Scholarlink

<p>Effective paddy field monitoring is essential for optimizing resource management and ensuring food security. This study explores the potential of dual-polarimetric Sentinel-1 SAR data for paddy monitoring and classification in the Cauvery Delta Zone, the rice bowl of Tamil Nadu, India, using advanced polarimetric decomposition techniques and machine learning models. Temporal variations of key polarimetric parameters, including Alpha (ranging from 8° to 37°), Anisotropy (0.4 to 0.9), and Entropy (0.4 to 1.0), were obtained across early and peak monsoon seasons, illustrating paddy phenological stages. The analysis combines the H/Alpha decomposition parameters with the Random Forest (RF) algorithm to classify paddy fields according to their distinct scattering mechanisms. The RF classifier demonstrated impressive accuracy at 90.3%, accompanied by an Area Under the Curve (AUC) value of 0.92, which signifies strong model performance in distinguishing paddy from other vegetation and land-cover types. This investigation underscores the relevance of SAR-based polarimetric decomposition techniques for precise paddy monitoring and the importance of combining polarimetric parameters with machine learning for extensive agricultural assessments. The results can improve irrigation management, maximize resource use, and encourage sustainable agricultural practices in the Cauvery Delta Zone.</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

Mapping and monitoring paddy crops with sentinel-1 derived polarimetric parameters and machine learning

  • 期刊论文

  • 发表日期：

    2025年6月27日

查看全文
  • Janardhanam Niraimathi,
  • Subbarayan Saravanan

摘要

Effective paddy field monitoring is essential for optimizing resource management and ensuring food security. This study explores the potential of dual-polarimetric Sentinel-1 SAR data for paddy monitoring and classification in the Cauvery Delta Zone, the rice bowl of Tamil Nadu, India, using advanced polarimetric decomposition techniques and machine learning models. Temporal variations of key polarimetric parameters, including Alpha (ranging from 8° to 37°), Anisotropy (0.4 to 0.9), and Entropy (0.4 to 1.0), were obtained across early and peak monsoon seasons, illustrating paddy phenological stages. The analysis combines the H/Alpha decomposition parameters with the Random Forest (RF) algorithm to classify paddy fields according to their distinct scattering mechanisms. The RF classifier demonstrated impressive accuracy at 90.3%, accompanied by an Area Under the Curve (AUC) value of 0.92, which signifies strong model performance in distinguishing paddy from other vegetation and land-cover types. This investigation underscores the relevance of SAR-based polarimetric decomposition techniques for precise paddy monitoring and the importance of combining polarimetric parameters with machine learning for extensive agricultural assessments. The results can improve irrigation management, maximize resource use, and encourage sustainable agricultural practices in the Cauvery Delta Zone.