Intelligent phytoprotection systems use modern information and communication technologies to protect crop growth while optimizing the required human labor. A fundamental challenge is to understand the complex biological environment through the set of sensors and develop intelligent phytoprotection applications at the system level. Recent technological advances in machine vision, data mining, image processing, cloud computing, edge computing, and Internet of Things (IoT) have significantly enabled AI-driven smart crop protection systems. Most of the current studies rely on large labeled datasets with high cost of data acquisition and annotation. As a necessary addition in the community, few-shot learning aims to learn from limited labeled data to obtain generalized models. Therefore, Internet of Things (IoT) data collection and processing is crucial, while data quality and security are essential for Deep Learning based on Big Data and Little-shot learning based on limited data in AI-driven intelligent crop protection. Instead of using grossly large amounts of redundant data, data information analysis and efficient learning methods based on limited data may be more useful for certain real-world tasks in intelligent crop protection. The future smart agricultural systems should use limited high-quality data and seriously consider data security, which will benefit the use of smart crop protection and agricultural equipment.
This Research Topic aims to collect researches focusing on data gathering, quality and security in AI-driven intelligent phytoprotection systems, especially advanced few-shot learning and deep learning based on data mining and information evaluation. Both original papers and review articles are welcome. Possible contributions include the following:
- Data gathering from smart IoT sensors in intelligent phytoprotection systems.
- Data security and privacy in the AI-driven intelligent phytoprotection systems.
- Data mining and information analysis for AI-driven phytoprotection systems.
- New techniques and theories for "few-shot learning" in phytoprotection systems
- Fusion and processing of multi-source data for deep learning in phytoprotection systems
- Advanced frameworks for few-shot learning or deep learning in phytoprotection systems
- Cloud/edge computing for deep learning or little-shot learning in phytoprotection systems
- Specific hardware deployment and model acceleration in intelligent phytoprotection systems
- Specific applications, e.g., plant pest and disease detection, yield prediction, irrigation strategy, online agricultural data acquisition, smart agricultural machinery, data communication, etc.
Intelligent phytoprotection systems use modern information and communication technologies to protect crop growth while optimizing the required human labor. A fundamental challenge is to understand the complex biological environment through the set of sensors and develop intelligent phytoprotection applications at the system level. Recent technological advances in machine vision, data mining, image processing, cloud computing, edge computing, and Internet of Things (IoT) have significantly enabled AI-driven smart crop protection systems. Most of the current studies rely on large labeled datasets with high cost of data acquisition and annotation. As a necessary addition in the community, few-shot learning aims to learn from limited labeled data to obtain generalized models. Therefore, Internet of Things (IoT) data collection and processing is crucial, while data quality and security are essential for Deep Learning based on Big Data and Little-shot learning based on limited data in AI-driven intelligent crop protection. Instead of using grossly large amounts of redundant data, data information analysis and efficient learning methods based on limited data may be more useful for certain real-world tasks in intelligent crop protection. The future smart agricultural systems should use limited high-quality data and seriously consider data security, which will benefit the use of smart crop protection and agricultural equipment.
This Research Topic aims to collect researches focusing on data gathering, quality and security in AI-driven intelligent phytoprotection systems, especially advanced few-shot learning and deep learning based on data mining and information evaluation. Both original papers and review articles are welcome. Possible contributions include the following:
- Data gathering from smart IoT sensors in intelligent phytoprotection systems.
- Data security and privacy in the AI-driven intelligent phytoprotection systems.
- Data mining and information analysis for AI-driven phytoprotection systems.
- New techniques and theories for "few-shot learning" in phytoprotection systems
- Fusion and processing of multi-source data for deep learning in phytoprotection systems
- Advanced frameworks for few-shot learning or deep learning in phytoprotection systems
- Cloud/edge computing for deep learning or little-shot learning in phytoprotection systems
- Specific hardware deployment and model acceleration in intelligent phytoprotection systems
- Specific applications, e.g., plant pest and disease detection, yield prediction, irrigation strategy, online agricultural data acquisition, smart agricultural machinery, data communication, etc.
