Towards Low-Cost Image-based Plant Phenotyping using Reduced-Parameter CNN
Towards Low-Cost Image-based Plant Phenotyping using Reduced-Parameter CNN
复制标题
DOI：
--
复制
发表时间：
2018-09
期刊：
影响因子：
--
通讯作者：
John Atanbori;F. Chen;A. French;T. Pridmore
中科院分区：
文献类型：
--
作者：
John Atanbori;F. Chen;A. French;T. Pridmore
分割是大多数植物表型分析应用的核心。目前最先进的植物表型分析应用依赖于深度卷积神经网络（CNN）。然而，这些网络具有许多层和参数，增加了训练和测试时间。依赖于这些深度CNN的表型应用程序通常也很难（如果不是不可能）部署在资源有限的设备上。我们提出了我们的工作，研究深度神经网络中的参数减少，这是将植物表型应用转移到现场和资源有限的低成本设备上的第一步。我们重新构建了四个基线深度神经网络（创建我们称之为“Lite CNN”），通过减少它们的参数，同时使它们更深，以避免过度拟合的问题。我们在“Lite”CNN上实现了最先进的、与基线相当的性能。我们还引入了一个简单的全局超参数（alpha），提供了一个有效的权衡参数大小和精度之间。
Segmentation is the core of most plant phenotyping applications. Current state-of-the-art plant phenotyping applications rely on deep Convolutional Neural Networks (CNNs). However, these networks have many layers and parameters, increasing training and test times. Phenotyping applications relying on these deep CNNs are also often difficult if not impossible to deploy on limited-resource devices. We present our work which investigates parameter reduction in deep neural networks, a first step to moving plant phenotyping applications in-field and on low-cost devices with limited resources. We re-architect four baseline deep neural networks (creating what we term "Lite CNNs") by reducing their parameters whilst making them deeper to avoid the problem of overfitting. We achieve state-of-the-art, comparable performance on our "Lite" CNNs versus the baselines. We also introduce a simple global hyper-parameter (alpha) that provides an efficient trade-off between parameter-size and accuracy.