Thailand is known worldwide as one of the major producers of agricultural products. Agricultural activities provide developing countries with food and revenue. In 2021, the Thai government introduced the Bio-Circular-Green Economic or BCG Model to move Thailand’s economic development forward and to lead the country out of the middle-income trap and create sufficient income for people, especially for those in the agricultural sector. The BCG model, conformed with the UN Sustainable Development Goals (SDGs) and integrated with the Sufficiency Economy Philosophy (SEP), uses Thailand’s strength in aspect of biological diversity and cultural richness, and employs technology and innovation to transform Thailand to a value-based and innovation-driven economy as well as to create economy and growth after the COVID-19 pandemic. To drive the BCG strategy, agricultural mechanization dealt with a wide variety of advanced technologies is of necessity as a supporting tool. In this review, the focus is to discuss the smart agricultural mechanization, which is related to precision agriculture technology, as well as advanced technology such as ICT, sensor, variable-rate technology, and AI.
Keywords: smart farming, mechanization, sustainable agricultural system, Bio-Circular-Green (BCG) model
Agricultural production in Thailand
Thailand is one of the major suppliers of agricultural products in the world market. Approximately 47% (23,878,920 ha) of the country’s total area was used for agricultural production in 2021 (OAE, 2023). About 44% of agricultural land use in Thailand was devoted to rice production, followed by: upland field crop (21%) such as para rubber, cassava, sugarcane, maize, and oil palm; fruit trees and perennial trees (26%); vegetables, cut flowers, and ornamental plants (1%); and other crops (9%).
Thailand’s agricultural production received a revenue of US$40,386 million in 2021, accounting for 8.71% of the nation’s gross domestic product (GDP), with 7.84%, 0.79%, and 0.08% of the crop and animal production, fishery and aquaculture, and forestry and logging sectors, respectively (NESDC, 2023). Considering the exporting and the importing values, it was found that Thailand’s agricultural product has surplus trade balance. The value of export in agricultural sector was more than the importing value around US$31,486 million in 2022. Meanwhile, non-agricultural product has deficit trade balance about US$51,642 million (OAE, 2023).
Natural rubber is the most exported value, followed by fruits and products, rice and products, cassava and products, chicken meat and products, sugar and products, and fish and products. These agricultural products have the export value more than US$3,400 million for each product in 2022 (OAE, 2023).
Thailand is divided into four main regions, including Central, Northern, Northeastern, and Southern regions. Rice is the most important crop grown throughout Thailand. Wet rice is cultivated in irrigated fields commonly found in every part of the Central Plain, while dry rice is grown predominantly in the North and Northeast. In general, weather conditions permit farmers to produce two crops a year. Where irrigation is available, rice is produced three crops a year. Some areas in the Central part utilized underground water, as a result farmers can accomplish five crops for rice production in two years. The main growing areas for cassava are Chon Buri and Rayong provinces in Eastern area of Thailand. However, substantial quantities were also produced in Northeastern area such as in Nakorn Ratchasima and Ubon Ratchatanee Provinces, and so on. The principal sugarcane cultivation areas are in and around Kanchanaburi Province in Western Bangkok. Also, sugarcane is grown in non-irrigated areas in the Northeast and in the North.
Agricultural mechanization in Thailand
Mechanization is a crucial input for agricultural production. It can increase land productivity by facilitating timeliness and quality of cultivation. More farm operations can be completed at the right time and more areas can be cultivated to produce more quantities of crops. Mechanization can also support opportunities that could cope with the problem of labor shortage. With the substantial difference in earnings between urban and rural areas, rural labor migration to urban areas accelerated. As a result, the number of farm households registered on the Thai government database r has been declining by 5.40% from 2019 to 2022 (NSO, 2023). The number of labors supply in the agricultural sector was approximately 20.55 and 20.53 million farmers in 2019 and 2022, respectively. Meanwhile, the numbers of non-agricultural labor forces in 2019 and 2022 were 17.63 and 19.37 million workers, respectively (OAE, 2023).
