INTRODUCTION:

The millennium development goals (MDGs) of diminishing hunger and food security are rooted in increasing agricultural productivity, especially from crop farming. The reason has been that agriculture is seen as the engine of economic growth and development in many low and medium-income economies, particularly in sub-Saharan Africa (SSA)¹. On the other hand, to achieve this goal, especially under changing climatic conditions, is posing new challenges and opportunities to small scale farmers worldwide ². Shortage in water is reported to affect a part of the globe, and the condition is deteriorating worse over time due to the growing world population and freshwater demands³. In the agriculture sector, mainly irrigation consumes a significant portion of the freshwater. Also, the lack of cost-effective smart irrigation systems, developing countries are using more water in contrast to the developed countries for achieving the same crop yield.

Besides, changing climatic conditions pose additional risks from both increasingly irregular rainfall patterns and rising temperatures. Literature shows that droughts and heavy rains are very likely to increase in frequency in large parts of the world. Hence, the need for extended scope of and improved access to state-of-the-art and just-in-time weather forecasts also for smallholder farmers, especially in the developing world. In a study⁴, it was shown that climate change could be expected to have a substantial impact on crop cultivation in Africa. The paper further highlighted that regional differences in these impacts require regional adaptations strategies such as improved nutrient management strategies or shifts in cropping calendars. Climate change-related adaptation and mitigation measures do also offer new possibilities to intensify agricultural production, particularly in temperature-limited environments such as high altitudes or seasonally hot regions⁵.

Therefore, there is a dire need to come up with advanced technologies based on smart strategies and decision support systems for the effective utilisation of rainwater. One of the possibilities to improve farmers’ decision-making in such challenging environments is the use of online or application (app)-based tools, either directly by farmers or through the help of extension services². These apps assist farmers in many aspects ranging from weather forecasts, smart irrigation and cropping calendars to more complex server-based crop growth models. According to D. C. Rose et al. ⁵, decision support tools, usually considered to be software-based, may be an essential part of the quest for evidence-based decision-making in agriculture to improve productivity and environmental outputs. These tools can lead users through exact steps and recommend the best decision paths or may act more as a source of information to advance the evidence base for decisions. Also, offer an intelligent irrigation system to minimise water wastage and improve moisture content in the soil during the dry season.

In a survey, it was revealed that inadequate finances, climate change (weather conditions and late rainfall), irrigation, and marketing incentives, inadequate agricultural extension agents, pest and diseases, bush fires and a lack of information contributes to the declining fortunes of the agriculture sector in low and medium countries (e.g., Ghana)⁶.

Climate change (CC): Climate change can disrupt food availability, reduce access to food, and affect food quality⁷. E.g., estimated rise in temperatures, changes in rainfall patterns, variations in extreme weather condition, and reductions in water accessibility leads to a reduction in agricultural productivity. Increase in recurrence and severity life-threatening weather condition can also disturb food delivery. Which subsequently leading to spikes in food prices after extreme events are expected to be more frequent in the future. Additionally, escalating temperatures can contribute to spoilage and contamination. Lack of Information: The study⁶ makes the case that lack of adequate information is one of the critical problems facing most small scale farmers in Ghana. Most farmers in remote communities have little or no access to information. The ones in suburban communities have limited access to information and lack what it takes to put into practice the information they receive. In situations where access to information such as the use of fertiliser, crop rotation, etc. is available, farmers are unable to understand due to illiteracy⁶. Thus, most of this information is in the English language. A survey conducted⁸ shows that in areas where climate has affected the rainfall pattern, the inability of farmers to put in place a better irrigation system leads to poverty. The inadequate agricultural extension (IAE): Agricultural extension programmes have been one of the primary channels of addressing food insecurity and rural poverty. Since it has the means to support rural adult learning, transfer technology, assist farmers in problem-solving and getting farmers involved in the agricultural knowledge and information system. However, IAE services have been identified in the literature as one of the cores limiting factors to the advancement of the agricultural sector and rural community development at large. i.e. Rural farmers (90% of farmers in Ghana) farming on small hectares of land faces a lack of adequate extension contacts¹. According to a report⁹, terrible road networks in communities where farming is very vibrant contributes to less productivity in agriculture.

