Precise semantic segmentation of crops and weeds is necessary for agricultural weeding robots. However, training deep learning models requires large annotated datasets, which are costly to obtain in real fields. Synthetic data can reduce this burden, but the gap between simulated and real images remains a challenge. In this paper, we present a pipeline for procedural generation of synthetic crop-weed images using Blender, producing annotated datasets under diverse conditions of plant growth, weed density, lighting, and camera angle. We benchmark several state-of-the-art segmentation models on synthetic and real datasets and analyze their cross-domain generalization. Our results show that training on synthetic images leads to a sim-to-real gap of 10%, surpassing previous state-of-the-art methods. Moreover, synthetic data demonstrates good generalization properties, outperforming real datasets in cross-domain scenarios. These findings highlight the potential of synthetic agricultural datasets and support hybrid strategies for more efficient model training.
翻译:农业除草机器人需要精确的作物与杂草语义分割。然而,训练深度学习模型需要大量标注数据集,在真实农田中获取成本高昂。合成数据可减轻此负担,但仿真图像与真实图像之间的差异仍是挑战。本文提出一种基于Blender的程序化合成作物-杂草图像生成流程,可在不同植物生长阶段、杂草密度、光照条件和相机角度下生成带标注的数据集。我们在合成与真实数据集上对多种先进分割模型进行基准测试,并分析其跨领域泛化性能。实验结果表明,使用合成图像训练会产生10%的仿真-现实差距,优于先前最优方法。此外,合成数据展现出良好的泛化特性,在跨领域场景中表现优于真实数据集。这些发现凸显了合成农业数据集的潜力,并为采用混合策略实现更高效的模型训练提供了依据。