Application of Artificial Intelligence in Horticulture

The horticulture industry requires people with artificial intelligence acumen to enable them to improve yield.

India, with its diverse soil and climate conditions and varied agro-ecological regions, provides a possibility to grow a large number of horticultural crops. Horticultural crops, which include fruits, vegetables, medicinal, aromatic, and ornamental plants, are key dietary healthy ingredients, sources of medicines and aroma, as well as have numerous aesthetic qualities for humans.

In the country, 90% of the total horticultural production is contributed by vegetables and fruits. Presently, our country is next to China in the area and production of fruits and vegetable crops, producing 10 percent of fruits and 14 percent of vegetables in global production. Horticulture crops play a unique role in India’s economy by improving the income of the rural people and generating employment for youth.

Based on new technologies and data analysis methods, the development of intelligent systems has improved the scope to enhance the cultivation and management aspects of horticultural crops. Surprisingly, horticulture, while being the least digital, has undergone a surge in agricultural technology research and commercialization. However, the sector is subjected to a broad range of biotic and abiotic stress affecting the yield and quality of the produce. In such circumstances, better and more efficient production approaches can be accomplished by using the recent technological refinements and solutions.

Artificial intelligence (AI) also called machine intelligence is a domain in computer science that instructs machines on how to replicate human physical actions and react like humans. Its applications in the horticulture industry are diverse. From planting, watering, and harvesting to experimenting with new systems, machines can often be created to be more efficient than people. The horticulture industry requires people with artificial intelligence acumen to enable them to improve yield. Artificial Intelligence technologies are helpful to yield healthier crops, provide information on prevailing weather conditions such as temperature, rain, wind speed, wind direction, solar radiation, pest control, monitor soil, and growing conditions, organize data for farmers, help with the workload and improve food supply chain (Manaware, 2020).

Application of AI in Horticulture

Horticultural operations are arduous and expensive and it is challenging to get and retain the labor force in this sector. The use of robots, drones, and automated systems is the future when it comes to enhancing efficiency and productivity within the horticulture industry. Some of the important uses of AI in horticulture are as:-

Disease diagnosis: Pre-processing of the image ensures the leaf images are segmented into areas like background, non-diseased part, and diseased part. The diseased portion is then trimmed and sent to remote labs for further diagnosis. It also helps in pest identification, nutrient deficiency recognition, and recommendations on disease diagnosis on a real-time basis. By this, there is a reduction in pesticide losses which also leads to reducing the contamination of the soil and groundwater as well as the chances of pesticide residues in the human food system. This also helps farmers to overcome the labor challenge.
Produce maturity identification: Images of different crops under white/UV-A light are captured to determine the proper stage of maturity of fruits. Farmers can create different maturity grades based on the crop/fruit category and add them into separate stacks before sending them to the market, especially in the case of highly perishable horticulture crops, and harvesting at proper maturity will enhance post-harvest shelf life.

Field management: Using high-definition images from airborne systems (drones or copters), real-time estimates can be made during the cultivation period by creating a field map and identifying spots where crops need water, fertilizer, or pesticides. This will sustain resource optimization up to a great extent.
Automation Systems in Irrigation: The smart Irrigation system is an Internet of things (IoT) based instrument which can automate the irrigation process by analyzing the moisture status of soil and the weather parameters. Irrigation is one of the most labor-intensive processes in farming which can be avoided by artificial intelligence because it is aware of historical weather patterns, soil quality, and the type of crops to be grown.

Automated irrigation systems are designed to utilize real-time devices which can constantly maintain desired soil conditions to increase water use efficiency and average returns. With close to 70% of the world’s freshwater being used in irrigation, automation can help farmers better manage these problems, and the objective of more crops per drop can be fulfilled.
Grading of fruits: The use of image processing for the grading of fruits has increased in recent years. Grading is an important step in the post-harvest process and involves the categorization of fruits, with consideration of the severity of the disease, defects, and contamination on fruits. Manually grading fruits is time taking and unreliable process. Consequently, it is needful to adapt the automated faster system in this regard. One such reliable method is the automatic image processing technique for sorting and grading of fruits.

