Development of non-destructive methods to estimate functional traits and field evaluation in tea plantations using a smartphone

Abstract

Smartphones are equipped with various types of sensors which make them a promising tool to assist diverse digital farming tasks because of their mobility, cost, accessibility, and computing power allow us to perform real-time practical applications. This paper presents the utilization of various non-destructive methods of nutrient and disease classification techniques using smartphone collected images, processed through various image segmentation algorithms. Both in vivo and in vitro estimations shows comparable results with both chlorophyll and nitrogen contents of a crop shoot. Moreover, the correlation between SPAD measured values and nitrogen of crop shoot showed a significant linear association (R 2 =0.7309), revealing the potency of in vivo observation for prediction of actual chlorophyll content in tea crop. SPAD values and yield have a strong linear relationship (R 2 =0.7103), in which SPAD-meter performed better detection at very low values. The study concluded that the proposed techniques could be used for automatic detection as well as classification of foliar diseases and nutrients in tea.

Plant phenotyping relevance

スマートフォン画像と画像セグメンテーションにより、茶葉の栄養・病害およびクロロフィル／窒素などの機能形質を非破壊推定する手法を開発・評価しており、形質取得法が研究の中心である。

Code and data availability

The supplied blocks contain no data availability, code deposit, or supplementary asset statements. The paper describes smartphone image collection, ImageJ-based RGB extraction, neural network/genetic algorithm segmentation, and ArcGIS kriging, but provides no public URL, repository, or identifier for the images, trait/

No evidence-backed public reproduction asset is currently recorded.