AGIT.in.th
Agriculture & Information Technology
ความสำเร็จของ AG-IT ในการพัฒนาการเกษตรและสังคมในญี่ปุ่น

Successful Information Technology (It) for Agriculture and Rural Development

Seishi Ninomiya
National Agricultural Research Center,
National Agricultural Research Organization
Kannondai, Tsukuba, Ibaraki 305-8666, Japan, 2004-09-01

This Bulletin discusses the role of information technology (IT) and its practical contributions to agriculture and rural development. It also presents Japan’s experiences in the use of IT in agriculture, and identifies the requirements and the issues needed to make practical use of IT systems for the agricultural domain. This paper also introduces new technologies that can fulfill the requirements and give solutions to resolve said issues, showing several successful applications of the technologies. Finally, it emphasizes on the importance of international collaboration in promoting the use of IT in agriculture and rural development.

Abstract

Introduction

Information technology (IT) doubtlessly contributes much to agriculture and rural development. Firstly, it can facilitate rural activities and provide more comfortable and safe rural life with equivalent services to those in the urban areas, such as provision of distance education, tele-medicine, remote public services, remote entertainment etc. Secondly, IT can initiate new agricultural and rural business such as e-commerce, real estate business for satellite offices, rural tourism, and virtual corporation of small-scale farms. Thirdly, it can support policy-making and evaluation on optimal farm production, disaster management, agro-environmental resource management etc., using tools such as geographic information systems (GIS). Fourthly, it can improve farm management and farming technologies by efficient farm management, risk management, effective information or knowledge transfer etc., realizing competitive and sustainable farming with safe products. For example, farmers must make critical decisions such as what to and when to plant, and how to manage pests, while considering off-farm factors such as environmental impacts, market access, and industry standards. IT-based decision support system (DSS) can surely help their decisions. Fifthly, IT can provide systems and tools to secure food traceability and reliability that has been an emerging issue concerning farm products since serious contamination such as BSE and chicken flu was detected. Finally, IT can take an important and key role for industrialization of farming or farm business enterprises, combining the above roles.

Japan’s It Experiences

IT policy for agriculture and rural development started late in 1980s in Japan but it had not been successful for a long time. The policy gave higher priority to hardware than software, resulting in insufficient data resources and poor applications that were not useful enough to convince farmers of the beneficial effect of IT in agriculture. Poor rural network infrastructure and IT literacy contributed to this failure. The latest statistics of the Ministry of Agriculture, Forestry and Fisheries (MAFF) (http://www.maff.go.jp/toukei/sokuhou/data/pcikou2003/pcikou2003.htm, December 2002) shows that 50% of farm households own personal computers (PCs) but only 10% of them use the PCs for farming. This number is much lower than that of other industries. This fact clearly indicates that farmers are not convinced about the benefits of IT in agriculture.

Following the e-Japan Strategy (http://www.kantei.go.jp/foreign/it_e.html) announced by the Japanese government in 2000, MAFF drew up an IT strategy for agriculture in the 21st century (http://www.maff.go.jp/soshiki/kambou/joho/it-sen/it-sen.htm). This strategy is substantially different from the former policy by emphasizing the importance of enrichment of digital contents and rural IT literacy issue.

After the e-Japan strategy, severe commercial competition has brought the country a rapid spread of very cheap broadband Internet (ADSL, CATV & FTTH) in urban areas. However, broadband connectivity is still quite poor in rural areas because of absence of commercial competition. This fact also limits the IT extension in the rural areas, as the Internet is apparently an inevitable core infrastructure in IT utilization.

It Applications in Agriculture and Rural Development

Looking at the present status in Japan, we can identify existing issues that we need to solve in order to extend IT to the agricultural domain. Agriculture stands on the very complex interaction between biological, climatic and geographical factors in addition to human economic activities. The information under such a complicated system is unpredictable, unstable, subjective, site-specific and reliant on empirical decisions given the inherent variability of biological phenomena. Agricultural information with these features is typically beyond the scope of the information science used in industrial information systems, and this has surely led to the failure of IT in agriculture. We should also consider how to easily collect field data. Though field data are the basis for farm decision support, few people realize the importance of it in developing several decision support programs.

Poor network infrastructure in rural areas is also one of the obstacles for IT in agriculture because the Internet is an important factor in whatever information system we develop nowadays and it usually helps reduce cost of system development and maintenance. Another difficult problem is computer literacy in the rural areas. Finally, we do have to convince farmers of the benefits of IT. The key factors to consider for IT in agriculture are as follow;

  • 1. How to adjust software to the special features of agricultural information;
  • 2. How to enrich digital contents;
  • 3. How to utilize the Internet, especially to reduce time and monetary costs;
  • 4. How to provide easy-to-use systems for computer literacy; and
  • 5. How to convince farmers of the potential benefits of IT.

