In the face emotion recognition, the emotion of the human faces is classified. In this the raspberry pi3 b+, USB camera, speaker to hear the background music. To detect the face the Harrcascade frontal face.xml model is used. To detect the face emotion the mini_XCEPTION.102-0.66. hdf5 model is used. The Keras is used to convert the images in an array format. Then the prediction is done to face emotion. Based on the emotion prediction the music will play in the background.
In the existing system, classification is done through simple image processing to classify images only.
In this proposed system, Deep learning is used with the help of Keras, contains several Models. Among those models, the facial emotion of the human is classified.
If the application needs USB Camera, it can be easily interfaced with the Raspberry Pi USB Port and the speaker. The speaker is used to play the background music for the face emotion detect by a USB camera.
This project is capable of performing the real-time face emotion recognition, using the pre-trained model to detect the face and the face emotion. Start the program to run the USB camera will ON then keep the face in the frame of the camera. From the video the frame is taken one by one as an image, then the images are sent to emotion prediction to detect the emotion of the human face. Based on the emotion detection the music will play in the background.
- Raspberry Pi
- USB Camera
- Power supply
- Raspbian OS with libraries installed
- SD Card Formatter
- Etcher/Win32 Disk imager
In the video demo, you can able to see the output obtained which classifies the face emotion with a good level of accuracy.