Soil is the foundation of global agriculture and food production and is considered a vital resource. Unfortunately, soil quality can be difficult to assess due to the inability to geographical restrictions and lack of technical resources.
Image Credit: Leonid Sorokin/Shutterstock.com
However, computer science innovations like artificial intelligence (AI) technologies can help protect and conserve soil quality, enabling faster and safer processing of enormous amounts of data obtained during physical soil sampling and remote imaging.
AI in Monitoring Soil Quality
Artificial intelligence and machine learning (ML) technologies that monitor soil quality and fertility utilize different algorithms for agriculture analysis.
Machine learning applications use supervised and unsupervised methods to support data analysis procedures, generating sufficient elements to provide a statistical solution to the problems requiring these techniques.
With the help of artificial intelligence technologies, particularly electronic applications for deep learning, farmers can find potential nutrient deficiencies in soil quality.
Different agricultural technologies like Farm Beats have been built where farmers only need to take a picture with their smartphone and then upload the image to an AI development system. After assessing the problem, farmers are provided with restoration techniques and other solutions that will help improve the soil quality and quantity of the crop.
Artificial Intelligence Technologies in Agriculture
Agricultural producers now have access to previously inaccessible agricultural data sources for making decisions, such as satellite and unmanned aerial vehicles (UAV), humidity sensor readings, and ground-based weather stations. Simultaneously, new monitoring and control systems are continually introduced to the market, providing more personalized, accurate analysis and predictions on soil quality.
In 2014, 190,000 measurements were taken daily on intelligent agricultural infrastructure integrated into farms. By 2050, the number of measurements will grow to 4.1 million per day. It is almost impossible to navigate this flow of information without the help of computer systems like artificial intelligence technologies.
The generalization, analysis, and processing of data from various monitoring technologies and the issuance of suggestions based on them are some of the duties of artificial intelligence in soil quality analysis and the broader agricultural industry.
AI Technologies to Monitor Soil Quality
Brazilian agricultural startup InCeres has developed an app that can predict soil quality and fertility based on soil application and nutrient uptake. The analysis is based on data on the chemical composition of the soil, weather conditions, crop types, and satellite images showing plant growth rates.
To predict soil fertility, the application developed by InCeres uses AI systems that analyze a vast amount of data and produces accurate forecasts for each specific area of farmland.
According to Leonardo Menegatti, principal researcher at InCeres, the standard method of chemical analysis over ten years will cost the farmer R$200(Brazilian real) per hectare, while the new approach will cost a total of about R$40 over the same period, saving 80%. The technology already offers solutions for the profitable management of agricultural businesses. As the program improves, the application will learn and predict the soil quality in the future.
From other soil quality analysis strategies, Varatharajalu and Ramprabu have presented an automated watering system that employs a soil moisture sensor, temperature sensor, pressure regulator sensor, and molecular sensor for enhancing crop growth.
Outputs from the sensors are converted to digital signals and transmitted to the multiplexer over a wireless network such as Zigbee or a hotspot.
Dr. Ali Al-Naji and Professor Javaan Chahl of the University of South Australia have worked on a device that accurately measures soil quality indicators like moisture with the help of a typical RGB digital camera. It utilizes a common video camera to analyze changes in soil color to detect moisture content.
The digital camera was linked to an artificial neural network (ANN) programmed to recognize different soil moisture levels under various weather situations.
Challenges that Face AI Systems in Agriculture
Data preparation is a significant obstacle to forecasting and estimating variables like soil quality.
Agricultural research is typically based on robust data sets that are reproducible and representative. As a result, even though we can create numerous Al models, temperature, soil moisture, photosynthetic rate, and ecological balance can all be affected by weather and climate change. Under such conditions, crop production could collapse substantially in a short amount of time, resulting in a need for frequent data collection.
It is inevitable that increased digitization across industries will be translated into agriculture. Intelligent machines like artificial intelligence and machine learning can transform basic data inputs into beneficial information.
When applied to the agricultural sector, even to monitoring variables like soil quality, it could transform the global food supply chain.
References and Further Reading
Al-Naji, A., Fakhri, A. B., Gharghan, S. K., & Chahl, J. (2021). Soil color analysis based on an RGB camera and an artificial neural network towards smart irrigation: A pilot study. Heliyon, 7(1), e06078 https://doi.org/10.1016/j.heliyon.2021.e06078
Dharmaraj, V., & Vijayanand, C. (2018). Artificial intelligence (AI) in agriculture. International Journal of Current Microbiology and Applied Sciences, 7(12), 2122-2128. https://doi.org/10.20546/ijcmas.2018.712.241
Varatharajalu, K., & Ramprabu, J. (2018). Wireless Irrigation System via Phone Call & SMS. Int. J. Eng. Adv. Technol, 8, 397-401. Retrieved from https://www.ijeat.org/wp-content/uploads/papers/v8i2s/B10821282S18.pdf
Mishra, S. (2022). Emerging Technologies—Principles and Applications in Precision Agriculture. In Data Science in Agriculture and Natural Resource Management (pp. 31-53). Springer, Singapore. https://doi.org/10.1007/978-981-16-5847-1_2
4EI. (2021). Technology Innovation is Transforming Soil Health Monitoring. [Online] 4 Earth Intelligence. Available at: https://www.4earthintelligence.com/insights/technology-innovation-transforming-soil-health-monitoring
World Bank. (2021). A Roadmap for Building the Digital Future of Food and Agriculture. [Online] The World Bank. Available at: https://www.worldbank.org/en/news/feature/2021/03/16/a-roadmap-for-building-the-digital-future-of-food-and-agriculture
InCeres. (2018). Artificial Intelligence for Soil Fertility Control. [Online] Available at: https://pesquisaparainovacao.fapesp.br/artificial_intelligence_for_soil_fertility_control/747
Disclaimer: The views expressed here are those of the author expressed in their private capacity and do not necessarily represent the views of AZoM.com Limited T/A AZoNetwork the owner and operator of this website. This disclaimer forms part of the Terms and conditions of use of this website.