Scholarlink

<p>The farmers make the majority of their income from farming, and in order to improve crop development, constant attention and continuous monitoring is necessary. Precision agriculture is made possible by the ability of farmers to plot information from field, arrange and evaluate it to monitor their crops remotely using satellite imagery data. Since researchers are more interested in sugarcane development success, this cash crop is employed in this study. In this study, various ML methods used to remotely sensed sugarcane crop data are surveyed. In order to monitor the crop for identifying dense and sparse vegetation the remote sensing quality indicators and Machine Learning approaches are being investigated and put into practice. The same farmer’s plot can be observed every month to detect changes, and in subsequent research, enhanced vegetation indices can be used to identify the source of the scanty vegetation in a certain plot. Also pre-processing technique used is novel method proposed as Improved Cloud cover detection and removal. A combination of vegetation indices used as discriminating features after pre-processing Sentinel-2 data is used to monitor at grand growth stages of sugarcane crop in conjunction with ML models. The performance indicators, such as overall Accuracy, Precision, Recall is computed for baseline models and the Ensemble techniques further gave accuracy of 94% for categorising remotely data.</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

Remote sensing and machine learning methods to analyse the vegetation of sugarcane crop

  • 期刊论文

  • 发表日期：

    2025年6月4日

查看全文
  • Mansi Kambli,
  • Bhakti Palkar

摘要

The farmers make the majority of their income from farming, and in order to improve crop development, constant attention and continuous monitoring is necessary. Precision agriculture is made possible by the ability of farmers to plot information from field, arrange and evaluate it to monitor their crops remotely using satellite imagery data. Since researchers are more interested in sugarcane development success, this cash crop is employed in this study. In this study, various ML methods used to remotely sensed sugarcane crop data are surveyed. In order to monitor the crop for identifying dense and sparse vegetation the remote sensing quality indicators and Machine Learning approaches are being investigated and put into practice. The same farmer’s plot can be observed every month to detect changes, and in subsequent research, enhanced vegetation indices can be used to identify the source of the scanty vegetation in a certain plot. Also pre-processing technique used is novel method proposed as Improved Cloud cover detection and removal. A combination of vegetation indices used as discriminating features after pre-processing Sentinel-2 data is used to monitor at grand growth stages of sugarcane crop in conjunction with ML models. The performance indicators, such as overall Accuracy, Precision, Recall is computed for baseline models and the Ensemble techniques further gave accuracy of 94% for categorising remotely data.