Scholarlink

<p>Nutrient deficiencies represent primary factors leading to significant declines in commercial crop yield, impacting various growth-related factors. Ensuring sufficient plant nutrition is a paramount issue for both the quantity and quality of crop yields; thus, early identification of nutrient deficiency is essential. Deep learning advancements show remarkable advances in nutrient deficiency identification and crop yield loss prediction. This research introduced the soft attention-based (SABC) model for nutrient deficiency identification using the mobile net V2 as the base model and time series plant severity (TSPS) module for nutrient deficiency (ND) and crop yield loss (CYL) identification for identified ND. Plant leaf images are collected from real-time crop field environments to identify ND and CYL in groundnut plants. The plant leaf data are gathered at different ages of the plant, including 30, 60, and 90-day age groups, and validated using the data validation method. Leveraging this data allows for monitoring plant nutrient deficiencies in real-time crop field environments. The SABC model obtained an accuracy of 97.85% on the groundnut dataset and 98.14% on the rice dataset; our SABC model exhibits a 2.69% improvement in the accuracy of the current literature. This methodology provides farmers with a valuable tool for identifying nutrient deficiencies in crop fields, potentially revolutionizing nutrition management practices in 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

Enhanced real-time monitoring of nutrient deficiencies and crop yield loss prediction using a soft attention-based model and time series data analysis

  • 期刊论文

  • 发表日期：

    2026年5月11日

查看全文
  • Kummari Venkatesh,
  • K. Jairam Naik

摘要

Nutrient deficiencies represent primary factors leading to significant declines in commercial crop yield, impacting various growth-related factors. Ensuring sufficient plant nutrition is a paramount issue for both the quantity and quality of crop yields; thus, early identification of nutrient deficiency is essential. Deep learning advancements show remarkable advances in nutrient deficiency identification and crop yield loss prediction. This research introduced the soft attention-based (SABC) model for nutrient deficiency identification using the mobile net V2 as the base model and time series plant severity (TSPS) module for nutrient deficiency (ND) and crop yield loss (CYL) identification for identified ND. Plant leaf images are collected from real-time crop field environments to identify ND and CYL in groundnut plants. The plant leaf data are gathered at different ages of the plant, including 30, 60, and 90-day age groups, and validated using the data validation method. Leveraging this data allows for monitoring plant nutrient deficiencies in real-time crop field environments. The SABC model obtained an accuracy of 97.85% on the groundnut dataset and 98.14% on the rice dataset; our SABC model exhibits a 2.69% improvement in the accuracy of the current literature. This methodology provides farmers with a valuable tool for identifying nutrient deficiencies in crop fields, potentially revolutionizing nutrition management practices in agriculture.