ARTIFICIAL INTELLIGENCE AND DEEP LEARNING IN FLORICULTURE AND LANDSCAPING: ADVANCES, APPLICATIONS, CHALLENGES AND FUTURE PERSPECTIVES

Abstract

Floriculture and landscaping are closely related branches of ornamental horticulture and share many underlying technologies but are seldom reviewed together. This review synthesises recent literature on the application of artificial intelligence (AI) and deep learning in the floriculture value chain, landscape planning and design. Convolutional neural networks, object-detection frameworks (YOLO, Faster R-CNN) and vision transformer and self-supervised models have been used in floriculture for cultivar identification, high-throughput phenotyping, flower and bud detection, growth and flowering prediction, disease and pest detection, quality assessment, robotic harvesting, grading and sorting and postharvest management including vase-life prediction. In landscaping, AI is assisting in generative design and visualisation, sensor-based plant selection and machine learning-GIS approaches to green-space planning. Recognition, grading and disease scoring are now image-based tasks often more than 90% accurate and faster than manual assessment. But even in the best greenhouse conditions, robotic harvesting still only succeeds in about 78 to 80 percent of cases and landscape tools are used mainly as decision making aids rather than as autonomous planners. Some of the limitations that hinder widespread use include lack of annotated ornamental datasets and high cultivar diversity, poor generalisation between laboratory and field conditions, cost limitations for small and medium growers, limited model interpretability, lack of standardised benchmarks and challenges in handling delicate plant material. Future directions include multimodal and generative decision-support systems that combine computer vision and large language models, explainable and lightweight edge models, IoT-linked digital twins, soft robotic manipulators and shared data ecosystems integrated with plant breeding. This review provides a consolidated roadmap for researchers, growers and policy makers toward intelligent, sustainable ornamental horticulture.