. Manager can start the authentication checking process where live data for verification will be asked by the system, then pre-processing will be performed. After this, features will be extracted and sent to the trained model which will predict some features and, in the end, it will provide some decision.

Benefits of the Project

Our system is mostly depending on machine learning models. We are doing image processing of different types of images, extracting different features. As a result, the reliability of our project highly depends on the reliability of the features extracted and the pipeline used for it. To ensure portability, the application will be developed in PYTHON language.

The system can be extended later with other functionalities required, such as it may consist of using IRIS recognition.

The system will be operational 24 hours a day and 7 days a week, as it will be working on specific domain and backend will be supported by highly pro-efficient libraries like Tensorflow, Keras and Open CV.

The bank vault system has security as its most important aim. Banks could go bankrupt if the vault’s security system becomes compromised. This system is centered on three biometric traits of authorized users to improve security and reduce the possibility of compromise within the bank. Once the individual has been enrolled into the system the user is then given access to the vault. Identification simply means a one-to-many matches requiring the user to provide his fingerprint, facial and voice as a means of identification. The acquired biometric sample, presented for identification, is recognized by using neural network models, if there is a match with all three traits enrolled, access is provided to the vault door, or otherwise it is declined.

Users like the bank manager shall be able to access the vault in an average time of 1-2 minutes.

Technical Details of Final Deliverable

CSS-System is basically a web-based application project, it aims to provide a clear and focused user environment for authentication of user’s biometric verification of fusion of serval models. Although, its end users are mainly specified as Bank Managers, Server room holders, or any company requiring a high-level security system will be able to use this application.

       Flow Diagram shows how the system should start and complete the progress.

1. Product Scope

   The strengths and limitations of biometric systems and their legal, social, and philosophical implications is to dispel the common misconception that a biometric system unambiguously recognizes people by sensing and analysing their biometric characteristics. No biometric technology is infallible; all are probabilistic and bring uncertainty to the association of an individual with a biometric reference, some of it related to the particular trait being scrutinized by the system. Variability in biometric traits also affects the probability of correct recognition.

   TODO(in CSS-system)	NOT TODO(in CSS-system)
   Verify sample data provided by current user.	Will not enroll new user at run time.
   Sustain the integration of models and provide results.	Single model will not allow the user to penetrate through the system.
   Will only allow the user to pass when the resultants from fusion of model provides green signal.	CSS-system will not contain the details of the application where deployed, it will only perform gate keeping role.

2. Product Functionality:

• Acquires data from user (as shown in figure).
• Feature extraction and matching algorithms.
• Process data at model end for facial, fingerprint and voice recognition.
• Store necessary data in database.

Because of the unavailability of the hardware, CSS-system will be presented as a prototype, the proposed sytem will reiqure following hardware to be completely deployed in market:

• Camera
• Microphone
• Fingerprint scanner
• Raspberry pi 

In case all hardware all managed, our system will transform in shape of desktop application.

TODO(in CSS-system)

Verify sample data provided by current user.

Sustain the integration of models and provide results.

Will only allow the user to pass when the resultants from fusion of model provides green signal.

Final Deliverable of the Project

HW/SW integrated system

Core Industry

IT

Other Industries

Core Technology

Artificial Intelligence(AI)

Other Technologies

Sustainable Development Goals

Industry, Innovation and Infrastructure, Responsible Consumption and Production, Peace and Justice Strong Institutions

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com