Data augmentation techniques for small image datasets?

Question

Currently, I am training small logo datasets similar to Flickrlogos-32 with deep CNNs. For training larger networks I need more datasets, thus using augmentation. The best I'm doing right now is using affine transformations(featurewise normalization, featurewise center, rotation, width height shift, horizontal vertical flip). But for bigger networks, I need more augmentation. I tried searching on kaggle's national data science bowl's forum but couldn't get much help. There's code for some methods given here but I'm not sure what could be useful. What are some other(or better) image data augmentation techniques that could be applied to this type of(or in any general image) dataset other than affine transformations?

Answer 1

A good recap can be found here, section 1 on Data Augmentation: so namely flips, random crops and color jittering and also lighting noise.

It covers the same kind of techniques:

• rotation
• translation
• zoom
• flips
• color perturbation

 For more details on this, study the Machine Learning Tutorial. Going through Datasets in Machine Learning will also be beneficial.

Hope this answer helps you!