In the following sections, several key technologies to solve these issues are introduced with their applications to agricultural domain.

Efficient and Low Cost Field Data Acquisitions

Undoubtedly, digital data contents are most important when we develop agriculture information systems. Actually, fundamental and widely used data such as market information, weather information, and agricultural material information is becoming available in the Internet through several services by national and local governments, semigovernmental agencies, academic institutes, commercial sectors etc. and both the quality and quantity are advancing, though the data’s inflexible interface and associated charges render it unsuitable for general use especially by end users.

In addition to the above-mentioned fundamental data, site-specific field data are definitely necessary when a farmer requires some site-specific decision support. For example, a growth model may require soil and fertilization information for accurate predictions. To obtain such information, the farmer has to record field data continuously for a considerably long period. In spite of such importance, providing tools for farmers to easily collect these data is often forgotten and neglected. The following systems have been recently suggested to solve this problem.

To support to record farm-working journal that is one of the most fundamental data from the fields, Kouno et al. (1998 and 2000) developed a system combined with a web camera and a metrological robot. The web camera automatically collects crop images used to remotely analyze plant growth and condition. Application of web camera (Fig. 1) to agriculture, which was originally suggested by Ninomiya et al. (1997), is now very common as reasonably cheap cameras and easy-to-use software become available.

Sugawara (2001) developed a mobile-phone-based farm-working journal (Fig. 2, Sugawara 2001) to collect field data. The software is web-based and we can directly upload farming data to a database from the fields.

Otuka and Yamakawa (2003) and Otuka and Sugawara (2003) developed PDA-based field data collection systems combined with GPS (Global Positioning System). They record farming data on PDAs at fields combined with location data automatically measured by GPS and synchronize the data with PCs so that they can handle collected data, using a PC-based GIS application. Matsumoto and Machida (2002) and Kamiya and Machida (2002) utilized voice-recognition technologies to record farming data in the fields. These systems are easy to use and their mobility strongly support farmers in situ data collections.

Recently, Fukatsu et al. (2003) developed a field monitoring system called FieldServer (Fig. 3, http://model.job.affrc.go.jp/FieldServer/FieldServerEn/default.htm). A FieldSever originally has ordinal sensors such as temperature, solar radiation, moisture and soil temperature. It has very flexible interface and can optionally have several types of sensor such as a web camera, an infrared sensor, wind speed, wind direction and leaf wetness. In addition to its sensing functions, FieldServer can serve as a wireless LAN access point so that each FieldSever can establish a wireless network with other FieldServers. This indicates that a whole region can be covered by the Internet accessible wireless hot spot, having several FieldServers deployed and just one link point to the Internet in the region (Fig. 4). Latest version of the FieldServer is completely autonomous without any requirement for electric supply.

A FieldServer is remarkably cheap (<US$300) and as accurate as an ordinal expensive weather robot as a sensing system. Using its wireless LAN hotspot function, this can easily be used in rural areas.

Field data acquisitions are becoming even more important because of the recent movement toward traceability of agricultural products, in which information must be easily traced to the original farming conditions, e.g. varieties, pesticide spray, harvest dates and producer names.

Case-Based Knowledge Management and Decision Support

Natural science is based on a reductive approach. In this approach, the whole is separated into elements and each element is theoretically explained one by one. Then, the elements are compiled again to explain the whole. But in agriculture, where many factors interact complicatedly with each other, this approach sometimes fails. We have not succeeded in fully reductively explaining even the growth of rice yet, whereas, empirical judgments by expert farmers are often sufficiently effective. The purpose of agricultural information systems is not to find out the truth, but to provide optimal decision support for farmers, and the reductive approach is not the only way. Case-bases provide one alternative approach.

A case-base is a kind of database that stores empirical cases and has a function to recommend relevant cases according to users’ decision making queries. Our group (Otuka and Ninomiya 1998, Otuka and Kitamurua 2002) developed a prototype case base system (Fig. 5), using a concept search engine that is based on latent semantic indexing (Deerwester 1990). In other words, this is a search based on meaning. Using the system, one can retrieve cases without entering any keywords. The user can enter normal sentences as queries to the system and the system searches for recommended cases corresponding to the queries, based on the context or concept of the queries. This is a typical non-reductive approach and seems to be a very powerful way to transfer knowledge for farm decision support. The cases result from highly complicated farm factors and contain plenty of useful information. However, they are usually preserved in plain texts and are very difficult to properly retrieve with keyword-based searches because the contexts of the cases themselves provide useful information.

