Recent methods for evaluating crop qater stress using AI techniques: A review

Authors: CHO, SOO BEEN - Chungnam National University; SOLEH, HIDAYAT - Gyeongsang National University; CHOI, JI WON - Gyeongsang National University; HWANG, WOON-HA - National Institute Of Agricultural Sciences (RDA); LEE, HOONSOO - Chungbuk National University; CHO, BYOUNG-KWAN - Chungbuk National University; Kim, Moon; Baek, Insuck; KIM, GEONWOO - Gyeongsang National University

Submitted to: Sensors
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 9/24/2024
Publication Date: 9/29/2024
Citation: Cho, S., Soleh, H.M., Choi, J., Hwang, W., Lee, H., Cho, B., Kim, M.S., Baek, I., Kim, G. 2024. Recent methods for evaluating crop qater stress using AI techniques: A review. Sensors. 24. Article e6316. https://doi.org/10.3390/s24196313.
DOI: https://doi.org/10.3390/s24196313

Interpretive Summary: This overview presents the latest artificial intelligence (AI) techniques, including machine learning and deep learning, for monitoring crop water stress using remote sensing. It first addresses various remote sensing methods, including color, thermal, hyperspectral imaging, and the crop water stress index. Next, it explores multiple approaches for analyzing large volumes of remote sensing data using artificial intelligence methods (such as machine learning, deep learning, and ensemble methods) to evaluate crop water stress under abnormal climatic conditions. This review provides valuable information for agricultural researchers and industrial users of remote sensing technologies, aiding in the effective monitoring of crop water stress. By leveraging timely in-field data, it supports better decision-making and actions that can enhance productivity and mitigate the adverse effects of climate change on crop production.

Technical Abstract: In recent years, the average global temperature has increased and led to abnormal climatic conditions that worsen the growth conditions for crops. It might be negatively impacting crop yield and quality, and posing a threat to global agricultural production. These changes induce water stress in crops, which is significantly affecting both yield and quality. To address this issue, non-destructive remote sensing techniques for crop water monitoring are essential. Hence, various remote sensing methods and artificial intelligence technologies are being utilized for the quantification of crop water stress. Due to the fact that AI technologies are efficient in analyzing large volumes of remote sensing data and effective in predicting specific target values with limited data under water stress conditions. In this study, we analyzed remote sensing technologies (RGB, thermal imaging, hyperspectral, CWSI) and artificial intelligence methods (machine learning, deep learning, ensemble, GAN, XAI) for evaluating crop water stress in response to abnormal climatic conditions.