Crop health assessment through hierarchical fuzzy rule-based status maps DOI Creative Commons
Danilo Cavaliere, Sabrina Senatore, Vincenzo Loia

et al.

Knowledge and Information Systems, Journal Year: 2024, Volume and Issue: 66(11), P. 7109 - 7136

Published: Aug. 5, 2024

Abstract Precision agriculture is evolving toward a contemporary approach that involves multiple sensing techniques to monitor and enhance crop quality while minimizing losses waste of no longer considered inexhaustible resources, such as soil water supplies. To understand status, it necessary integrate data from heterogeneous sensors employ advanced devices can assess status. This study presents smart monitoring in agriculture, involving be both stationary (such moisture sensors) mobile sensor-equipped unmanned aerial vehicles). These collect information visual maps production conditions, comprehensively the area spot any potential vegetation problems. A modular fuzzy control scheme has been designed interpret spectral indices vegetative parameters and, by applying rules, return status about The rules are applied incrementally per hierarchical design correlate lower-level (e.g., temperature, indices) with higher-level vapor pressure deficit) robustly determine main have led it. case was conducted, collection satellite images artichoke crops Salerno, Italy, demonstrate incremental integration health monitoring. Subsequently, tests were conducted on vineyard regions interest Teano, efficacy framework assessment plant stress. Indeed, comparing outcomes our those cutting-edge machine learning (ML) semantic segmentation indeed revealed promising level accuracy. Specifically, classification performance compared output conventional ML methods, demonstrating consistent achieves an accuracy over 90% throughout various seasons year.

Language: Английский

Crop health assessment through hierarchical fuzzy rule-based status maps DOI Creative Commons
Danilo Cavaliere, Sabrina Senatore, Vincenzo Loia

et al.

Knowledge and Information Systems, Journal Year: 2024, Volume and Issue: 66(11), P. 7109 - 7136

Published: Aug. 5, 2024

Abstract Precision agriculture is evolving toward a contemporary approach that involves multiple sensing techniques to monitor and enhance crop quality while minimizing losses waste of no longer considered inexhaustible resources, such as soil water supplies. To understand status, it necessary integrate data from heterogeneous sensors employ advanced devices can assess status. This study presents smart monitoring in agriculture, involving be both stationary (such moisture sensors) mobile sensor-equipped unmanned aerial vehicles). These collect information visual maps production conditions, comprehensively the area spot any potential vegetation problems. A modular fuzzy control scheme has been designed interpret spectral indices vegetative parameters and, by applying rules, return status about The rules are applied incrementally per hierarchical design correlate lower-level (e.g., temperature, indices) with higher-level vapor pressure deficit) robustly determine main have led it. case was conducted, collection satellite images artichoke crops Salerno, Italy, demonstrate incremental integration health monitoring. Subsequently, tests were conducted on vineyard regions interest Teano, efficacy framework assessment plant stress. Indeed, comparing outcomes our those cutting-edge machine learning (ML) semantic segmentation indeed revealed promising level accuracy. Specifically, classification performance compared output conventional ML methods, demonstrating consistent achieves an accuracy over 90% throughout various seasons year.

Language: Английский

Citations

2