The case-based approach can be applied to several types of cases collected in many ways. E-mails exchanged in mailing lists are good examples of cases. Images can be cases as well as ordinary texts. An example (Fig. 6) was suggested by Xin et al. (1998). In the system they developed, farmers request that extension services diagnose disease and pests by sending queries and images taken by digital cameras by e-mail, and extension services reply to these. These question-and-answer (Q&A) and images are automatically stored as cases for a forthcoming automated Q&A system where the stored images and queries will be automatically matched with new queries and images to find out the proper answer.

A new approach called case-based modeling has also been suggested recently (Hoshi et al. 2000). To model some phenomena such as crop growth, we usually take either mechanistic modeling or statistical modeling. Case-base modeling is partially close to statistical modeling but it is not always based on probability as statistical modeling does.

User Interface

IT literacy is also tremendously important. In addition to reinforcement of IT education and training, easy-to-use interface is a big challenge. Regular PC interface with a keyboard is definitely unacceptable by majority of farmers. Several technologies are now available to provide easy-to-use systems for those end users. For example, mobile phone-based interface is surely promising as one of the solutions for this issue. The coverage of mobile phones is expanding even in developing countries and small number of keys makes it easy to use. Because farmers do not require complicated decision supports, its simple screen is usually acceptable. The second and third generation mobile phones provide seamless connectivity to the Internet and can substitute for regular PCs if we can have applications suitable to the mobile phone interface. Actually, mobile phones are now used not only for data collection but also for in situ decision support in the fields. For example, Laurenson et al. (2002) and Sasaki et al. (2002) developed mobile-phone based applications to access weather database so that farmers can always check weather conditions in their fields. More practical applications such as a pesticide warning system have been also prototyped. Its easier interface than PC, and its mobility to fields have been welcomed by farmers.

Another example is the so-called Navigation Search Engine (Fig. 7), which does not require any keyboard usage. A user clicks a mouse, selecting the queries automatically given by the engine and can finally reach an appropriate search result without knowing much about the topic.

Geographical information system (GIS) technology is also promising in agriculture especially as a user interface to integrate several types of data sets, because agricultural information typically extends spatially, and it is often necessary and convenient to handle it at a regional scale. An application of GIS to agricultural decision support developed by Fulcher (1999) is a good example. The system predicts the benefit to a farmer and the environmental stress based on the various scenarios about crop conversion planned by the user, integrating many layers of geographical data such as soil conditions and the watershed. Fulcher’s group also provides several Web-based GIS systems to the public (Lee 1998, http://www.cares.missouri.edu/), indicating the importance of sharing Internet-based resources such as huge data sets and expensive GIS systems. For successful GIS applications, how to collect datasets for several layers is most important. This is always a big issue because these datasets usually belong to different organizations. Grid concept, which will be discussed later in this paper, is important in this regard. Once datasets and some analytical methods become available, GIS can be a powerful tool/interface for agricultural and rural DSS.

Recently, video conferencing facility is becoming popular and much cheaper than before. This can be a good interface to distance learning systems.

Datamining

We have now a huge amount of data in agricultural production and experiments that have been recorded for over 100 years since modern agriculture started. These long-term data may be sources of critical information that will give us new knowledge in agricultural production. At present, most of the data are still in printed materials and, though the performance of OCR (Optical Character Recognition) has rapidly improved, considerable labor is still required to digitize those data for an agricultural information system. However, once digitized, a modern technology called data-mining is available to analyze the resulting huge amount of data. It is based on several statistical procedures and graphical presentation of data and analyzed results, and named in an analogy with mining gold from gigantic mountains. As mentioned already, the background of agriculture is complicated and a huge number of factors are related to it. Datamining technology is expected to help us to mine unknown facts from huge amounts of data.

Eagbusiness and Traceability

Emarket Place and Direct Marketing

New businesses in agriculture are emerging, utilizing IT. Web-based marketing is typical of these initiatives. Web-based marketing was originally initiated as a direct marketing system between farmers and consumers. This style of business called B2C has been steadily growing as the Internet grows. Internet malls that virtually combine several farmers growing various commodities seem to be particularly promising, though have not yet achieved critical mass. Recently, a new style of business called B2B has been growing rapidly. This bridges farmers and wholesalers, substituting for fresh markets by providing virtual market places over the Internet. Web-based marketing is typically successful in Korea.