Overview of Agricultural Mechanization in Thailand
A substantial contribution to agriculture in the last century has been the escalation from manual and stock-animal farm work to powered farm equipment. Historically, Thai farmers used simple tools, animal-drawn implements, and water wheels. The growth of agricultural mechanization in Thailand has been reviewed (Thepent, 2015) and could be briefly summarized as follows. In 1891, mechanization with power technology started in Thailand. The early growth of agricultural mechanization was linked with imported machineries for trial in 1920, a single axle tractor in 1947, the Japanese 2-wheeled tractors or power tillers during 1955-1957, as well as established the Agricultural Engineering Division (AED) under the Ministry of Agriculture and Co-operative, and then released the design of an axial flow pump, namely “Debaridhi water pump” for local manufacture. The second phase was relevant that the Division released the design of a 4-wheeled tractor powered by a 25-horsepower engine, named “Iron Buffalo”, and the first prototype of a rice combine harvester was designed in 1958. A firm producing 2-wheeled tractors in Ayudthaya province began manufacturing a simple 4-wheeled tractor in 1967-1969. The AED constructed the prototype for an axial flow rice thresher, via its blueprint supported from the International Rice Research Institute (IRRI) in 1975. During 1978 to 1982, a rice transplanter (12 rows and power operated) was imported from China by a local firm that also produced it. The third phase of farm mechanization in Thailand was that local firms around Bangkok started to fabricate a Thai-made rice combined harvester in 1985-1987. Around the early 1990s, these firms successfully developed Thai-made rice combined harvesters. It was accepted for use by farmers and popularly used in hiring services, especially in the Central plain and then its use spread in other regions of the country. The fourth phase of agricultural mechanization during 1995 to 2015 was that steel plows, seed drills, mowers, mechanical reapers, and threshers contributed to the development of modernized agriculture. Tractors enable the farmers to perform land preparation, sow and harvest on large agricultural lands with less manpower.
Recently, powered machinery such as tractors, has replaced many jobs formerly carried out by men or animals. As of now, there are two methods of practicing farming apparatus use in Thailand: as a proprietor and/or through custom hiring. Most farmers are looking forward to appropriate and high efficiency harvesters for rice and sugarcane harvesting due to labor shortage during harvesting season, especially in the central plain region and the lower part of the northern region.
Current Situation of Mechanization and Precision Agriculture in Thailand
Recently developed technologies emphasized agricultural machinery to minimize operation time, input, and energy consumption as well as maximize productivity in agriculture via the concept of precision agriculture. The application of precision agriculture in Thailand is briefly discussed as follows.
Precision agriculture originated in 1990s (National Research Council, 1997), when global positioning system (GPS), yield monitoring, geographical information system (GIS) and other software started to use to collect field data. These technologies enable the possibility to locate areas in large fields that vary in features, farm input, plant health, and yield. The site-specific farming techniques considered with such information are able to practice in planning and operation management of agricultural production via agronomic knowledge and practice integrated with new mechanical technologies. Precision agriculture approaches to the variability of natural resources, such as soil condition and landscape variability; differences in management techniques, including chemical recommendations for crop protection and crop quality; and the availability of engineering technology such as farm machinery operated with positioning system, and variable rate application (VRA) equipment (Ahamed et al., 2016).
In Thailand, some farm practices and technologies were developed during 1995 to 2005 to efficiently use the agricultural inputs according to spatial variability of plot following the concept of precision farming. For example, precision fertilization was launched by the Department of Soil Science, Faculty of Agriculture, Kasetsart University since 2004 (Attanandana et al, 2004). This tailor-made fertilization that is dependent on soil series and soil’s nutrients N-P-K used lesser amount of fertilizers when compared with traditional fertilization. It could reduce production cost and increase 10-15% of the yield (Attanandana, 2005). Also, Sriswasdi et al., (2008) developed a mobile app to provide fertilizer recommendation according to soil condition of farmer’s plot. By implementing this app to the rice farmers in aggregated area of 160,000 ha, the app’s recommendation could help farmers to save their fertilizer cost and increase their yield. In addition, some automatic fertilizer applicators based on soil analysis result were applied in some major crops such as for sugarcane farming (Saicomfu & Jansuwan, 2019). In addition, some tailor-made fertilizing machines were designed to use in agronomy and horticulture crops (DOA, 2022) and so on. Accordingly, consumption of chemical fertilizers, one major imported input of agricultural production in Thailand, was decreased. The imported values of chemical fertilizers have decreased by 33.85% from 2018 to 2022. It was around US$686 million in 2022.