Several studies have attempted to minimise these challenges; for example, an optimized solar-powered pump was proposed for efficient irrigation in farms to maximise production and reduce the cost of electric energy¹⁰. Likewise, water from fish ponds was proposed for adequate irrigation using advance technology¹¹. Distillery spent-wash irrigation technology was proposed for sprouting and growth of Asteraceae flowers¹². A hydrogels technology was put forward¹³ as a means to provide irrigation in small scale farming. Similarly, a PIC microcontroller was adopted to propose an intelligent irrigation system for senna crop farming in India¹⁴. An examination of the feasibility of organic farming of fruits in some part of India was undertaken, to improve food security¹⁵. The above study shows that improvement in agriculture is a global concern, due to the part it plays in a different economy. For example, agriculture is said to be the pillar of our economy (Ghana), employing an enormous 42% of the Ghanaian population¹ who are 18 years and older. There is a need to create an enabling environment and grow agriculture as a business. Hence, this paper seeks to a proposed conceptual framework for smart farming based on internet of things (IoT) service and machine learning techniques. Focused on helping farmers make an informed decision to increase crop yield; to promote the millennium development goals (MDGs) of reducing hunger and food security. Expressly, the following specific objectives are set

1. To offer a recommend sowing date for different crops model based on weather condition and rainfall pattern, to educate farmers which crop to grow as a specific period of the year.

2. Propose a water-saving, irrigation-based on various techniques, e.g., thermal imaging, Crop Water Stress Index (CWSI), direct soil water measurements and more to prevent water wastage.

3. Provide a knowledge base of Agric extension services in the various local language to allow easy understanding of the information by farmers.

MATERIAL AND METHODS:

Internet of Things (IoT) is faster-growing technology, which has made tremendous improvement in different economic sectors worldwide¹⁶. In this paper a combination of IoT services, machine learning (ML) algorithms and sensor devices is proposed to achieve the stated goal, to offer an intelligent based management system to help farmers in their day-to-day activities. Figure 1 shows the proposed conceptual framework. Literature has shown that prediction of soil moisture is possible using sensors placed at the field and weather forecasted data. Hence, an IoT based architecture would be used to collect, transmit and process the physical parameters (soil moisture, air temperature, air relative humidity, soil temperature, radiation) of farming land along with the weather forecast information to manage the irrigation efficiently. So, we will consider evaporation of soil moisture based on air temperature, air relative humidity, soil temperature, and radiation. The parameters are considered for analysing the soil moisture drain (change/ difference) pattern based on the recorded data of soil moisture.

ML is a subclass of artificial intelligence (AI) where machines (computers) made to train on a dataset to learning hidden patterns within the data and then given anonymous data to make a prediction based on the pattern learned. ML techniques can be classified into supervised learning, unsupervised techniques, semi-supervised learning and reinforcement learning¹⁷. Several algorithms based on these classifications have been applied in areas different sector of life to achieve compelling results. To mention a few, in finance^18–22, health²³, education^24–26, energy^27–29, agriculture³⁰ and more. A hybrid of supervised and unsupervised ML techniques based on support vector machine (SVM), Convolutional neural networks (CNN) and long short-term memory (LSTM) is proposed. For estimating the difference/change in soil moisture due to weather conditions and further predict the raining pattern in the future. The sensor node data is expected to be collected wirelessly collected over the cloud using web-services and a web-based information visualisation and decision support system that provides real-time information intuitions based on the examination of sensors data and weather forecast data.