Related to the direct marketing business, a Japanese farmer started a very interesting trial. He equipped his greenhouse with a web camera system (Fig. 8), originally in order to analyze the growth of his crops. He then went on to utilize the system for his sales promotion (Fig. 8). He sold plantlets of melons to consumers and undertook the management of those until harvest. While the plants were growing, he let the consumers access his web camera to observe their plants. This idea was popular with urban people, giving them a virtual experience of farming.

Food Traceability

After the BSE came out, traceability of foods became a big issue. Several systems have been developed and IT is more or less a key technology to guarantee traceability. One simple example is the VIPS developed by Sugiyama (Fig. 9, http://vip2.nfri.affrc.go.jp/, site in Japanese). The system provides consumers farming records on crops that they purchase by using a code number and URL printed on the product label. New systems also provide information of distribution route of products to realize true traceability. In the near future, RFID (radio frequency identification) tags may replace legacy printed tags such as bar codes because of easier non-contact wireless readability and more information capacity. In any case, field data is the basis of farm product traceability and the importance of field data acquisition should be emphasized in this regard as well, considering reduction of farmer’s workload on recording. A mobile phone-based recording system that is linked to traceability has already been operational in a region of Japan for some crops.

Virtual Corporation

Small-scale farming is typical in the Asian region. It is a main cause of the inefficient agricultural productivity and the lack of its global competitiveness. A simple solution is to merge small-scale lands to a big scale one. The land ownership, however, makes it difficult as the number of landowning farmers increased based on modernization. One solution we can expect is to virtually integrate those small-scale farmers while keeping their financial independency. For example, a group of farmers can purchase chemicals with cheaper price than they can when they order individually. Or, if they can share machineries, the total cost on them can be reduced. We can expect similar cost reduction in marketing, logistics, risk management etc. as merits of scale. To realize such cooperation, the help of IT is inevitable in many ways.

Distributed System and Grid Technology

Grid-Based Decision Support

The most important advantage of the Internet is its use in information sharing between distributed resources. Such information sharing can greatly increase the amount of data available to users. Unfortunately, the will to share information in agricultural information systems is still weak and even at one site the same data sets can be duplicated ineffectively. However, new technologies to utilize the Internet make it possible for us to develop a distributed system called DataGrid for agriculture, which provides improved access to programs and effective utilization of available databases. The basic idea of DataGrid is acceptance of heterogeneity and autonomy of distributed resources.

In a Grid-base decision support system (Fig. 10), the network provides users with the necessary access to dynamically-linked programs and in situ data (Ninomiya 2001). This approach provides the following benefits:

  • 1. Multiple users can share a single executable module, avoiding duplication of software development and maintenance.
  • 2. Multiple programs can share the same data set, avoiding duplicated data maintenance and management.
  • 3. Data sets and programs are dynamically linked in the Internet, providing diverse functionality to users.
  • 4. Programs and data sets are managed by their owners, facilitating updates and maintenance.
  • 5. Developers have powerful sub model and data access components from which to assemble new programs.
  • 6. Above features reduce cost of system development and maintenance.

Data Mediation and Broker

These benefits match the requirements of agricultural decision support system (DSS) for diverse data sets and multiple subprograms that we have to handle, but to realize this Grid-based DSS framework, we face a major problem that similar kinds of data are stored in heterogeneous ways. For example, the weather databases published on the Internet differ in their logical structure, data management software and access method. Adapting each DSS to each database would result in inefficiency and redundancy as shown in Fig. 11.

We proposed the concept of data mediation as a solution for this issue, implementing the weather data mediation software MetBroker (Fig. 12, http://www.agmodel. org/). MetBroker can absorb the differences between weather databases, providing a consistent interface to client applications such as crop growth prediction models. MetBroker does not require any modification of the original databases (Fig. 12, Laurenson et al. 2000 and 2002).

MetBroker provides applications with details of available data, receives requests from client applications specifying the elements, resolution and period required, queries remote databases and returns results to the client program. Requests can be either for a single station or the geographical area of interest. In the latter case, MetBroker returns results from multiple databases to the client, which is unprecedented functionality. The results of geographical requests can be used for spatial interpolation. MetBroker utilizes a powerful metadata structure to provide catalogues of available data to client applications, and identify which databases should service geographical requests. Fig. 13 shows a comparatively simple application of MetBroker that retrieves weather data sets from several heterogeneous database based on the request given by end users.

MetBroker currently offers access to over 5000 stations in 14 databases in 7 countries. MetBroker’s original design makes it easy to add a new database, and MetBroker-linked DSS can use the new database immediately without any modification.

