Identification Services with Machine Learning

Challenge Background

Presently in Sudan, there is no standardized Civilian Identification Record despite the rising crime rates this nation experiences, this reality adversely hinders the process of Identifying Personnel and impairs both the pursuit of Justice and hinders the process of Personnel Authentication and Verification. We firmly believe that introducing this technology would promote introducing and adopting a standardized personnel Identification system and greatly improve national security standards. Such a system would help keep track of civilian identities and provide easy and ubiquitous access to this system. The same framework could also be replicated or implemented on different scales for private Identification services and for creating complex ID-based security systems.

The Problem

Sudan Experiences a lot of crime and despite the rising crime rate, it still lacks a standardized Digital Civilian Record System. Completely reliant on paper-based authentication, the current system is both resource exhaustive and time-consuming. Taking days in pursuit of paper trails, where it would take seconds digitally. The process of Record Retrieval, Personnel verification, background checks, and clearing are all done manually and based on paper trail-based approaches. In this solution we propose the use of a cloud-based digital identification system that uses a relational database system and relies on Machine Learning to Identify, Assess, Examine, and manage security operations, allowing for the creation and deployment of an easy-to-use system for Identification and RecordKeeping.

Goal of the Project

• Develop a recognition model that can accurately identify faces.
• Test the accuracy of our model and attempt to improve it.
• Integrate our final model into a suitable database in an application.
• Deploy an API or demo of the proposed system.
• Test the final product and measure its effectiveness.

Project Timeline

What you'll learn

- Data collection - Data Processing - Labelling of Data - ML Model for extraction of Face - ML Model for identification and comparison of Faces against Known Databases - A Database system for storing the Face-Recognition Database Content - Testing of Results and Fine Tuning the model - Deployment of the whole system