Recent trends and advances in hyperspectral imaging techniques to estimate solar induced fluorescence for plant phenotyping

• Published in: Ecological indicators 2022-04, Vol.137, p.108721, Article 108721
• Main Authors: Mangalraj, P., Cho, Byoung-Kwan
• Format: Article
• Language: English
• Subjects: Hyperspectral imaging; Machine learning; Plant phenotyping; Solar induced fluorescence
• ISSN: 1470-160X; 1872-7034
• DOI: 10.1016/j.ecolind.2022.108721

[Display omitted] •The recent trends in SIF estimation at multiscale and multilevel are discussed in correspondence to the end applications.•The relationship of traits and the factors influencing the impact of SIF-GPP Correlation is described.•Multiple machine learning models used to identify the correlation for developing complex applications are explained.•We have provided future directions and challenges associated with SIF estimation. Inevitable environmental changes empower the researchers to understand and analyze the plant traits for improving the ecosystem. Solar-induced fluorescence (SIF) is one of the functional traits to analyze the vegetation and assess plant phenotyping. Estimation of SIF through hyperspectral imaging technique gaining its popularity in the recent days than any other estimating techniques due to its contiguous spectrum property, which allows us to obtain more information. Another merit of hyperspectral images is that they can be used to acquire data on different scales. In our review, we have focused on three major areas as follows, a.) Hyperspectral imaging techniques in estimating SIF in different scales varying from Ground Scale to Orbital Scales. b.) Correlation of other functional traits and factors influences the SIF estimation c.) Machine learning techniques used to interpret the SIF traits for Agricultural Monitoring. Moreover, the aforementioned areas are becoming crucial in the recent trend, and we confine our review with the state-of-the-art techniques exclusively from 2010 to 2021. We comprehend the details in the review to provide insights on the breakthrough made in hyperspectral imaging for SIF estimation, allowing the reader to deepen their understanding in the areas of plant phenotyping, which would enable them to explore the field for future research.