Balance trade of Thai agricultural machinery industry
The agricultural machinery industry in Thailand consists of only a few large companies, all of which are foreign-owned. Local manufactures are small, therefore not being able to produce machines with high quality (Embassy of India, 2015). Most of the agricultural equipment used in Thailand is locally produced such as tractors, power tillers, disc ploughs, disk harrows, water pumps, sprayers, threshing machine, reapers, combine harvesters, cleaning equipment, dryers, rice milling machines, equipment for specific processing, and so on.
The Iron and Steel Institute of Thailand (2020) reported that in 2020 some agricultural machines were imported from overseas such as China (35.3%), Japan (24.4%), India (5.3%), and US (4.9%). Over the past six years, amount of importing value of agricultural machinery was increased around 23.53% from 2017 to 2022 (Table 1). Meanwhile, Thailand’s exports of agricultural machinery were growing around 17.48% from 2017 to 2022. In 2022, the export amounted to US$1,168.64 million while the import was US$1,357.77, resulting in a trade deficit of US$189.13 million. Thailand's status as a regional export hub allows local machinery and parts firm supplied to Thailand’s neighbors in ASEAN countries. The trend of export growth is likely to continue as demand from those countries.
PROSPECTS OF SMART AGRICULTURAL MECHANIZATION IN THAILAND
As part of the Thailand 4.0 strategy, the Thai government supports the application of technical innovation in major economic sectors, including the agriculture sector. Starting from 2017, the target of agricultural production system was also 4.0, the Thai agriculture sector has rapid growth in digital platforms and applications, developed by both the public and private sectors (Ministry of Industry, 2017). It is essential to modernize agriculture through the mechanization of its operations. Use of agricultural machinery has an extraordinary potential for poverty alleviation by increasing land and labor productivity in Thailand for sustainable development. Innovative policy such as the Bio-Circular-Green Economic Model was introduced by the Thai government (NSTDA, 2021). This model coupling with the advanced technologies including ICT, sensors technology, variable-rate technology, yield monitoring and mapping technology are the key drivers of smart agricultural mechanization in Thailand. The following section briefly discusses the policy and technology for the prospect of smart agricultural mechanization in Thailand in the post COVID-19 pandemic.
Bio-Circular-Green Economic (BCG) model
In 2021, the Bio-Circular-Green Economic Model or BCG was introduced and promoted by the Thai government as a new economic model for inclusive and sustainable growth. The BCG model is as the Government’s key policy to lead the country out of the middle-income trap and create sufficient income for people, especially for those in the agricultural sector. The model uses Thailand’s strength in the aspect of biological diversity and cultural richness, and employs technology and innovation to transform Thailand to a value-based and innovation-driven economy as well as creating economy and growth after the COVID-19 pandemic. The model conforms with the UN Sustainable Development Goals (SDGs) and is integrated with the Sufficiency Economy Philosophy (SEP) which is also the key principle of Thailand’s social and economic development. The BCG model has been applied to focus on promoting four industries comprised of agriculture and food; medical and wellness; bioenergy, biomaterial and biochemical; and tourism and creative economy (NSTDA, 2021).
The Thai government is still moving Thailand’s economic development forward with the BCG Model. Science, technology and innovation are employed to enhance the capacity and competitiveness of each stakeholder in the value chain, both upstream and downstream, in all four industries mentioned above. In the food and agriculture sector, two main goals are involved. The first goal is the improvement of resource- and land-use efficiency, the reduction of production loss, and waste utilization. The second goal is to elevate from low-value commodities to value-added products, diversified products, differentiated products, and premium products, as well as the development of high-value functional ingredients and novel food products for groups of people such as patients, elder, and so on. In addition, converting biomass and agricultural by-products to high-value commodities such as bioplastics, biomedicine and so on, is launched as one goal of the BCG model in the bioenergy, biomaterial, and biochemical sector.
