Enhancing Detection of Fraudulent People at Immigration Terminals using Facial Expression Conversion

A recent Australian study concluded that their immigration officers often mismatch 1 out of 7 people. As fradulents create new ways of decepting the technology to get into a country, we propose a new method to detect fraudulent people trying to enter a country using AI/ML. The system uses facial rec

Project Title

Enhancing Detection of Fraudulent People at Immigration Terminals using Facial Expression Conversion

Project Area of Specialization

Artificial Intelligence

Project Summary

A recent Australian study concluded that their immigration officers often mismatch 1 out of 7 people. As fradulents create new ways of decepting the technology to get into a country, we propose a new method to detect fraudulent people trying to enter a country using AI/ML. The system uses facial recognition to detect the mood and can convert it into any other. Immigration officers may use this system to match the photo on the passport with the person standing at the terminals to better judge whether he is the same person or not.

The system uses ML algorithms of neural networks to detect a face from a picture and is trained from thousands of images to detect expressions, which is further used to convert the picture in the desired expression. This part is extremely useful because people have neutral emotion photos on passports but when they arrive at the terminals  they are usually happy or anxious, so this system can be used to create a neutral emotion picture from a happy /sad one.

Project Objectives

The main objective of the proposed idea is to create a highly responsive and efficient system based on Machine Learning and Aritificial Intelligence that can detect expression from an image and create images of the same image but with different expressions. The system must be highly time efficient in a way that it does the job quickly. This is important because there are lines of people at immigration terminals and you don’t want your system to take 2 minutes at each person , rather the proposed system should do the job in a few seconds atmost.

The proposed system’s UX (user experience) must be good that a user interacting with it the first time should not have problems with it; and remembers the system the next time he uses it (Learnability). Along with UX , the system should have appealing UI (user interface) so that the user is attracted to it; and it appeals the user to use it. The main interface to be used would be GUI (Graphical User Interface). Furthermore it should be so easy to use that the user has the least cognitive load i.e (the system does most of the job and is easy to navigate and user doesn’t get stuck anywhere)

Project Implementation Method

Recent advances in machine learning yielded new tchniques to train deep neural networks, which resulted in highly successful applications in 
many pattern recognition tasks such as emotion and speech recognition. Machine learning can help in different fields as medical, entertainment, security. 
In this project we will identify the human face expression and can convert that image in multiple available human face expressions. As implementation 
gives the opportunity to see the plans become a reality so, here is the implementation method of our project. First of all we will design the UI screens 
for the complete web system. Then using the MongoDB compus we will design the database for the system. A user will upload the image that he/she wants to 
change, then after selecting the desired expression from available options system will detect the human face expression in image update the dataset and 
then convert the convert the image in selected expression of user. The system can be used for security purpose for example to detect the criminals with 
different images having different expressions. 

Benefits of the Project

The proposed system has countless benefits, some of them are given below:

1. The system can be used to better identify criminals by creating images of them in different expressions
2. Half part of the system can be used in LAWs( Lethal Autonomous Weapons System) to program the weapon not to attack a person/child with a crying face. The system can report back to the base whether to attack or not
3. It can be developed as an app on android or iOS for the amusement of people, they can see how they look in different emotions.
4. It can be further used to create datasets of people in different expressions which will further enhance the system itself.
5. The system can be used in helping detecting Deep Fakes
6. It can be used at airports to detect people with stolen passports(passports of other people) travelling into a country .
7. It has huge usage in film industry. Video processing can use these images generated by the system to create deepfakes where the actor himself might not be available or is too risky to let him do a thing (eg dangerous stunts that a actor should not do)
8. Further enhancement of the project can help medical doctors to detect how muscles contract and relax in different emotions. A Facial Expression Recognition and Conversion system of human muscles only will further help the scientists in medicine.

Technical Details of Final Deliverable

The final deliverable will be a Web API that can be used to detect and create images of different facial expressions from an image. The system will contain numerous webpages with css, javascript and a database to store data (MongoDB). Flask will be used as the language as an intermediary between the pages and the database. In terms of hardware the system would require a good processor, RAM (Random Access Modules) and a graphic card to run at airport systems.  SSD would be preferred over HDD for faster access of data.  An internet connection would be required that should be fast enough to run the system properly. 4G or 5G is preferred.

Final Deliverable of the Project

Software System

Core Industry

IT

Other Industries

Security

Core Technology

Artificial Intelligence(AI)

Other Technologies

Sustainable Development Goals

Good Health and Well-Being for People, Decent Work and Economic Growth

Required Resources

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