Historically, the agriculture sector is the oldest and largest industry in the world, as it provides food and resources for the survival of humans. It is predicted that by 2050, the global population will exceed 20 million people. This increase in population will also raise the need for food. Hence, there is a requirement to use advanced technologies to enhance crop production. In recent years technology has had a huge impact on agricultural productivity and solved the problems faced by the agricultural field.
Artificial Intelligence (AI) is a cutting-edge technology that is adopted as smart or precision agriculture is a solution to problems faced by the agriculture sector. AI has revolutionized agriculture. Just like other industries, AI in agriculture relies on automated tools (machines) such as cameras and robots, sensors and drones to make practical effects that increase and regulate the yield of the farm. These cutting-edge tools collected data on a variety of variables, including temperature, nutrient status, soil moisture and diseases. The computer is then supplied with this data and analysis is made so that the crop grows under the normal condition. AI plays a key role in reducing inputs in farming activities and enhancing yield, thereby paving the way for eco-friendly farming practices.
Here are some common applications of AI in agriculture:
Detecting Disease and pest
AI technology is altering the way we detect diseases and pests in crops. It allows the farmers to detect agricultural issues rapidly and precisely. By using AI algorithms, farmers can predict when pests or diseases are likely to occur, allowing them to take preventive measures to protect their crops. This can help to reduce the need for harmful pesticides and other chemicals. AI-based drones with advanced cameras fly over the field and take images of the plants. AI algorithms analyse the images to detect abnormalities such as discoloured leaves which might indicate disease or nutrient deficiency. It has been used to identify apple black rot disease with more than 90% accuracy.
AI technology known as convolutional neural networks (CNNs) had great accuracy exceeding 92% in detecting plant problems. PlantVillage is an app that diagnoses multiple diseases using AI tools. Farmers can wave their phone over a specific leaf and if the plant has a disease or pest attack, the app will identify it and suggest the best ways to manage the problem.
Automated and precise irrigation
Approximately 85% of freshwater resources are utilised by the agriculture industry and this percentage is growing steadily with the increase in population and food consumption. To ensure optimum water consumption, it is essential to develop and utilize advanced irrigation technologies. AI-based irrigation relies on different sensors in the soil and environment to collect real-time data. Moisture sensors detect moisture content and water level in the soil to estimate the water requirement of crops. Sensors also measure environmental temperature and humidity to determine evaporation and the best sowing time. Similarly, rainfall-detecting sensors are used to avoid over-irrigation. In scorching weather, AI can increase irrigation according to the crop requirement. Soil moisture sensors notify the moisture content to farmers on their phones via SMS.
The farmers can command ON or OFF the water supply, according to the requirement of the crop. Drones and satellites take pictures using advanced cameras and determine the areas for irrigation by analyzing the color and texture of the crop. Drip irrigation is an automated method that uses smart valves that direct the water flow. It reduces the water loss. AI can also detect the parts of the irrigation system, such as clogging pipes, valves and pumps, with the help of a flow detector, enabling immediate maintenance.
Monitoring soil health
AI can be used for chemical analysis to determine missing nutrients in the soil. Nutrient scanners and fertility meters take data from soil samples and give farmers precise estimations of deficient nutrients and general soil health. Farmers can use this information to determine the best time to fertilize their crops and how much fertilizer is required. This not only ensures that the crop is receiving necessary nutrients but also avoids over-fertilization.
AgroCare, an AI-based software, helps the farmers to identify the amount of primary nutrients and offers suggestions to improve soil health. Similarly, a German company designed an AI-powered app called Plantix that can detect nutrient deficiency in soil, including plant diseases and pests. These apps help farmers to improve their fertilizer application, maintain soil health, and improve yield and production.
Yield prediction
AI is also revolutionizing the way farmers predict crop yields. By analyzing the nutrient content in the soil and infestation of pests or disease, farmers can predict the performance of crops more accurately. This information can help farmers to make better decisions about planting and crop management, which leads to higher yields. Studies revealed that Sentinel satellites can detect the crop yield by analyzing the crop vigor, health and stress because these characteristics are directly correlated with crop yield. In Australia, support vector machines are used to estimate the wheat yield and CNN was used to predict the rice yield in China.
Autonomous weeding robots
Manually hoeing weeds is labour-intensive work and using herbicides also damages the crop. Scientists have developed robots to eliminate the weeds from the field using AI. These robots use computer vision and image recognition systems to distinguish between weeds and crops based on leaf color, shape and size. This system removes the weeds using mechanical tools or applies herbicides only to weeds, reducing the use of chemicals. It protects the crop from herbicides and environmentally friendly agricultural practices.
AI in Plant Breeding
Phenotypic data collection in the field for breeding programs is labor-intensive and time-consuming. AI has altered the data collection and plant breeding techniques. The high-throughput technique uses AI-powered drones, cameras and sensors to collect phenotypic data from thousands of plants simultaneously. The data is recorded with high accuracy, hence improving the selection intensity. AI is being used to develop biotic (disease and pests) and abiotic (drought and heat) resistance varieties by analysing the genotypic data and identifying the genes and markers related to the specific trait.
Diseases, pests and drought-resistance genes can also be identified using AI. It helps to detect the genetic markers linked to specific traits and helps the breeder to select the plants at early stages for those traits. In a hybrid breeding program, AI is used to identify the best parental combinations/hybrids based on the genotypic data of plants. Hence, genomic and marker-assisted selection using AI accelerates the breeding program.
Conclusion
The agricultural sector confronts numerous challenges such as harsh weather, inefficient irrigation systems, and biotic and abiotic stress. Farmers can overcome these challenges by integrating artificial intelligence into agricultural practices. Farmers may use sensors, robotics, and drones to monitor their crops more accurately and effectively. AI enables farmers to make better decisions and optimise crop management using real-time data. It results in better yield production, pest management and increased profits.
