oversampling multivariate time series data

For some classification needs. I have multivariate time series data composed from 4 stelite images in form of (145521 pixels, 4 dates, 2 bands) I made a classification with tempCNN to classify the data into 5 classes. However there is a big gap between the class 1,2 with 500 samples and 4,5 with 1452485 samples.

I' am wondering if there is a method that help me oversamling the two first classes to make my dataset more adequate for classification.