Authors

Abstract

Accurate measurement of plant phenotypic data can provide a comprehensive understanding of plant physiology and help to study the relationship between plant genes and the environment. The application of visible light and other multi-source and multi-dimensional imaging sensing technology can provide a rich data source for plant phenotype identification and monitoring. With the continuous development and application of computer technology in the field of plant phenotype analysis, deep learning technology has made remarkable achievements in plant phenotype identification and monitoring. On the basis of reviewing the relevant research results at home and abroad at this stage,this paper firstly describes the common ways of plant phenotype image acquisition; then it discusses in detail the current status of the application of deep learning technology in the fields of classification, detection and segmentation of plant phenotypes, crop development and yield prediction, as well as plant drought and pest stress, etc.; and finally it discusses the challenges and future development goals of the deep learning method in the monitoring and recognition of plant phenotypes.This paper aims to provide theoretical support and technical reference for the development and application of deep learning technology in the field of agricultural plant phenotyping.

Abstract in Chinese

精确测量植物表型数据可以全面了解植物生理状况，有助于深入研究植物基因与环境之间的关系。应用可见光等多源多维度成像感知技术可为植物表型识别与监测提供丰富的数据源，随着计算机技术在植物表型分析领域的不断发展应用，深度学习技术在植物表型识别与监测方面已取得了显著成绩。在梳理现阶段国内外相关研究成果的基础上，本文首先阐述了常见的植物表型图像采集方式；之后详细探讨了深度学习技术在植物表型的分类、检测与分割，作物生长发育及产量预测，以及植物干旱与病虫害胁迫等领域的应用现状；最后讨论了深度学习方法在植株表型监测与识别中的挑战与未来发展目标。本文旨在为深度学习技术在农业植物表型领域的发展与应用提供理论支持和技术参考。