Scholarlink

<p>This study examines the hypothesis that leaf color properties can serve as robust indicators for estimating potassium&#xa0;(K) concentrations in apple leaves. Potassium is an essential macronutrient that plays a&#xa0;critical role in water transport, photosynthesis, and disease resistance, and its precise monitoring is fundamental for optimizing crop yield in precision agriculture. To this end, a&#xa0;nondestructive potassium prediction model was developed using second-order polynomial regression applied to color features extracted from RGB (red–green–blue), HSV (hue–saturation–value), and LAB (Lab, lightness a–b) color spaces. Leaf images from 20 ‘Royal Gala’ apple trees were captured under standardized indoor lighting conditions (300 lux) with an iPhone&#xa0;6s, and color data were correlated with potassium concentrations determined through standard laboratory analyses. Model performance was evaluated using the coefficients of determination (<i>R</i><sup>2</sup>), root mean square error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE). The RGB-based model achieved the highest predictive accuracy (<i>R</i><sup>2</sup> = 0.9996, RMSE = 66.04), outperforming models based on HSV and Lab* features. These findings indicate that RGB-based image processing offers a&#xa0;rapid, low-cost, and nondestructive approach for potassium monitoring, with strong potential for integration into precision agriculture systems. Furthermore, the study outlines methodological considerations for adapting this modeling approach to other nutrients and plant species.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A Nondestructive Leaf Nutrient Content Prediction Model and Leaf Color Properties in Apples

  • 期刊论文

  • 发表日期：

    2025年9月18日

查看全文
  • Hamit Armağan

摘要

This study examines the hypothesis that leaf color properties can serve as robust indicators for estimating potassium (K) concentrations in apple leaves. Potassium is an essential macronutrient that plays a critical role in water transport, photosynthesis, and disease resistance, and its precise monitoring is fundamental for optimizing crop yield in precision agriculture. To this end, a nondestructive potassium prediction model was developed using second-order polynomial regression applied to color features extracted from RGB (red–green–blue), HSV (hue–saturation–value), and LAB (Lab, lightness a–b) color spaces. Leaf images from 20 ‘Royal Gala’ apple trees were captured under standardized indoor lighting conditions (300 lux) with an iPhone 6s, and color data were correlated with potassium concentrations determined through standard laboratory analyses. Model performance was evaluated using the coefficients of determination (R2), root mean square error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE). The RGB-based model achieved the highest predictive accuracy (R2 = 0.9996, RMSE = 66.04), outperforming models based on HSV and Lab* features. These findings indicate that RGB-based image processing offers a rapid, low-cost, and nondestructive approach for potassium monitoring, with strong potential for integration into precision agriculture systems. Furthermore, the study outlines methodological considerations for adapting this modeling approach to other nutrients and plant species.