Smart agriculture: precision farming through sensor-based crop monitoring and control system
The escalation of the global population and the depletion of natural resources have propelled the evolution of smart agriculture and precision farming, underpinned by sensor-based crop monitoring and control systems, which are anticipated to revolutionize the agricultural sector. Notab...
Saved in:
| Main Authors: | , |
|---|---|
| 格式: | Article |
| 语言: | English |
| 出版: |
Universiti Malaysia Pahang Al-Sultan Abdullah Publishing
2024
|
| 主题: | |
| 在线阅读: | http://irep.iium.edu.my/119724/7/119724_Smart%20agriculture%20precision%20farming.pdf http://irep.iium.edu.my/119724/ https://journal.ump.edu.my/mekatronika/article/view/10562/3430 |
| 标签: |
添加标签
没有标签, 成为第一个标记此记录!
|
| 总结: | The escalation of the global population and the depletion of natural resources have propelled the evolution of smart agriculture and precision farming, underpinned by sensor-based crop monitoring and control systems, which are anticipated to revolutionize the agricultural sector. Notably, prevalent smart agriculture systems predominantly emphasize either IoT components for data monitoring and control or machine learning components for data analysis. Consequently, this project endeavours to develop a system that seamlessly integrates both IoT and machine learning components, culminating in an advanced system capable of real-time crop monitoring and growth prediction. Collaborating with the Urban Farming Farm under the auspices of the Kulliyyah of Economics and Management Science, an IoT system comprising soil moisture, temperature, and humidity sensors, alongside an actuator, is devised to facilitate data acquisition and required intervention specifically for Okra Fruit during the pre-harvesting stage. Subsequently, four distinct algorithms are trained with the collected dataset to ascertain the most optimal algorithm for predicting crop growth and harvesting time, resulting in the selection of the Random Forest Regression model, which attains the highest model score of 86%. Upon its integration into the comprehensive system for monitoring new data and predicting fruit growth, the model achieves an impressive 98% accuracy score. Future endeavours for this project aim to enhance its applicability and predictive capabilities through the incorporation of diverse datasets from various plant species, the expansion of crop predictions to encompass the entire growth cycle, the integration of additional sensors, and the enhancement of the system's scalability to cover larger areas. |
|---|