Scholarlink

<p>Effective soil type classification is essential for improving agricultural methods, encouraging sustainable land usage, and increasing resource efficiency. This research introduces a novel approach termed Xception Convolutional Neural Network with Addax Snow Leopard Optimization Algorithm (XcovNet_Ax-SLOA) to address the soil classification challenges in an Internet of Things (IoT) environment. The Ax-SLOA algorithm integrates the Addax Optimization Algorithm (AOA) and the Snow Leopard Optimization Algorithm (SLOA) to better optimize routing paths and ensure reliable data transmission from IoT sensors to the base stations. At the base station, the input data collected from the dataset is first normalized using median normalization. Feature extraction is then performed by Shapley Additive exPlanation (SHAP). Finally, the soil type classification is done by the presented XcovNet_Ax-SLOA. The results of the experiments demonstrate that the proposed XcovNet_Ax-SLOA method achieved an accuracy of 96.566%, a specificity of 94.355%, and a sensitivity of 97.355%. Additionally, the Ax-SLOA-based routing provides better network performance with a throughput of 94.211 Mbps and energy of 96.393&#xa0;J. The proposed method shows strong applicability in precision agriculture and environmental monitoring, where accurate soil classification and effective data communication are important.</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

Ax-SLOA: Addax Snow Leopard Optimization Algorithm with XCovNet for Soil Type Classification in IoT

  • 期刊论文

  • 发表日期：

    2026年2月7日

查看全文
  • K.P. Sriram,
  • P. Kola Sujatha,
  • Kannan Arputharaj

摘要

Effective soil type classification is essential for improving agricultural methods, encouraging sustainable land usage, and increasing resource efficiency. This research introduces a novel approach termed Xception Convolutional Neural Network with Addax Snow Leopard Optimization Algorithm (XcovNet_Ax-SLOA) to address the soil classification challenges in an Internet of Things (IoT) environment. The Ax-SLOA algorithm integrates the Addax Optimization Algorithm (AOA) and the Snow Leopard Optimization Algorithm (SLOA) to better optimize routing paths and ensure reliable data transmission from IoT sensors to the base stations. At the base station, the input data collected from the dataset is first normalized using median normalization. Feature extraction is then performed by Shapley Additive exPlanation (SHAP). Finally, the soil type classification is done by the presented XcovNet_Ax-SLOA. The results of the experiments demonstrate that the proposed XcovNet_Ax-SLOA method achieved an accuracy of 96.566%, a specificity of 94.355%, and a sensitivity of 97.355%. Additionally, the Ax-SLOA-based routing provides better network performance with a throughput of 94.211 Mbps and energy of 96.393 J. The proposed method shows strong applicability in precision agriculture and environmental monitoring, where accurate soil classification and effective data communication are important.