Facial expression recognition based on MATLAB k-svd dictionary learning
Facial expression recognition is an important problem in computer vision, and it has a wide range of applications. This article will introduce the method of using MATLAB k-svd dictionary learning to realize facial expression recognition.
- Dataset preparation
This article uses the CK+ dataset, which contains facial expression videos taken by multiple people at different times, with a total of 593 images. For convenience, we crop out the facial regions of the characters and resize them to the same size. Finally, we get a size of 120 × 100 120\times100120×A dataset of 100 grayscale images.