Review: Application of Artificial Intelligence in Phenomics.

Area of Science:

• Agricultural Science
• Computer Science
• Plant Biology

Background:

• Plant phenomics is rapidly advancing due to innovations in high-throughput phenotyping technologies.
• Artificial intelligence (AI), particularly computer vision, machine learning, and deep learning, is increasingly integrated into scientific research.
• Non-invasive imaging techniques benefit from AI for improved data collection and analysis efficiency.

Purpose of the Study:

• To review state-of-the-art papers on AI-applied plant phenotyping published between 2010 and 2020.
• To provide an overview of current phenotyping technologies and AI integration.
• To discuss limitations and future directions in AI-driven plant phenotyping.

Main Methods:

• Literature review of over 100 papers on AI and plant phenotyping (2010-2020).
• Analysis of AI integration in non-invasive imaging and field phenotyping tools.
• Examination of AI's role in data collection, management, and analysis.

Main Results:

• AI, including machine learning and deep learning, significantly improves the efficiency of plant trait analysis via image processing.
• AI has driven the development of software and open-source tools for field phenotyping, facilitating data sharing and community research.
• The integration of AI enables the handling of large datasets crucial for accurate phenotype studies.

Conclusions:

• AI is a transformative force in plant phenotyping, enhancing data analysis and collection capabilities.
• Open-source tools and community-driven research are accelerating progress in AI-applied plant phenotyping.
• Addressing current limitations and exploring future directions is essential for continued advancement in the field.