Agriculture not only provides fuel for billions of people, but also jobs opportunity applications. One of the most important distinguishing factors humans and everything else, is cognition. Artificial Intelligence is the process of making a computer or robot smart and think the way a human think and solve problems. According to Professor McCarthy, artificial intelligence is “the science and technology of developing intelligent machines, especially intelligent computer software.” Mainstream “AI” has existed for decades in the form of rule-based programs that provide basic manifestations of “intelligence” in certain contexts. However, progress has been limited because the methods for solving many real-world problems are too complex for humans to program by hand. What if the burden of making complex predictions, processing data, and specifying models could be shifted from the programmer to the programmer? This is the promise of modern artificial intelligence.
Machine learning is an area of artificial intelligence in which progress is rapid and dynamic. Machine learning solves complex problems that humans can solve by shifting the decision burden to an algorithm. As AI pioneer Arthur Samuel wrote in 1959, machine learning is “an area of study that allows computers to learn without being explicitly programmed.” The goal of machine learning is to develop a predictive model for a specific use case by writing a program for each type of object that needs to be identified. To solve the problem of writing a separate program for each discovery, Deep Learning went beyond the basics. Deep learning is an aspect of machine learning that saves the programmer the time and effort required to perform graphical or optimization tasks. Deep learning has changed the world of artificial intelligence. An artificial neural network is a powerful but highly flexible learning with three layers ie. an input layer, an output layer and multiple layers called “deep neural network”. The neural system of living organisms, such as the brain, inspired ANN as an information processing technique for information processing. Using this method, with increasing efficiency, we can now:
- Image Processing
- Real-time inter-language interpretation
- Use speech to control equipment
- Predict how a genetic mutation will affect DNA transcription.
- Proper cultivation
- Phenotyping and crop analysis
- Detection of tumors on medical images
According to (FAO, 2017), global population growth is slowing, in some areas the population will continue to increase after 2050 and into the next century, with more people living in cities than in rural areas, and this difference is predicted to increase as the population increases. Agriculture feeds the world, and the world’s population is growing rapidly. In 2019, the world population is 7.7 billion, and by 2050, the population will increase by two billion, bringing the world population to nine billion. The environmental burden created on the planet by increasing human population and industry, including agriculture, is contributing to global warming. Various forms of cruelty lead to soil deterioration, leading to deterioration of crop quality; Chemical runoff contributes to the formation of dead tissue and threatens marine life. Thus, the use of artificial intelligence in agriculture can be critical in solving critical problems such as pest and disease attacks, inadequate pesticide use, inadequate drainage and irrigation, weed control and yield prediction, to name a few. The complex interaction of soil, seed and crop chemicals is due to agricultural production.
Precision farming Suitable agriculture is based on the motto “In the right place, at the right time and in the right place”. Precision farming replaces the repetitive and labor-intensive farming sector with more precise and controlled methods than traditional ones. According to Gibbons (2000) and Waheed et al. (2006), advanced data processing technologies for timely crop management in season, such as variable rate technology, airborne and satellite remote sensing, multispectral and hyperspectral terrestrial, computer simulation, global positioning systems (GPS); geographic information systems (GIS). innovative methodological approaches on which real farming is based. According to Cox (2002), livestock use as well as spatially diverse farm activities enabled by satellite-based global positioning system (GPS) are placed under the general heading of precision agriculture (or farming). correct). Ulla et al. (2017) discussed precision cultivation, data collection, integration of different technologies, and efficiency analysis to improve production efficiency while reducing costs. Yangdong et al. (2017) described precision horticulture as a way to increase profitability and productivity in resource use, thus achieving this goal in the face of various problems faced by agribusinesses, mainly due to climate change, land degradation, agricultural land availability, labor shortage and increasing costs. With declining equipment costs, increasing computing power, and increasing interest in nondestructive food analysis techniques, interest in modeling and computer vision has increased in agriculture in recent years.
Ulla et al. (2017) were instructed to consider agricultural problems, different precision farming systems based on artificial intelligence and machine learning, and future directions. According to the study, there are useful technologies for precision farming such as GPS/GNSS, mobile devices, robotics, unmanned tractor, irrigation, unmanned aerial vehicle (UAV), Internet of Things (IoT), sensors, seeding seeds at variable rates, climate simulations, Data collection, data analysis, management decisions and cultivation are the four main phases of precision cultivation. Over the last two decades, new AI-based technologies have been developed for efficient farming, such as artificial neutral networks (ANNs) and fuzzy logic controllers to control temperature and humidity in synthetic greenhouses. Similarly, the use of neural networks and fuzzy logic in crop allocation for crop mapping is useful as it ultimately allows the identification of crop water demand. In addition, Fuzzy Logic can be used to sort crops such as apples, tomatoes, lettuce, cauliflower and even mangoes. Such methods involve image capture or data input, image extraction, and subsequent classification and/or scoring. Based on parameters such as size, shape, color, aroma etc., the final assessment of yield is done on a scale, for example, in the range from 1 to 10. Similarly, it is possible to classify dates trees based on conditions and yields, which they can provide to help farmers make the most of their resources.