images captured by smartphones can be sent to plant pathologists in remote laboratories for further disease identification. Lab experts can directly suggest cure and
prevention for the diseases. Another agricultural use of crowdsourcing is fertilizer
calculation. There are commercial mobile device-based optical applications to estimate the color level of rice leaves and compare the crowdsourced images and
recommend the required amounts of nitrogen fertilizer for the field (Pongnumkul
et al. 2015). Atmospheric data from smartphone sensors and amateur weather
stations have been already utilized by some applications (Muller et al. 2015).
Decision tree algorithm is implemented in a crowdsourcing mobile system to help
generate accurate and reliable decision on agricultural plant diseases (Singh et al.
2014). Mobile4D is an integrated mobile system for crowdsourcing-based disaster
alerting and reporting system which could be used in minimizing the impact of small
disasters on crops and livestock (Frommberger and Schmid 2013). CrowdHydrology
is another crowdsourcing project to encourage citizen scientists to voluntarily send
hydrologic measurements via text messages to a server which distributes the information on the Web (Lowry and Fienen 2013).
Crowdsourced datasets are released as a common knowledge database free for the
entire human being. It has become a reliable way for consumers to retrieve data at a
very low cost. The crowdsourced data has already been used in agriculture as a
significant supplement. It has greatly changed the whole situation of data supplying
market. Its coverage will undoubtedly be further expanded in the next few years.
4.4 Conclusion
This chapter summarized state-of-the-art data sources of agro-geoinformatics and the
sourcing methods. The data sources can be divided into four classes: satellite,
airborne, in situ reports, and human reports. The details of each source are investigated and introduced. Basically, the satellite data source has the best spatial coverage
and long observation history. The airborne and in situ datasets are mostly casespecific or site-specific. Human reports are brief descriptions and mostly concise
terms and numbers answering basic questions. The different sources are not
completely overlapped and can be integrated to obtain an immersive understanding
of the truth in the crop fields. The data sourcing has three major options: conventional, cloud-based, and crowdsourcing. Conventional sourcing has a very tough
docking process. Cloud service simplifies the uploading and distributing of big data.
Crowdsourcing greatly lowers the cost of data collection and retrieval. Nowadays,
the three sourcing methods coexist and share the market. The latter two are gradually
seizing the major portion. The future development is towards the Internet-based,
mobile friendly, big data, low-cost, robustness, and high-performance data
distribution.
4 Agro-geoinformatics Data Sources and Sourcing
61
prevention for the diseases. Another agricultural use of crowdsourcing is fertilizer
calculation. There are commercial mobile device-based optical applications to estimate the color level of rice leaves and compare the crowdsourced images and
recommend the required amounts of nitrogen fertilizer for the field (Pongnumkul
et al. 2015). Atmospheric data from smartphone sensors and amateur weather
stations have been already utilized by some applications (Muller et al. 2015).
Decision tree algorithm is implemented in a crowdsourcing mobile system to help
generate accurate and reliable decision on agricultural plant diseases (Singh et al.
2014). Mobile4D is an integrated mobile system for crowdsourcing-based disaster
alerting and reporting system which could be used in minimizing the impact of small
disasters on crops and livestock (Frommberger and Schmid 2013). CrowdHydrology
is another crowdsourcing project to encourage citizen scientists to voluntarily send
hydrologic measurements via text messages to a server which distributes the information on the Web (Lowry and Fienen 2013).
Crowdsourced datasets are released as a common knowledge database free for the
entire human being. It has become a reliable way for consumers to retrieve data at a
very low cost. The crowdsourced data has already been used in agriculture as a
significant supplement. It has greatly changed the whole situation of data supplying
market. Its coverage will undoubtedly be further expanded in the next few years.
4.4 Conclusion
This chapter summarized state-of-the-art data sources of agro-geoinformatics and the
sourcing methods. The data sources can be divided into four classes: satellite,
airborne, in situ reports, and human reports. The details of each source are investigated and introduced. Basically, the satellite data source has the best spatial coverage
and long observation history. The airborne and in situ datasets are mostly casespecific or site-specific. Human reports are brief descriptions and mostly concise
terms and numbers answering basic questions. The different sources are not
completely overlapped and can be integrated to obtain an immersive understanding
of the truth in the crop fields. The data sourcing has three major options: conventional, cloud-based, and crowdsourcing. Conventional sourcing has a very tough
docking process. Cloud service simplifies the uploading and distributing of big data.
Crowdsourcing greatly lowers the cost of data collection and retrieval. Nowadays,
the three sourcing methods coexist and share the market. The latter two are gradually
seizing the major portion. The future development is towards the Internet-based,
mobile friendly, big data, low-cost, robustness, and high-performance data
distribution.
4 Agro-geoinformatics Data Sources and Sourcing
61