Figure 1 Propose Conceptual Framework

CONCLUSION:

The current paper proposed a conceptual framework based on IoT and machine learning techniques to help farmers in developing countries overcome challenges in small scale farming. Such challenges are, irregular rain pattern, access to adequate information, inadequate agricultural extension agents, bush fires destroying crops pest and diseases, and more, which affect low productive and food security. It is believed that the successful implementation of the proposed system will go a long way to help achieve the MDGs of reducing hunger and food security. Also, it will provide closed-loop control of the water supply to realise a fully autonomous irrigation scheme where possible for small scale farming communities. Again, to provide the farmer with the future rainfall pattern so that formers can know the type of crop to plant at a specific period. Furthermore, it is expecting to offer an accumulation of expert knowledge to a farmer in a more familiarised language. In future, the authors look forward to financial support to bring the proposed framework to reality, by the implementation of the proposed system.

ACKNOWLEDGEMENT:

The author is grateful to all who contribute in any form to the success of this study.

CONFLICT OF INTEREST:

The author declares no conflict of interest.

REFERENCES:

1. Danso-Abbeam, G., Ehiakpor, D. S. and Aidoo, R. Agricultural extension and its effects on farm productivity and income: insight from Northern Ghana. Agric. Food Secur. 7, 74 (2018).

2. Cotter, M. et al. Creating the data basis to adapt agricultural decision support tools to new environments , land management and climate change — A case study of the RiceAdvice App. 423–432 (2020) doi:10.1111/jac.12421.

3. Goap, A., Sharma, D., Shukla, A. K. and Krishna, C. R. An IoT based smart irrigation management system using Machine learning and open source technologies. Comput. Electron. Agric. 155, 41–49 (2018).

4. Van Oort, P. A. J. and Zwart, S. J. Impacts of climate change on rice production in Africa and causes of simulated yield changes. Glob. Chang. Biol. 24, 1029–1045 (2018).

5. Rose, D. C. et al. Decision support tools for agriculture : Towards effective design and delivery. AGSY 149, 165–174 (2016).

6. Asiamah, D. A. and Darko, E. T. Part I: Why Ghana’s agricultural growth has slowed down. https://farmerline.co/2016/11/01/part-i-why-ghanas-agricultural-growth-has-slowed-down/#:~:text=Inadequate finances%2C climate change%2C poor,declining fortunes of the sector. (2016).

7. Brown, M. E. et al. Climate Change, Global Food Security, and the U.S. Food System. BMC Public Health vol. 5 https://ejournal.poltektegal.ac.id/index.php/siklus/article/view/298%0Ahttp://repositorio.unan.edu.ni/2986/1/5624.pdf%0Ahttp://dx.doi.org/10.1016/j.jana.2015.10.005%0Ahttp://www.biomedcentral.com/1471-2458/12/58%0Ahttp://ovidsp.ovid.com/ovidweb.cgi?T=JSandP (2015).

8. Bullappa, H. and Shivkumar, K. The Impact of Irrigation on Poverty with Special Reference to-Asian Countries. Asian J. Manag. 9, 1109 (2018).

9. Balasubramanian, M., Anandh, K. S. and Kumar, T. P. Study and Assessment of Rural Roads Network in Agriculture Productivity. Int. J. Technol. 5, 125 (2015).

10. Tawale, N. and Pote, G. R. Design and Development of Optimized Solar Powered Water Pump for Farming and Irrigation. Int. J. Technol. 4, 265–282 (2014).

11. Amaliya, N. K. and Kumar, S. P. Status of Pond Waters for Irrigation in Kanyakumari District, Tamilnadu, India. Asian J. Res. Chem. 8, 253 (2015).

12. Chandraju, S., Thejovathi, C. and Chidan Kumar, C. S. Impact of distillery spentwash irrigation on sprouting and growth of tagetes (Asteraceae) flowering plant. J. Chem. Pharm. Res. 3, 376–381 (2011).

