Building the face emotion recognition model
In this section, we will implement the application of FER using CNN. For coding purposes, we will be using the TensorFlow, TFLearn, OpenCV, and Numpy libraries. You can find the code by using this GitHub link: https://github.com/jalajthanaki/Facial_emotion_recognition_using_TensorFlow. These are the steps that we need to follow:
Preparing the data
Loading the data
Training the model
Preparing the data
In this section, we will be preparing the dataset that can be used in our application. As you know, our dataset is in grayscale. We have two options. One is that we need to use only black and white images, and if we are using black and white images, then there will be two channels. The second option is that we can convert the grayscale pixel values into RGB (red, green, and blue) images and build the CNN with three channels. For our development purposes, we are using two channels as our images are in grayscale.
First of all, we are loading the dataset and...