In addition to MetBroker, we have already developed SoilBroker, DEMBroker (Laurenson et al. 2002) and ChizuBroker, which mediate soil databases, digital elevation databases and map databases, respectively. These types of databases are heterogeneously available over the Internet and important for agricultural decision support.

Applications of Data Brokers

Several applications that utilize the data brokers have been developed, mainly using the Java applet technology (Laurenson et al. 2002). Fig 14 shows crop models for Japanese pear and paddy rice. Both obtain weather data needed to predict growth through the mediation of MetBroker. It indicates that one can simulate how Japanese pear grows in Florida for example, combining the model with the weather data of Florida mediated by MetBroker without changing the model applets at all.

Fig. 15 shows a Java applet that displays weather data on the locations of observation stations over a map of the region. This is a typical application on a Grid system that combines different types of data dynamically. That is, weather data are mediated by MetBroker and map data in a different place from weather databases are mediated by ChizuBroker. This application also uses a function of MetBroker called spatial query that retrieves weather data from multiple databases which cover the same area.

Fig. 16 shows an applet that illustrates risk of extreme meteorological events by displaying circle charts of the event probabilities (Laurenson et al. 2001). This applet has a function to predict weather conditions of non-observation points by interpolation of observed data, combining with the data mediation by DEMBroker.

International Collaboration among the Asian Countries

We have plenty of knowledge and technologies that could be commonly shared among Asian countries, considering their similarities in terms of small-scale farming and cropping systems. Therefore, international cooperation is inevitable and highly desirable especially in the Asian countries, having the similar background of agriculture and rural development. Two major international activities for agricultural information technology in the Asia-Pacific region started in 1998. One is the Asian Federation for Information Technology in Agriculture (AFITA; http://www.jsai.or.jp/afita/) and the other is the Agricultural Working Group of the Asia Pacific Advanced Network consortium (APAN/AG-WG; http://apan.net/).

Afita and Related Activities

The AFITA holds an international conference every two years and has held three conferences already in Japan, Korea and China. The countries where local societies for IT in agriculture and rural development are available are the full members of AFITA; these are China (Chinese Society for Agriculture Engineering, http://www.csae.org.cn, http://asscomp.caas.net.cn, National Engineering Research Center for Information Technology in Agriculture, http://www.nercita.org.cn/admin.asp), India (Indian Society of Agricultural Information Technology, http://www.insait.net/), Indonesia (Indonesian Society for Agricultural Information), Japan (Japanese Society of Agricultural Informatics, http://www.jsai.or.jp), Korea (Korean Society for Agricultural Information, http://www.ksais.or.kr/), Taiwan (Taiwan Agricultural Information and Technology Association), and Thailand (Thai Agricultural Information Network, http://www.thaiag.net/thaiagEng.html).

AFITA also accepts institutional participation from countries without local societies as associate members; these are ICARD of Vietnam, MARDI of Malaysia, UPLA of the Philippines, MAU of Mongolia, BAU and BOU of Bangladesh. AFITA also has participants from international organizations such as FAO, CGIAR, APAARI, etc.

The next AFITA conference (AFITA/WCCA2004, http://www.afitaandwcca2004.net/) was held in Bangkok in August 2004 jointly with WCCA (World Congress of Computer in Agriculture and Natural Resources). WCCA is a conference organized by the compound of AFITA, EFITA (European Federation of Information Technology in Agriculture, http://www.efita.net/) and PanAFITA (Pan American Federation of Information Technology in Agriculture) and held every year synchronously with an international conference held by one of the compound element. In 2004, CIGR2004 (International Commission of Agricultural Engineering, http://www.ucd.ie/cigr/) will be also held in Beijing. Section III of the conference will be fully devoted to IT in agriculture.

Apan

The APAN is an NGO consortium that provides high performance Internet links among the Asia-Pacific countries for research and educational purposes. The APAN also intermediates international research and educational collaborations that utilize the APAN infrastructure. The APAN agricultural working group (APAN/AG-WG) (http://agri-wg.jp.apan.net/) encourages use of the high performance network for agricultural research and its application.

The APAN/AG-WG has undertaken several projects. The Bio-Mirror project to promote the mirroring of bio-sequence databases in the world (http://bio-mirror.jp.apan.net/) is a particularly successful project. Database mirroring of FAO/WEICENT and the International Rice Research Institute (IRRI) located databases are also planned. APAN’s high performance network is suitable for distance learning as well. The availability of APAN’s fast network infrastructure make videoconferencing possible (Raab 1999).