To drive the BCG strategy, the agricultural mechanization is of necessity to support agricultural technology, precision agriculture technology, smart farming technology, postharvest technology, food and bio-product production and safety technology, waste management technology, as well as advance technology such as ICT, sensor, and AI technology to achieve the aforementioned goals. Agricultural machinery was formerly only aimed towards the mechanization of agricultural operations. However, integration with those technologies navigates the agricultural mechanization technology for the broader targets. Not only complete farm operation at the right time by facilitating timeliness and quality of cultivation to increase land productivity, but produce abundant food, bio-products, and bioenergy without depleting the natural resources or polluting its environment and also to emphasize the agricultural mechanization technology. To address these issues, therefore, nowadays smart agricultural mechanization deals with a wide variety of related subjects which are presented briefly as follows.
Information and Communication Technology (ICT) plays as one key role to support production, facilitate management, and enhance productivity of agriculture and agricultural industry. ICT helps farmers to gain information and knowledge to improve their cultivation and create chance for their markets. Vital information related to sowing, crop protection, and improving soil fertility enables farmers to stay up to date on the latest crop cultivation practices and then improve their productivity. ICT can provide convenient access and enables farmers to acquire weather-related recommendation, additional information and services that enable informed decision-making as well as alerts them to prepare for irregular events such as floods, drought, or even pest and disease outbreaks, thus preventing significant crop loss.
Increasing in mobile technology and web-enabled smartphones, lead to increasing of internet use in Thailand. As in 2022, there were around 57 million internet users in Thailand (Statista, 2023). Increasing use of digital platforms and development of web application and mobile app for a variety of agricultural purposes at different points of the value chain are playing vital roles in reshaping Thai agriculture. These digital tools enable extensive linkages between agricultural sector, food industry, consumer, and stakeholders. For instance, Thai farmers can make direct purchases of their production inputs at a lower price, share and let hire their farming resources, as well as sell their products directly to consumers to obtain increased profits. These opportunities are conveniently accomplished via the social media app such as Facebook which around 51.87 million internet users in Thailand were using (Statista, 2023), also the third-party delivery apps, and other platform-to-consumer services.
As sensors have advanced, ICT can now connect with agricultural production system in a variety of ways using the concepts with clouds. ICT enables all stakeholders to detect environmental impact to boost production via analytics of data obtained from sensor. Moreover, ICT involves farm machineries in areas of machine optimization. The remote access to on-board machinery optimization is now required for variable application of input such as planting, fertilizing, weeding, and pesticides (Ahamed et al., 2016). A summary of sensors for smart farm mechanization is discussed briefly as following section.
Sensors for smart agricultural mechanization
An overview of sensors for intelligent system employed in farm machinery includes sensors for positioning and image sensors.
· Sensors for positioning
The global positioning system (GPS) is a satellite-based navigation system. It consists of a network of 24 satellites currently orbiting in space at a distance of 11,000 miles from the Earth (Farrel and Barth, 1999). When four or more satellites are detected by the GPS receiver, the receiver can triangulate its location from the known positions of the satellites. GPS can be used for precision farming, yield mapping, field planning, and agricultural vehicle guidance. It enables gathering of significant input/production data with positioning information, and then leading to efficient manipulation and analysis of collected data to achieve soil/plant strategies which can enhance production.
Satellite-based GPS sensor is widely used in Thailand. Its application in agriculture includes field mapping and yield mapping, which widespread used in sugarcane and sugar industry. The GPS is installed both on the mechanical sugarcane harvester and the truck operated with the harvester in order to check and record sugarcane yield for each plot and make use of such mapping for planning about sugarcane cutting order, scheduling of workers and machines, and so on.
Meanwhile, some farming operations require high precision for positioning, such as row crop bed preparation, planting, fertilizing and so on. In these operations, GPS accuracy is not sufficient. For this reason, the Real Time Kinetic (RTK) GPS has been proposed. The RTK-GPS receiver takes in the normal signals from the Global Navigation Satellite Systems (GNSS), included satellites from GPS (USA), GLONASS (Russia), Beidou (China), and Galileo (Europe), along with a correction stream to achieve one centimeter positional accuracy in real time. The RTK-GPS requires a base station located at a known surveyed point to transmit correction via radio frequency to one or more mobile receivers within a ten-kilometer range of the base station. This enables using of high accuracy application in agriculture, for instance, precise guidance system of vehicle along crop rows. In Thailand, the RTK-GPS is moderately growing. It was widely used in surveying and land reforming tasks. Its utilization in aspect of variable rate application for fertilizing in sugarcane farming (Saicomfu & Jansuwan, 2019) and horticulture crops (DOA, 2022) was also found. As well, correction of GPS elevation data obtained from the RTK-GPS was deployed for land levelling on rice plot, and topography mapping of land.
