PROJECT: Deep Learning and GUI Development for Emotional Facial Recognition
-- This the the readme file for the Emotion Detection Project -- 
-- This project was designed Austin Lynch for the purposes to UCSB's IEEE software project class -- 

** HOW TO USE **
1. Install Necessary librarys in Matlab listed below
2. Import the data excel file from following folder link:
   https://docs.google.com/spreadsheets/d/172ZCF12ufH1fADBiHSMoya5wh9Tsmn8P/edit?usp=sharing&ouid=109210790893542256917&rtpof=true&sd=true
2. Make sure to have face data excel file in current folder to build database 
3. Construct database folders to store data images (* Note Folder Names Matter *)
    a. Have sub-folder in current folder names "database"
    b. Within this sub-folder have 7 different folders names: "Suprise", "Sad",
       "Neutral", "Happy", "Fear", "Disgust", "Angry"
    c. These folders should initially be empty
3. Run the Emotions.m code file to build and run the GUI
4. When the GUI is up
    a. First Build the Database by pressing the button. This will take around 2-3 minutes
       to build if not already. 
    b. Second Train the Neural Network. A Training Plot will show up showing progress. 
       This will take 2-3 minutes as well and network will be saved to current folder when 
       done. 
    c. You can now run the GUI with emotion detection now that the database is done and 
       network trained and stored 
    d. When done Reading face, press stop button the the program will stop 

** NECESSARY MATLAB LIBRARYS FOR CODE TO RUN **
    1. Image Processing Toolbox
    2. Deep Learning Toolbox
    3. Matlab Support package for USB Webcams
    4. Computer Vision Toolbox

** FILES **
- Emotion.m - 
This file Builds the GUI and hold call back functions for when buttons pressed. 
This is the main file of the project. 

- ClassifyFrame.m - 
This is a supporting file for Emotion.m that takes trained Neural network and image
as inputs and returns the image with labeled bounding box and classification label from 
network.

- TrainCNN.m - 
This is a supporting file for Emotion.m that builds the neural network and trains it
if it is not already trained and saved or uploads the previously trained neural network 
file. 

** FUNCTIONS **
 - BuildDataBaseButtonPushed - 
This function is housed in Emotion.m and is a call back function for when the Build Database 
Button is pressed. It checks to see if the database is already built and if not is then 
read through the data base excel file and build the database of images to be used in the
neural network. 

 - TrainCNNButtonPushed - 
This function is housed in Emotion.m and is a call back function for when the TrainCNN
Button is pressed. It first check to see if the network has been previously trained and
saved and if not then it builds and Trains the network by calling on the function housed in
TrainCNN.m

 - StartButtonPushed - 
This function is housed in Emotion.m and is a call back function for when the Start
Button is pressed. It is connects to the built in webcam on the computer and snapshots
a frame from this feed then uses the trained neural network classify it and output the 
results

 - StopButtonPushed - 
This function is housed in Emotion.m and is a call back function for when the Stop
Button is pressed. It changes the stop indicator variabe to stop the looping of the frame 
capturing and classification in the start function and reset the GUI values.