AI Based Human Emotion Recognition System

Project Title

AI Based Human Emotion Recognition System

Project Area of Specialization Artificial IntelligenceProject Summary

“AI BASED HUMAN EMOTION RECOGNITION SYSTEM” is a Machine Learning Computer Vision system based on Faster RCNN Model. In this system we are going to detect the human emotions with Machine Learning. It will provide the user emotion detecting. Now a days the Companies values the customer’s feedback. Most of the companies failed to judge the customers satisfaction about their products due to the lack of data and written feedbacks. Our approach will these types of companies to know what their customers wants. In our project we are going to use facial features and build a neural network based on the data we collect and then derives our results.

The process is pretty streamline, we first collect data with the expertise of the people in reading the human emotions. Then we are going to divide our data into test and train and finally we get our desired output. The idea behind all this come from the current situation that the world facing and the innovation and us taking place in the world of computer vision.

In a facial recognition system -- which maps an individual's facial features mathematically and stores the data as a faceprint -- face detection data is required for the algorithms that discern which parts of an image or video are needed to generate a faceprint.  Once identified, the new face print can be compared with stored face prints to determine if there is a match [1]. Similarly, we going to develop our system with the custom algorithms and the data sets and by storing the correct fractures and amp them right we can detect the human emotions with the local demographics of people.

There are serval other techniques for the object detection in the machine learning and deep learning but we are going to use the faster RCNN. Because it has many advantages over the other techniques like YOLO. YOLO is little faster and easy to use but for small object detection YOLO struggle and the Faster RCNN is better.

Project Objectives

Following are the objectives of this project:

1. More accurate model for local emotion recognition
2. RCCN approach for emotion recognition
3. Real time computer vision application

Data visualization for the detected emotions

Project Implementation Method

Components, Libraries, Web Services and stubs
•    pandas
•    matplotlib
•    tensorflow
•    keras – 2.0.3
•    numpy
•    opencv-python
•    sklearn
•    h5py
Webservices such as:
•    HTTPS
•    FTP
And Flask API
Tools and Techniques
Tools 
•    Pycharm
•    VScode
•    Google Collab
•    Google drive
Technologies 
•    Python
•    Bootstrap
•    Html
•    PHP
•    Java script 
•    CSS
 

Benefits of the Project

Safety measure 
 Emotion recognition is already used in schools and other institutions because it helps prevent violence and improve the overall safety of the location. 

 Staff Assistance 
 Some companies are using AI with emotion recognition API  as an HR assistant. By assessing voice tones, facial expressions, and keywords, the system helps determine if a candidate is honest and truly interested in the position, and provides reports that recruiters can review. 
 
 Customer service 
 Currently, the system installed in the customer service center is on sale. With the help of a camera equipped with artificial intelligence, you can compare the  emotions of customers before and after entering the center to determine how satisfied they are with the service they receive. Also, if your score is low, you can advise your employees on improving quality of service.
 
 Children with various disabilities 
 There is a project aimed at helping children with autism interpret the feelings of those around them using the Google Glass Smart Glasses system. When a child interacts with another person, vision and sound are used to provide clues about the emotions of the other person. 

 Target Group Commitment 
 Companies  also use emotion recognition to determine  business outcomes based on the emotional response of their viewers. Apple has also released a new feature on the iPhone called Animoji, whose emojis are designed to mimic human facial expressions. 

Test video game 
 Video games are tested for user feedback  to determine if a company has achieved its goals. Emotion detection can be used during these testing stages to understand the emotions users are experiencing in real time and incorporate that feedback into the final product manufacturing.
 
 Healthcare 
 The healthcare industry  is certainly taking advantage of facial emotion recognition these days. They use it to find out if a patient needs a medicine or  to let a doctor know who to meet first.
 

Technical Details of Final Deliverable

The suggested Deployment environment is: -
1.    Workstation/Server: A powerful workstation or server is required with a power 4 cores+ CPU and an NVidia Tesla series GPU.
2.    Cloud Computing (Optional): If user don’t want to buy a powerful computer, he can get an AWS or google AI cloud subscription for the task of processing.
3.    OS: For processing environment Linux or windows is recommended (As they are tested) for the web app (Android, IOS, Mac, windows or Linux) with internet connection can be used. 
4.    Camera: For This a HD camera is must with minimum good throughput is must.
 

Final Deliverable of the Project HW/SW integrated systemCore Industry ITOther Industries Security Core Technology Artificial Intelligence(AI)Other Technologies Big DataSustainable Development Goals Sustainable Cities and CommunitiesRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	55000
FYP Document	Miscellaneous	1	6000	6000
FYP Poster	Miscellaneous	1	2000	2000
GPU	Equipment	1	25000	25000
HD Camera	Equipment	1	7000	7000
Web Hosting	Equipment	1	15000	15000