Wild fire in tropical forests occurs unexpectedly and brings serious problems. Often it is quite difficult to detect it from land. The ANDES (http://www.affrc.go.jp/ANDES/) project is to detect such wild fire, utilizing satellite images acquired by NASA and NOAA in the USA. These images are sent to a super computer in MAFFIN (Ministry of Agriculture, Forestry and Fishery Research Network System in Japan) via APAN where they are analyzed to extract information regarding fires in tropical forest areas in the Southeast Asian countries. The results of the analyses are sent through the APAN network to the countries where fire has been detected.

The international sharing of meteorological databases via MetBroker is also planned under APAN. Considering the Asian farming conditions mentioned above, we should be able to share many agricultural programs but the heterogeneity in the meteorological databases blocks it. This project will provide a fundamental broker system to solve the problem, applying MetBroker as a key technology. This idea can be extended to other types of databases (soil, crop, etc.) as well. FieldServer introduced above is also becoming a subject of a new project in APAN.

Multilingual Information Exchange

In spite of the importance of data sharing, the big diversity in languages and characters in the Asian region prevents us from sharing information. We often realize that we cannot consult even dictionaries because of unfamiliar characters. To overcome this obstacle, machine translation is highly expected. The progress of these technologies is outstanding especially between similar languages. For example, Japanese-Korean translation and vice-versa are now rather practical. To apply this technology to agricultural and rural development, we have to enrich dictionaries and ontology of the specific terms in the field.

Laurenson et al. (2002) developed a menu localization system based on their broker concept (Fig. 13). This system provides a multilingual menu term server that supplies an appropriate set to menu terms of a requested language by client applications. Once a set of menu terms in a language is ready, we do not need to duplicate or recompile the program for the language at all. In other words, once an application is developed to obtain menu terms from the server, the application can be automatically localized to all the languages available at the server.

Recently, a new APAN project on multilingual information exchange was initiated (http://ml-wg.cpe.ku.ac.th) to provide a tool to share the data resource in different languages. The first step is to develop multilingual dictionary of agricultural terminology, multilingual thesaurus and ontology by cooperating with the Agrovoc project of FAO. Because fully automated translation of full text is still quite challenging, the first target is to provide a function to share tabulated information in different languages.

Conclusion

This paper is rather topic-introductory rather than a systematized presentation. It is mainly because the field of IT in the agricultural domain is quite new to be introduced systematically. We believe that, particularly in the case of agriculture, there is a great potential to benefit from IT. In fact, many people dream of agriculture empowered by IT. However, when they are asked what practical measures would empower agriculture in this way, most are at a loss for explanation. It is mainly because there is no general answer to the question. Agriculture is typically site-specific, depending on climatic and soil conditions, cropping style, market requirements, and so on. Therefore, it is the decision makers using IT who are best placed to adapt flexible technologies according to their individual situations (Laurenson and Ninomiya 2002).

References

Deerwester, S., S.T. Dumais, G.W. Furnas, T.K. Lamaier and R. Harshman. 1990. Indexing by latent semantic analysis. American Spc. Infor. Sci. 41: 391-407.

Fukatsu, T. and M. Hirafuji. 2003. Development of Field Servers for a field monitoring system, Agricultural Information Research 12: 1-12, in Japanese with an English summary.

Fulcher, C. 1999. An Internet-based integrated resource management system. IWS;99 Internet Workshop 99. APAN Pre-Proceeding: 88-92.

Kamiya, T. and T. Machida. 2002. Studies on the automatic journal system for field work by using GPS, Agricultural Information Research 11:263-272, in Japanese with an English summary.

Hoshi, T., T. Sasaki, H. Tsutsui, T. Watanabe and F. Tagawa. 2000. A daily harvest prediction model of cherry tomatoes by mining from past averaging data using topological case-based modeling, Computer and Electronics in Agriculture 29: 149-160.

Kouno, T., K. Roy, T. Machida, S. Moriizumi and S. Ninomiya. 2000. Visible agriculture by FARMWEB, an advanced Web based farming database system – In case of acquiring three dimensional (3D) images of melon fruit. Agricultural Information Research 9: 1-14.

Kouno, T., S. Ninomiya, T. Machida and S. Moriizumi. 1998. Development of farm-working journal database on WWW. Agr. Info. Res. 7: 105-117., in Japanese with an English summary.

Kouno, T., T. Machida and R. Tagami. 2002. Improvement of crop protection support system “PaDB”- Addition of cultivation calendar and distributed databases, Agricultural Information Research 1: 51-64.