To use GPS and RTK-GPS, especially for navigating the short travelling of agricultural vehicle, it must be aware of their limitations related to time synchronization between the satellite and the receiver, precise real-time location of satellite, degradation of signals when the number of satellites from which the signals received is less than four number, electromagnetic noise and periodic signal blockage by trees or buildings. In these situations, the reliable sensors such as dead-reckoning sensors and inertial sensors should be substituted (Ahamed et al., 2016).
· Image sensors
In the agricultural sector, the parameters like canopy, yield, surface color and quality of product are the important measures from the farmers’ and also agricultural scientists’ points of view. However, manual visual inspection is labor intensive, expensive, and prone to human error, leading to variation in result obtained. Image processing can help to get result well within short time and at affordable cost, since it is one effective tool which can be applied to measure those parameters with accuracy and economy. Applications of image processing in agriculture is growing vastly and broadly through the image processing sources of radiation such as Gamma ray imaging, X-ray imaging, imaging in UV band, imaging in visible band, imaging in IR band, imaging in Microwave band and imaging in Radio band (Gonzalez & Woods, 2008).
Image sensor can be used as the effective tool to acquire information about an object or any phenomenon without physical contact with the object. It is a valuable implementation of remote and proximal sensing. Remote sensing is widely attached to the use of satellite, airborne or UAV platforms using multi- or hyper-spectral imagery. In terms of proximal sensing, the sensor is close to the object and is installed on various platforms such as handheld, fixed installations, as well as attached on robotics and tractor. The types of sensors range from simple RGB camera to multispectral and hyper-spectral high resolution imaging systems or even thermographic camera.
In Thailand, many applications of image sensors have been found in agriculture in three main aspects comprised of: 1) in crop monitoring and productivity assessment; 2) in fruit and vegetable grading; and 3) in weed detection.
1) Application in crop monitoring and productivity assessment:
Remote Sensing (RS), involves identification of earth surface features and estimation of geo-biophysical properties using electromagnetic radiation. The RS data generally obtain from images of the satellites and unmanned aerial vehicle (UAV). Imaging techniques with different spectrum such as infrared, hyper spectral imaging, and X-ray imaging were useful in determining the vegetation indices, canopy measurement, irrigated land mapping, etc. The result obtained from spectral data together with multi-source data in spatial and temporal variations can be used to monitor and assess changed crop intensity and then crop productivity.
2) Application in fruit and vegetable grading:
In case of fruit and vegetable grading, accurate sorting and grading processes are needed to meet consumers’ expectations under quality and safety standard of food. Then image processing has been applied to check quality attributes of the product. It is a nondestructive, accurate and reliable method to gain target of sorting, grading of fresh products, detecting of defects appeared on the surface such as dark spots, cracks and bruises on fresh fruits, as well as, identifying the maturity of products via surface color determination.
3) Application in weed detection:
Weeds grow on farms which compete with crops for water, light, nutrients and space, causing reduction in crop yield. Weed control is important, but research related to weed detection by using image processing technique is growing slightly in Thailand. There were some applications of weed detection based on image processing. For instance, factors affecting the discrimination of weeds from Chinese Kale seed were assessed via image processing technique (Dathamart et al., 2019). Data obtained from feature extraction techniques along with AI classifier algorithms were applied for weed identifying and checking weed specie. Data acquired from image sensors installed on UAV and AI classifier algorithms were also conducted in Thailand for segmenting of soil, crop, and weed, and then detecting and estimating weed coverage.
Variable-rate application (VRA) is one management approach for addressing the spatial variability within a field. It is an important component for site specific crop management. The equipment used to perform VRA is commonly called variable-rate technology (VRT), which is the new technology for determining non-uniform cropping input, such as seed, fertilizer, and pesticide. There are two types of VRA: Map-based VRA and Sensor-based VRA to adopt the variable-rate application with farm equipment.
