[论文解读] Extracting Pasture Phenotype and Biomass Percentages using Weakly Supervised Multi-target Deep Learning on a Small Dataset

Badri Narayanan, Mohamed Saadeldin+3 more|arXiv (Cornell University)|Jan 8, 2021

Genetic and phenotypic traits in livestock被引用 6

一句话总结

本文提出了一种使用迁移学习和数据增强的弱监督多目标深度学习方法，仅基于261张田间图像即可估算草、三叶草、白三叶草、红三叶草和杂草的生物量百分比。尽管数据集较小，该方法在草上的平均绝对误差（MAE）为6.77%，三叶草为6.92%，白三叶草为6.21%，红三叶草为6.89%，杂草为4.80%，在大多数组件上优于基线模型。

ABSTRACT

The dairy industry uses clover and grass as fodder for cows. Accurate estimation of grass and clover biomass yield enables smart decisions in optimizing fertilization and seeding density, resulting in increased productivity and positive environmental impact. Grass and clover are usually planted together, since clover is a nitrogen-fixing plant that brings nutrients to the soil. Adjusting the right percentages of clover and grass in a field reduces the need for external fertilization. Existing approaches for estimating the grass-clover composition of a field are expensive and time consuming - random samples of the pasture are clipped and then the components are physically separated to weigh and calculate percentages of dry grass, clover and weeds in each sample. There is growing interest in developing novel deep learning based approaches to non-destructively extract pasture phenotype indicators and biomass yield predictions of different plant species from agricultural imagery collected from the field. Providing these indicators and predictions from images alone remains a significant challenge. Heavy occlusions in the dense mixture of grass, clover and weeds make it difficult to estimate each component accurately. Moreover, although supervised deep learning models perform well with large datasets, it is tedious to acquire large and diverse collections of field images with precise ground truth for different biomass yields. In this paper, we demonstrate that applying data augmentation and transfer learning is effective in predicting multi-target biomass percentages of different plant species, even with a small training dataset. The scheme proposed in this paper used a training set of only 261 images and provided predictions of biomass percentages of grass, clover, white clover, red clover, and weeds with mean absolute error of 6.77%, 6.92%, 6.21%, 6.89%, and 4.80% respectively.

研究动机与目标

• 开发一种非破坏性、高效的混合牧场中草、三叶草和杂草生物量百分比估算方法。
• 通过利用迁移学习和数据增强，解决农业图像分析中标签数据有限的挑战。
• 通过弱监督和标签填补，提升在小样本数据集上存在缺失标签时的多目标回归性能。
• 利用深度学习实现对视觉上相似的三叶草种类（白三叶草与红三叶草）的准确区分。
• 为未来在智能农业中开展半监督和弱监督学习奠定基础。

提出的方法

• 微调在ImageNet上预训练的VGG-16模型，用于生物量百分比的多目标回归。
• 应用数据增强和模型正则化，以缓解261张图像的小型训练集上的过拟合问题。
• 采用标签填补技术——基线法、回归法、均值法和中位数法，处理训练样本中白三叶草和红三叶草的缺失值。
• 通过差异化的样本权重实现弱监督，训练过程中优先考虑更可靠的标签。
• 使用平均绝对误差（MAE）作为多目标预测的损失函数进行模型训练。
• 在公开的Grass Clover Image数据集（2019年生物量预测挑战赛）上评估性能。

实验结果

研究问题

• RQ1 仅使用261张标注的田间图像，深度学习模型能否实现准确的多目标生物量百分比预测？
• RQ2 迁移学习结合数据增强和正则化，在小样本农业数据集上是否能有效克服过拟合？
• RQ3 采用差异化样本权重和标签填补的弱监督方法，是否能提升在不完整或噪声标签数据上的性能？
• RQ4 该模型能否在混合牧场中有效区分视觉上相似的三叶草种类（白三叶草与红三叶草）？
• RQ5 与Biomass Prediction Challenge中的基线模型相比，该方法在关键生物量组分上的表现如何？

主要发现

• 所提模型在草上的平均绝对误差（MAE）为6.77%，三叶草为6.92%，白三叶草为6.21%，在这些组分上优于基线模型。
• 模型在红三叶草上的MAE为6.89%，杂草为4.80%，红三叶草略高于基线，但杂草表现与基线相当。
• 标签的均值填补技术在整体性能上表现最佳，显著降低了大多数生物量组分的误差。
• 通过有效的标签填补和弱监督，模型在标签噪声和缺失数据上表现出强鲁棒性。
• 结果表明，迁移学习和数据增强即使在标签数据极少的情况下也能实现优异性能，验证了该方法在真实农业应用中的可行性。
• 本研究为未来在农业非破坏性表型估计中开展半监督和弱监督学习奠定了坚实基础。