Scholarlink

<p>Soil is a continuous body covering a segment of land surface on earth where crops or plants may grow. Soils are classified by various processes designed for evaluating soil types. However, soil type classification and assessment is generally a costly and time-consuming process. Here, NBBWO and SpinalNet_NBBWO are devised for routing and soil type classification. Initially, the Internet of Things (IoT) is simulated, and then routing is accomplished based on NBBWO, which combines Namib beetle optimization (NBO) with beluga whale optimization (BWO). The routing process is conducted by considering fitness parameters including energy, delay, trust, and distance. Afterwards, soil type classification is performed at the base station. To conduct soil type classification, first, features are chosen based on chord distance. Then, data are augmented by the oversampling technique. Lastly, soil type classification is carried out by SpinalNet, and its training process is performed by NBBWO.</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

Namib Beetle Beluga Whale Optimization-Enabled SpinalNet for Soil Type Classification in the Internet of Things

  • 期刊论文

  • 发表日期：

    2025年8月1日

查看全文
  • S. Sivakumar,
  • S. Premkumar,
  • T. S. Arthi,
  • G. Sakthi

摘要

Soil is a continuous body covering a segment of land surface on earth where crops or plants may grow. Soils are classified by various processes designed for evaluating soil types. However, soil type classification and assessment is generally a costly and time-consuming process. Here, NBBWO and SpinalNet_NBBWO are devised for routing and soil type classification. Initially, the Internet of Things (IoT) is simulated, and then routing is accomplished based on NBBWO, which combines Namib beetle optimization (NBO) with beluga whale optimization (BWO). The routing process is conducted by considering fitness parameters including energy, delay, trust, and distance. Afterwards, soil type classification is performed at the base station. To conduct soil type classification, first, features are chosen based on chord distance. Then, data are augmented by the oversampling technique. Lastly, soil type classification is carried out by SpinalNet, and its training process is performed by NBBWO.