· Map-based VRA
In the map-based VRA system, a real-time sensor is not required. The variable-rate application is based on the spatial information contained in an electronic map of the field. The system requires a positioning system such as RTK-GPS to determine the position of the equipment in the field and use this location to determine the appropriate application rate. The amount of total application can be determined in advance prior to operating on the field. Ground speed controller and actuators are also required for the map-based VRA system. The requirements for a high-accuracy GPS to determine location and GIS for data processing make the system expensive.
· Sensor-based VRA
The sensor-based VRA system uses data from real-time sensors such as image sensor, etc. These real-time sensors work in field to measure soil property, crop property, and light reflectance of crop and weeds. The VRA control system uses sensor data instead of application rate map and then electronically and automatically control site-specific field operations. The sensor-based VRA system takes measurement continuously. The system has lower cost because it does not use GPS.
Controller for the sensor-based VRA system uses microprocessor to read the sensor input and calculate the output rate by using the stored algorithm. Actuator responds to the output signal from the controller, and then regulate the amount of material applied to the soil or crop or weed, etc. on the field.
The sensor-based variable rate applicator can be grouped by the type of product that is applied to the filed such as seeds, dry chemicals (granular fertilizer, lime), and liquid chemicals (liquid fertilizer, liquid pesticide). These three major types are applied in spatially variable or variable-rate operation using the applicator. The smart sprayer is one type of real-time applicator for site-specific weed management developed at the University of Illinois at Urbana-Champaign (Tien et al., 1999). In Thailand, real time weed recognition system for identifying and locating weed grown in sugarcane field has been discussed to design and develop. This recognition system will be used as machine vision system attached on smart sprayer in the future.
Yield monitoring and mapping
Yield monitoring and mapping are considered as accept change of precision farming. This can be used for making future management decisions both on a single field plot and large area. A yield monitoring system incorporated with GPS sensor, collects data on the crop yield and farm performance for a given year. The system measures and records information including harvested grain flow, grain moisture, and location. In Thailand, yield monitoring and mapping are becoming important practices. For instance, on-the-go yield sensor attached to the paddy rice combine harvester to check the amount of the harvested paddy rice, as well as the information system to gather such data were designed and developed in 2018 (DOAE, 2019).
Several sensing methods have been utilized to measure the mass flow through the mechanical harvester. These sensors include mass flow, velocity flow, and volumetric flow sensors. These sensing technologies such as impact sensing, radiation or gamma source sensing, volumetric-flow sensing, photoelectric sensor, paddle wheel, moisture content sensor, a differential global positioning system (DGPS) and so on, will be highly taken into consideration in the near future for Thailand.
Smart agricultural mechanization is currently considered as a necessity in Thailand to support agricultural and food production, as well as bio-products and bioenergy productions, without depleting the natural resources or polluting the environment according to the concept of the BCG model to transform Thailand to a value-based and innovation-driven economy as well as creating economy and growth after the COVID-19 pandemic. Smart agricultural mechanization is not based on a single technology for the improvement of a single practice. Instead, it is practiced and utilizes several technologies for the application of several management techniques. It integrates with agricultural knowledge and new technologies such as ICT, sensors, variable-rate technology, yield monitoring and mapping, and so on. In the future, these affordable technologies will be highly taken into consideration in designing and research in an attempt to develop the smart agricultural mechanization that fits Thai agriculture.
Agricultural Engineering Research Institute (AERI). (2022). Research of agricultural engineering research institute. Department of Agriculture, Ministry of Agriculture and Co-operative, Bangkok, Thailand. Retrieved from https://www.doa.go.th/aeri/?page_id=5248.
Ahamed, T., Noguchi, R., Takigawa, T., & Tian, L. (2016). Bioproduction engineering: Automation and precision agronomics for sustainable agricultural systems. Nova Publishers, New York, USA.
Ahmed, I., Adnan, A., Gul, S., & Islam, M. (2008). Edge based real time weed recognition system for selective herbicides. Paper presented at the Proceedings of International Multiconference of Engineers and Computer Scientists 2008 Vol.1, Hong Kong.