Laurenson, M., U. A. Gowdara and S. Ninomiya. 2002. ResourceServer a localization system for distributed computing applications, APAN/Shanghai 2002, http://agri-wg.jp.apan.net/2002Shanghai/Index.html.

Larenson, M. and S. Ninomiya. 2002. DEMBroker – Consistent access for software applications to digital elevation models, Proc. AFITA2002.

Laurenson, M., T. Kiura and S. Ninomiya. 2000. Accessing on-line weather databases from Java, Internet Workshop 2000, Tsukuba, Japan: 193-198.

Laurenson, M., T. Kiura and Ninomiya, S.. 2001. A Tool for estimating the risk of extreme climatic events. Agricultural Information Research 10: 79-90.

Laurenson, M., T. Kiura and S. Ninomiya. 2002. Providing agricultural models with mediated access to heterogeneous weather databases. Applied Engineering Agric. 18: 617-625.

Laurenson, M and S. Ninomiya. 2002. Successful Agricultural Decision Support Systems Agricultural Information Research 11: 5-25.

Laurenson, M., A. Otuka and S. Ninomiya. 2002. Developing agricultural models using MetBroker mediation software. J. Agric. Meteorol. 58: 1-9.

  • Lee, B.L. and Y.C. Kim. 1998. Web interface for GIS in agriculture, Agricultural Information Technology in Asia and Oceania 1998. Proc. 1st AFITA conf.: 107-111.
  • Matsumoto K. and T. Machida. 2002. Studies on development of work journal supporting systems with speech recognition, Agricultural Information Research 11: 255-262.
  • Ninomiya, S.. 2001. The present status and perspective of research projects for agricultural information technologies in MAFF. J. Jpn. Soc. Agric. Machine. 63,4: 3-11, in Japanese.
  • Ninomiya, S., T. Kiura, K. Eto, T. Nanseki and M. Ueda. 1997. Field Image acquisition system on WWW. J. SHITA 9: 12-19, in Japanese with an English summary.
  • Otuka, A. and S. Ninomiya. 1998. Conceptual Retrieval of Agricultural Cases with Latent Semantic Indexing Approach, Proceeding of International Conference on Engineering of Decision Support Systems in Bio-Industries, Montpellier, France: 37-38.
  • Otuka, A. and C. Kitamura. 2002. Agriculture case base; Decision support by harnessing knowledge and experience, Bul. Natl. Agr. Res. Centr. 1: 45-56.
  • Otuka, A. and T. Yamakawa. 2003. An application for insect field surveys using a hand-held computer, Agricultural Information Research 12: 113-124.
  • Otuka A. and K. Sugawara. 2003. A labor management application using hand-held computers, Agricultural Information Research 12: 95-104.
  • Raab, R.T.. 1999. Realizing APAN’s potential to facilitate distance learning applications: A CG perspective on what needs to be done. APAN Osaka Meeting Report. http://agri-wg.jp.apan.net/Osaka/index.htm.
  • Sasaki, S.,T. Kiura and T. Furuya. 2002. Development of rain fall database for cellular phone, Agricultural Information Research 11: 231-238, in Japanese with an English summary.
  • Sugawara, K.. 2001. Farming diary system using internet-enabled cellular phones, Internet Workshop 2001 Proc.II Applications 247-252.
  • Tanaka, K., M. R. Laurenson, K. Sugahara, T. Ohtani, T. Watanabe and S. Umemoto. 2001. Web based IPM system for Japanese pear diseases in Japan. II. Forecasting models for scab infection associated with weather retrieval system. Proc. KSPP International Conference on Plant Disease Forecast: Information Technology in Plant Pathology.
  • Tanaka, K., T. Watanabe and M. Hirafuji. 2000. Model Design using the Interface and use of models by the modelbase system, Proc. AFITA2000, 267-272.
  • Xin, J. H. Beck, L. Halsey, J. Fletcher and F. Zazueta. 1998. Using digital cameras and the Internet to identify plant insect and disease problems. Computer in Agriculture 1998: 352-358.

Index of Images

  • Figure 1 A Web-Based Camera Server System, Field Eye: User Interface for the Remote Camera System (Left and Middle), User Interface for the Automatically Acquired Image Database System (Right).Figure 1 A Web-Based Camera Server System, Field Eye: User Interface for the Remote Camera System (Left and Middle), User Interface for the Automatically Acquired Image Database System (Right).

  • Figure 2 Mobile Phone-Based Web Applications; Weather Database Access (Left) and Farm Management Diary (Right).Figure 2 Mobile Phone-Based Web Applications; Weather Database Access (Left) and Farm Management Diary (Right).

