Ruiz-Munoz Jose F, Nimmagadda Jyothier K, Dowd Tyler G, Baciak James E, Zare Alina
Department of Electrical and Computer Engineering University of Florida Gainesville Florida USA.
Department of Material Sciences and Engineering University of Florida Gainesville Florida USA.
Appl Plant Sci. 2020 Jul 30;8(7):e11374. doi: 10.1002/aps3.11374. eCollection 2020 Jul.
High-resolution cameras are very helpful for plant phenotyping as their images enable tasks such as target vs. background discrimination and the measurement and analysis of fine above-ground plant attributes. However, the acquisition of high-resolution images of plant roots is more challenging than above-ground data collection. An effective super-resolution (SR) algorithm is therefore needed for overcoming the resolution limitations of sensors, reducing storage space requirements, and boosting the performance of subsequent analyses.
We propose an SR framework for enhancing images of plant roots using convolutional neural networks. We compare three alternatives for training the SR model: (i) training with non-plant-root images, (ii) training with plant-root images, and (iii) pretraining the model with non-plant-root images and fine-tuning with plant-root images. The architectures of the SR models were based on two state-of-the-art deep learning approaches: a fast SR convolutional neural network and an SR generative adversarial network.
In our experiments, we observed that the SR models improved the quality of low-resolution images of plant roots in an unseen data set in terms of the signal-to-noise ratio. We used a collection of publicly available data sets to demonstrate that the SR models outperform the basic bicubic interpolation, even when trained with non-root data sets.
The incorporation of a deep learning-based SR model in the imaging process enhances the quality of low-resolution images of plant roots. We demonstrate that SR preprocessing boosts the performance of a machine learning system trained to separate plant roots from their background. Our segmentation experiments also show that high performance on this task can be achieved independently of the signal-to-noise ratio. We therefore conclude that the quality of the image enhancement depends on the desired application.
高分辨率相机对植物表型分析非常有帮助，因为其图像能够实现诸如目标与背景区分以及地上部植物精细属性的测量和分析等任务。然而，获取植物根系的高分辨率图像比收集地上部数据更具挑战性。因此，需要一种有效的超分辨率（SR）算法来克服传感器的分辨率限制、减少存储空间需求并提升后续分析的性能。
我们提出了一种使用卷积神经网络增强植物根系图像的SR框架。我们比较了训练SR模型的三种方法：（i）使用非植物根系图像进行训练，（ii）使用植物根系图像进行训练，以及（iii）先用非植物根系图像对模型进行预训练，再用植物根系图像进行微调。SR模型的架构基于两种最先进的深度学习方法：快速SR卷积神经网络和SR生成对抗网络。
在我们的实验中，我们观察到SR模型在未见过的数据集中提高了植物根系低分辨率图像信噪比方面的质量。我们使用了一组公开可用的数据集来证明，即使使用非根系数据集进行训练，SR模型也优于基本的双立方插值法。
在成像过程中纳入基于深度学习的SR模型可提高植物根系低分辨率图像的质量。我们证明了SR预处理提升了经过训练以将植物根系与其背景分离的机器学习系统的性能。我们的分割实验还表明，在该任务上的高性能可以独立于信噪比实现。因此，我们得出结论，图像增强的质量取决于期望的应用。