Attanandana, T. (2005). SimRice program. Department of Soil Science, Faculty of Agriculture, Kasetsart University, Thailand. Retrieved from http://www.ssnm.agr.ku.ac.th/main/Download.htm
Attanandana, T., Verapattananirund, P., & Yost, R. (2004). Capacity building of the farmers to improve soil resources and economic conditions in Thailand. Department of Soil Science, Faculty of Agriculture, Kasetsart University, Thailand. Retrieved from https://kukr.lib.ku.ac.th/db/kukr/search_detail/download_digital_file/192693/23520
Dathamart, C., Kaewtrakulpong, K., Sermsak, R., Jedsadathumsathit, S., Phaosang, T., & Fuprasert, Y. (2019). Factors affecting the discrimination of weed from Chinese kale seed by image processing technique. Khon Kaen Agriculture Journal, 47(6): 1113-1118. doi:10.14456/kaj.2019.101
Department of Agricultural Economics (MOAC). (2018). Loss reduction in paddy rice harvesting process: A case study of the rice harvest in the promoted area of large scale farm. A Final Report, the Agricultural Research Development Agency (ARDA), Thailand.
Embassy of India. (2015). Thai market for agricultural machinery. Embassy of India, Bangkok, Thailand
Farrel, J. A., & Barth, M. (1999). The global positioning system and inertial navigation, McGraw-Hill, New York.
Gonzalez, R. C. & Woods, R.E. (2008). Digital image processing, 3rd Edition. Pearson Prentice All, New Jersey, US, 793p.
Iron and Steel Institute of Thailand. (2020). Thailand machinery outlook: March, 2020. Office of Industrial Economics, Ministry of Industry, Thailand
Iron and Steel Institute of Thailand. (2023). Thailand machinery outlook: March, 2023. Office of Industrial Economics, Ministry of Industry, Thailand
Ministry of Industry. (2017). The achievement in implementation of the government policy and industrial strategy: Fiscal year 2016. Ministry of Industry, Thailand.
National Research Council. (1997). Precision agriculture in the 21 century: Geospatial and information technologies in crop management. National Academy Press, Washington D.C.
National Science and Technology Development Agency (NSTDA). (2021). Bio-Circular-Green economy: Action plan 2021-2027. National Science and Technology Development Agency, Ministry of Higher Education, Science, Research, and Innovation, Thailand.
National Statistical Office (NSO). (2023). Statistics in agriculture and fishery. National Statistical Office, Ministry of Information and Communication Technology, Thailand. Retrieved from http://statbbi.nso.go.th/staticreport/page/sector/th/11.aspx
Office of Agricultural Economics (OAE). (2023). Agricultural statistics of Thailand 2022. Office of Agricultural Economics, Ministry of Agriculture and Co-operative, Bangkok, Thailand.
Office of the National Economic and Social Development Council (NESDC). (2023). National accounts of Thailand 2021. Office of the National Economic and Social Development Council, Office of the Prime Minister, Bangkok, Thailand.
Saicumfu, A., & Jansuwan, C. (2019). A study and development of an automatic fertilizer applicator based on soil analysis result. Paper presented at the Proceeding of the 57th Kasetsart University Annual Conference, Bangkok, Thailand.
Sriswasdi, W., Luengsrisagoon, S., Lorsuwansiri, N., Wuttilerdcharoenwong, S., Khunthong, V., Suksaengsri, T., Kawtrakul, A., Seebungkerd, N., Tananon, U., Narkwiboonwong, W., & Pusittigul, A. (2008). A smart mobilized fertilizing expert system: 1-2-3 Personalized Fertilizer. Paper presented at the World Conference on Agricultural Information and IT, IAALD AFITA WCCA 2008, Tokyo University of Agriculture, Japan.
Statista. (2023). Number of Thailand facebook users. Retrieved from https://www.statista.com/statistics/490467/number-of-thailand-facebook-users/
Thepent, V. (2015). Agricultural mechanization in Thailand. Policy Brief: issue No.6 (2015), Center for Sustainable Agricultural Mechanization, United Nations, Beijing, P.R.China.