Generation of microbial colonies dataset with deep learning style transfer

TL;DR

This paper presents a novel method combining traditional computer vision and neural style transfer to generate synthetic microbiological images, enabling effective training of deep learning models with fewer real data resources.

Contribution

The authors introduce a data augmentation strategy that synthesizes realistic microbiological images using style transfer, reducing the need for extensive real datasets for training deep learning models.

Findings

Synthetic dataset enables training of accurate microbial detection models

Comparable detection performance achieved with fewer real images

Method is flexible and potentially applicable to other domains

Abstract

We introduce an effective strategy to generate an annotated synthetic dataset of microbiological images of Petri dishes that can be used to train deep learning models in a fully supervised fashion. The developed generator employs traditional computer vision algorithms together with a neural style transfer method for data augmentation. We show that the method is able to synthesize a dataset of realistic looking images that can be used to train a neural network model capable of localising, segmenting, and classifying five different microbial species. Our method requires significantly fewer resources to obtain a useful dataset than collecting and labeling a whole large set of real images with annotations. We show that starting with only 100 real images, we can generate data to train a detector that achieves comparable results (detection mAP = 0.416, and counting MAE = 4.49) to the same detector but trained on a real, several dozen times bigger dataset (mAP = 0.520, MAE = 4.31), containing over 7k images. We prove the usefulness of the method in microbe detection and segmentation, but we expect that it is general and flexible and can also be applicable in other domains of science and industry to detect various objects.