Inside Out: Transforming Images of Lab-Grown Plants for Machine Learning Applications in Agriculture

TL;DR

This paper presents a method using a contrastive unpaired translation GAN to generate realistic field images of plants from indoor images, enhancing training datasets for improved plant detection in agriculture.

Contribution

It introduces a novel application of CUT-GAN for translating indoor plant images into field-like images, aiding data augmentation for machine learning in agriculture.

Findings

Synthetic images improve plant detection accuracy

Method extends to multi-plant field image generation

Enhanced models outperform those trained only on real data

Abstract

Machine learning tasks often require a significant amount of training data for the resultant network to perform suitably for a given problem in any domain. In agriculture, dataset sizes are further limited by phenotypical differences between two plants of the same genotype, often as a result of differing growing conditions. Synthetically-augmented datasets have shown promise in improving existing models when real data is not available. In this paper, we employ a contrastive unpaired translation (CUT) generative adversarial network (GAN) and simple image processing techniques to translate indoor plant images to appear as field images. While we train our network to translate an image containing only a single plant, we show that our method is easily extendable to produce multiple-plant field images. Furthermore, we use our synthetic multi-plant images to train several YoloV5 nano object detection models to perform the task of plant detection and measure the accuracy of the model on real field data images. Including training data generated by the CUT-GAN leads to better plant detection performance compared to a network trained solely on real data.