Identification of nutrient deficiency in bean plants by prompt chlorophyll fluorescence measurements and Artificial Neural Networks

Abstract

The deficiency of macro (N, P, S, Ca, Mg and K) and micro (Zn, Cu, B, Mo, Cl, Mn and Fe) minerals has a major effect on plant development. The lack of some nutrient minerals especially of nitrogen, potassium, calcium, phosphorus and iron is a huge problem for agriculture and early warning and prevention of the problem will be very useful for agro-industry. Methods currently used to determine nutritional deficiency in plants are soil analysis, plant tissue analysis, or combined methods. But these methods are slow and expensive. In this study, a new method for determining nutrient deficiency in plants based on the prompt fluorescence of chlorophyll a is proposed. In this paper bean plants are grown on a complete nutrient solution (control) were compared with those grown in a medium, which lacked one of these elements - N, P, K, Ca and Fe. In this article the mineral deficiency in nutrient solution was evaluated by the stress response of the plants estimated by leaves photosynthetic activity. The photosynthetic activity was estimated by analysis of the chlorophyll fluorescence using JIP-test approach that reflects functional activity of Photosystems I and II and of electron transfer chain between them, as well as the physiological state of the photosynthetic apparatus as whole. Next the fluorescence transient recorded from plants grown in nutrient solution with deficiency of N, P, K, Ca and Iron, as an input data in Artificial Neural Network was used. This ANN was train to recognise deficiency of N, P, K, Ca and Iron in bean plants. The results obtained were of high recognition accuracy. The ANN of fluorescence transient was presented as a possible approach to identify/predict the nutrient deficiency using the fast chlorophyll fluorescence records.

Plant phenotyping relevance

クロロフィル蛍光測定とANNを組み合わせ、植物の栄養欠乏状態を推定・識別する方法を提案しており、表現型取得・判定手法が研究の中心である。

Code and data availability

The paper describes chlorophyll fluorescence measurements and a Python ANN for nutrient deficiency recognition in bean plants, but contains no data availability statement, no public dataset deposit, and no code repository or URL. The ANN code is mentioned ('The code for our ANN is written on Python') without any public

No evidence-backed public reproduction asset is currently recorded.