  • Figure 3 Fieldserver, Wireless Lan Autonomous Field Monitoring System.Figure 3 Fieldserver, Wireless Lan Autonomous Field Monitoring System.

  • Figure 4 Coverage of a Region by Fieldservers.Figure 4 Coverage of a Region by Fieldservers.

  • Figure 5 A Web-Based Prototype of Case Base with Conceptual Search Engine. <BR>Figure 5 A Web-Based Prototype of Case Base with Conceptual Search Engine. <BR>

  • Figure 6 Plant Condition Diagnosis System.Figure 6 Plant Condition Diagnosis System.

  • Figure 7 Navigation Search: Users Do Not Need to Use Keyboards.Figure 7 Navigation Search: Users Do Not Need to Use Keyboards.

  • Figure 8 Web Marketing with Remote CameraFigure 8 Web Marketing with Remote Camera

  • Figure 9 An Agricultural Product Identification System. Id and a Url Are Printed on Product Packages. a Consumer Can Access the Producer's Home Page, Using the Url and Id to Obtain the Production History and Producer's Profile.Figure 9 An Agricultural Product Identification System. Id and a Url Are Printed on Product Packages. a Consumer Can Access the Producer’s Home Page, Using the Url and Id to Obtain the Production History and Producer’s Profile.

  • Figure 10 Schematic Diagram of an Agricultural Grid System.Figure 10 Schematic Diagram of an Agricultural Grid System.

  • Figure 11 If a Weather Data Mediation Program Is Not Available, the Number of Code Modules Is Proportional to N X M.Figure 11 If a Weather Data Mediation Program Is Not Available, the Number of Code Modules Is Proportional to N X M.

  • Figure 12 A Weather Data Mediation Program, Metbroker. It Decreases the Number of Code Modules Proportional to N + M.Figure 12 A Weather Data Mediation Program, Metbroker. It Decreases the Number of Code Modules Proportional to N + M.

  • Figure 13 A Simple Java Applet to Obtain Weather Data Mediated by Metbroker. Users Can Obtain Weather Data from Several Heterogeneous Weather Databases by Using This Application through Setting Query Options. Automatic Menu Localizer (See Text) Is Available on This Application (Right, English, Left; Korean).Figure 13 A Simple Java Applet to Obtain Weather Data Mediated by Metbroker. Users Can Obtain Weather Data from Several Heterogeneous Weather Databases by Using This Application through Setting Query Options. Automatic Menu Localizer (See Text) Is Available on This Application (Right, English, Left; Korean).

  • Figure 14 Crop Models As Metbroker Applications Such As a Japanese Pear Growth and Disease Prediction and a Rice Growth Prediction.Figure 14 Crop Models As Metbroker Applications Such As a Japanese Pear Growth and Disease Prediction and a Rice Growth Prediction.

  • Figure 15 Metbroker Application Which Shows Spatial Weather Data from a Region Over the Map. the Same Application Can Display Weather of Completely Different Databases. the Left Shows the Data for the Wakayama Prefecture from a Prefectural Local Database and Amedas for the Region, and the Right Shows the Data for the Korean Peninsula from the Seoul National University Weather Database and the Amedas DB of Japan. the Map Data Is Dynamically Downloaded from a Map Server in the Us, Using Chizubroker.Figure 15 Metbroker Application Which Shows Spatial Weather Data from a Region Over the Map. the Same Application Can Display Weather of Completely Different Databases. the Left Shows the Data for the Wakayama Prefecture from a Prefectural Local Database and Amedas for the Region, and the Right Shows the Data for the Korean Peninsula from the Seoul National University Weather Database and the Amedas DB of Japan. the Map Data Is Dynamically Downloaded from a Map Server in the Us, Using Chizubroker.

  • Figure 16 Metbroker Applet That Displays Extreme Weather Risk Using Circle Charts Over the Location of Observation Points. It Estimates the Conditions of Non-Observed Points As Well, Using Mesh-Interpolation. Metbroker, Dembroker and Chizubroker Are Jointly Used to Mediate Three Different Types of Databases.Figure 16 Metbroker Applet That Displays Extreme Weather Risk Using Circle Charts Over the Location of Observation Points. It Estimates the Conditions of Non-Observed Points As Well, Using Mesh-Interpolation. Metbroker, Dembroker and Chizubroker Are Jointly Used to Mediate Three Different Types of Databases.

    Ref.http://www.agnet.org/library/eb/549/

Leave a Reply