Plant Phenotyping in Plant Factories: Progress, Challenges and Prospects

ABSTRACT

Plant factories with artificial lighting (PFALs) are enclosed vertical cultivation systems equipped with customisable LED spectral regulation, whose stacked spatial structure and controlled light environments substantially differ from open fields and conventional greenhouses, bringing unique and specialised challenges to precise plant phenotyping. Driven by high‐throughput sensing technologies and artificial intelligence (AI), plant phenotyping in PFALs has achieved considerable advances. This review comprehensively summarises progress in PFALs plant phenotyping, covering sensing techniques including RGB imaging, multispectral imaging, hyperspectral imaging, and 3D point cloud reconstruction, as well as advanced analytical models integrating machine learning, deep learning, large language models, and AI agents for phenotypic data mining. We further highlight the important bottlenecks restricting the industrial application of PFALs phenotyping: commercial phenotyping devices are generally bulky, costly, and incompatible with compact vertical cultivation layouts, while the shortage of PFALs‐specific phenotypic datasets and scenario‐adaptive optimisation models hinders large‐scale industrial translation. To clarify the differentiated technical characteristics of PFALs phenotyping, we systematically compare the discrepancies in phenotyping strategies, technical demands, and application scenarios between PFALs and traditional greenhouses. On this basis, we propose targeted solutions and prospective development directions for PFALs phenotyping, focusing on miniaturised and integrated phenotyping equipment, optimised AI‐driven analytical frameworks, and multi‐omics intelligent breeding. Distinct from previous general phenotyping reviews, this paper specifically targets the vertical, controlled PFALs cultivation scenario, systematically sorting out its exclusive technical challenges and corresponding coping strategies, and providing actionable insights for the future intelligent upgrading and industrialisation of PFALs / vertical farm plant phenotyping technology.