13. Singh, H., Das, A. and Singh, S. Superabsorbent Polymers-A Potential Solution for Irrigation in Agriculture. Res. J. Pharm. Technol. 12, 2566 (2019).

14. Rajkanna, U., Karthickkumar, T., Jayaraman, L. and Mathankumar, M. Senna crop irrigation. Res. J. Pharm. Technol. 11, 2656–2658 (2018).

15. Sharma, H. L. and Thakur, V. Feasibility of organic farming of fruits in Himachal Pradesh. Int. J. Technol. 6, 278 (2016).

16. Mohapatra, R. K., Mohapatra, B. N. and Panda, P. P. Application and Security in Internet of Things (IOTs). Int. J. Technol. 9, 1 (2019).

17. Nti, I. K., Adekoya, A. F. and Weyori, B. A. A systematic review of fundamental and technical analysis of stock market predictions. Artif. Intell. Rev. 53, 3007–3057 (2019).

18. Nti, I. K., Adekoya, A. F. and Weyori, B. A. Predicting Stock Market Price Movement Using Sentiment Analysis: Evidence From Ghana. Appl. Comput. Syst. 25, 33–42 (2020).

19. Nti, I. K., Adekoya, A. F. and Weyori, B. A. Efficient Stock-Market Prediction Using Ensemble Support Vector Machine. Open Comput. Sci. 10, 153–163 (2020).

20. Nti, I. K., Adekoya, A. F. and Weyori, B. A. A comprehensive evaluation of ensemble learning for stock-market prediction. J. Big Data 7, 20 (2020).

21. Nti, I. K., Adekoya, A. F. and Weyori, B. A. Random Forest Based Feature Selection of Macroeconomic Variables for Stock Market Prediction. Am. J. Appl. Sci. 16, 200–212 (2019).

22. Adekoya, A. F. and Nti, I. K. The COVID-19 outbreak and effects on major stock market indices across the globe: A machine learning approach. Indian J. Sci. Technol. 13, 3695–3706 (2020).

23. Akyeramfo-Sam, S., Addo Philip, A., Yeboah, D., Nartey, N. C. and Nti, I. K. A Web-Based Skin Disease Diagnosis Using Convolutional Neural Networks. Int. J. Inf. Technol. Comput. Sci. 11, 54–60 (2019).

24. Nti, I. K. and Quarcoo, J. A. Self-motivation and Academic Performance In Computer Programming Language Using a Hybridised Machine Learning Technique. Int. J. Artif. Intell. Expert Syst. 8, 12–30 (2019).

25. Nti, I. K., Adekoya, A. F., Opoku, M. and Nimbe, P. Synchronising Social Media into Teaching and Learning Settings at Tertiary Education. Int. J. Soc. Media Interact. Learn. Environ. 6, 1 (2020).

26. Mohamed, A., Rizaner, A. and Hakan, A. Using data Mining to Predict Instructor Performance. Procedia - Procedia Comput. Sci. 102, 137–142 (2016).

27. Rodríguez, F., Florez-Tapia, A. M., Fontán, L. and Galarza, A. Very short-term wind power density forecasting through artificial neural networks for microgrid control. Renew. Energy 145, 1517–1527 (2020).

28. Zhang, J. et al. A reinforcement learning based approach for on-line adaptive parameter extraction of photovoltaic array models. Energy Convers. Manag. 214, 112875 (2020).

29. Nti, I. K. et al. Predicting Monthly Electricity Demand Using Soft-Computing Technique. Int. Res. J. Eng. Technol. 06, 1967–1973 (2019).

30. Nti, K. I., Eric, G. and Samuel, Y. Detection of Plant Leaf Disease Employing Image Processing and Gaussian Smoothing Approach. Int. J. Comput. Appl. 162, 20–25 (2017).

Received on 16.12.2020 Accepted 03.05.2021

International J. Technology. 2021; 11(1):1-5.

DOI: 10.52711/2231-3915.2021.00001