Scholarlink

<p>Abiotic stresses such as drought, salinity, extreme temperature, nutrient deficiency, heavy metals, and flooding significantly threaten global agricultural productivity by disrupting plant physiological, biochemical, and molecular processes. Early and accurate detection of these stresses is crucial for minimising yield losses; however, traditional monitoring approaches are often slow, subjective, and unable to capture subtle pre-symptomatic changes. Recent advances in artificial intelligence (AI) have transformed plant stress research by enabling rapid, data-driven analysis of complex biological signals. This review synthesizes current progress in AI-driven plant stress detection, highlighting the integration of machine learning and deep learning algorithms with advanced detectors and sensing platforms, including hyperspectral imaging (HSI), thermal cameras, Internet of Things (IoT)-based soil and water sensors, nanoscale biosensors, wearable plant devices, and remote sensing systems such as drones and satellites. These AI-powered technologies allow continuous and non-invasive monitoring of plant health, providing insights into stress-specific signatures associated with various abiotic stresses. This study evaluates the advantages of AI-based systems, such as early detection, high-throughput phenotyping, and real-time decision support, alongside prevailing challenges related to data standardization, model interpretability, environmental variability, and accessibility for low-resource farming systems. Finally, future perspectives are discussed for the potential of multimodal data fusion, digital twins, edge AI devices, and AI-integrated breeding pipelines to enhance crop resilience in a&#xa0;changing climate. Overall, this review demonstrates how AI is revolutionising plant stress diagnostics and paving the way for sustainable, predictive, and precision agriculture.</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

Artificial Intelligence-Assisted Detection of Abiotic Stress in Agricultural Crops: Sensors, Computational Models, and Outlook

  • 期刊论文

  • 发表日期：

    2026年4月28日

查看全文
  • Komal Sharma,
  • Upma Bhatt,
  • Vineet Soni

摘要

Abiotic stresses such as drought, salinity, extreme temperature, nutrient deficiency, heavy metals, and flooding significantly threaten global agricultural productivity by disrupting plant physiological, biochemical, and molecular processes. Early and accurate detection of these stresses is crucial for minimising yield losses; however, traditional monitoring approaches are often slow, subjective, and unable to capture subtle pre-symptomatic changes. Recent advances in artificial intelligence (AI) have transformed plant stress research by enabling rapid, data-driven analysis of complex biological signals. This review synthesizes current progress in AI-driven plant stress detection, highlighting the integration of machine learning and deep learning algorithms with advanced detectors and sensing platforms, including hyperspectral imaging (HSI), thermal cameras, Internet of Things (IoT)-based soil and water sensors, nanoscale biosensors, wearable plant devices, and remote sensing systems such as drones and satellites. These AI-powered technologies allow continuous and non-invasive monitoring of plant health, providing insights into stress-specific signatures associated with various abiotic stresses. This study evaluates the advantages of AI-based systems, such as early detection, high-throughput phenotyping, and real-time decision support, alongside prevailing challenges related to data standardization, model interpretability, environmental variability, and accessibility for low-resource farming systems. Finally, future perspectives are discussed for the potential of multimodal data fusion, digital twins, edge AI devices, and AI-integrated breeding pipelines to enhance crop resilience in a changing climate. Overall, this review demonstrates how AI is revolutionising plant stress diagnostics and paving the way for sustainable, predictive, and precision agriculture.