Generative AI-based Pipeline Architecture for Increasing Training Efficiency in Intelligent Weed Control Systems

TL;DR

This paper introduces a synthetic image generation pipeline using generative AI to augment training datasets for weed detection, improving model performance and data efficiency in agricultural applications.

Contribution

The study presents a novel GenAI-based pipeline combining SAM and Stable Diffusion for creating diverse synthetic images to enhance deep learning models in weed control.

Findings

Synthetic data improves model accuracy with less real data

Models trained on mixed datasets outperform those trained only on real data

Approach reduces reliance on costly real-world datasets

Abstract

In automated crop protection tasks such as weed control, disease diagnosis, and pest monitoring, deep learning has demonstrated significant potential. However, these advanced models rely heavily on high-quality, diverse datasets, often limited and costly in agricultural settings. Traditional data augmentation can increase dataset volume but usually lacks the real-world variability needed for robust training. This study presents a new approach for generating synthetic images to improve deep learning-based object detection models for intelligent weed control. Our GenAI-based image generation pipeline integrates the Segment Anything Model (SAM) for zero-shot domain adaptation with a text-to-image Stable Diffusion Model, enabling the creation of synthetic images that capture diverse real-world conditions. We evaluate these synthetic datasets using lightweight YOLO models, measuring data efficiency with mAP50 and mAP50-95 scores across varying proportions of real and synthetic data. Notably, YOLO models trained on datasets with 10% synthetic and 90% real images generally demonstrate superior mAP50 and mAP50-95 scores compared to those trained solely on real images. This approach not only reduces dependence on extensive real-world datasets but also enhances predictive performance. The integration of this approach opens opportunities for achieving continual self-improvement of perception modules in intelligent technical systems.