Drone-Based Technology:

This technology is providing new ways of enhancing crop yields through in-depth field analysis, long-distance crop spraying and high-efficiency crop monitoring and is quickly becoming invaluable for farmers. Drones monitor the condition of the soil and determine whether it requires watering or planting. Cameras monitor vegetables and fruits and are being taught to identify aberrations or problems such as dehydration and undesirable insects. All of this data can be used to foresee the ideal time to harvest. Drone technology has given agriculture a high-tech makeover. Further drones can scan the ground, and spray in real-time for even coverage. The result of aerial spraying is five times prompter with drones than conventional machinery. Here are ways drones will be used throughout the crop cycle. Some of the applications of drones in horticulture are discussed as:-

Crop monitoring: Ineffective crop monitoring is a huge obstacle. With drones, time-series animations can reveal the growth of a crop and predict production inefficiencies, enabling better management practices.

Use of robotics in horticulture

Robots in the horticultural sector are widely utilized in harvesting, drone spraying, field monitoring, sorting, grading, packing of final horticultural produce, nurseries and greenhouses to some extent. H2L developed a sorting robot for tulip cultivation. Several robots are already being designed for fruit cultivation as well. Robots do the heavy and repetitive work in no time at all, without ever getting exhausted. The other uses of robotics areas:-

Setting pots in the greenhouse: Numerous applications can be developed to make the cultivation and delivery of potted plants easier through robotics and, in doing so, heighten the productivity. The amazing advantage of a robotic arm is that it can perform multiple tasks in a 3D field, such as putting plants in a carrier and, in the next movement, taking them out of a carrier and putting them on tables or in trays. This is performed with the greatest precision, even if the plants are not located precisely in the right place.
Fruit harvesting robot: These robots require to pick up fruits without damaging the branches or leaves of the tree. The robots must be capable to access all areas of the tree being harvested and distinguish between fruits and leaves by using video image capturing. The camera is mounted on the robot arm and the colours detected are compared with properties stored in the memory. If a match is obtained, the fruit is harvested. If the fruit is hidden by leaves, an air jet can be used to blow leaves out the way for a clearer view. The pressure applied to the fruit is sufficient for removal from the tree, but not enough to bash the fruit. The shape of the gripper relies on the fruit being picked. A drone is used to pick apples and the company Bx recently showed a video robot ‘dog’ Spot from Boston Dynamics walking through an apple orchard with a camera and sensors.

The Quad Duster system: When a kiwifruit flower is well pollinated, it produces an abundance of seeds. The more seeds, the heavier the fruit, which is often what the grower desires. The Quad Duster system was developed to enhance the pollination of kiwifruit – pollination is essential for fruit yield and quality. The Quad Duster allows the uniform distribution of dry pollen into the vine canopy. Spreading the pollen as a dry powder, as opposed to it being mixed in a water solution, allows for bees to collect and transfer dry pollen that doesn’t land directly on the stigma of the flower during application. In 2015, Quad Dusters were used in about 25% of New Zealand’s kiwifruit industry, and this is set to grow. Insufficient pollination of kiwifruit flowers has been linked to small and deformed fruit that is not suitable for export markets and therefore reduces returns to growers.
The Robotic Apple Packing Cell: The Robotic Apple Packing Cell went into commercial trial in the 2016 season. The robot cells are made to work in standard industry pack houses. The system (of six individual packer robots) packs 120 apples per minute, orienting the fruit so all of the stems are lying horizontally in the trays and pointing in the same direction, all while having the color side of the apple facing up for optimum presentation. Camera technology is employed in this system, but it is the vision algorithms used to orient the stem in the right direction that is the really smart technology in this robot. One robot accomplishes the job of 2–3 people. The robot operates pneumatics (suction cups) to handle the fruit. This is mild and assures that no human hands touch the fruit during packing (reducing contamination issues).
Multipurpose Orchard Robotics: Robotics Plus is working on the Multipurpose Orchard Robotics project. The project aims to automate the harvesting and pollination of kiwifruit and apples by developing a central system that further modules can be added for different tasks such as pollination, spraying, and harvesting. There is also a lot going on around this subject where greenhouse horticulture is concerned.

Robots are being developed for picking tomatoes and various crop operations such as leaf pruning and pollination. Work is also being done on approaches that can predict harvests and that can take measurements of crops. All these data and measurements are important for the grower to be able to manage their crop in the greenhouse as effectively as possible. A good number of companies are investing in this aspect around the world.
Robots to assess orchard conditions: Researchers at the Australian Centre for Field Robotics (ACFR) at the University of Sydney have developed two ground-based mobile robots nick-named Mantis and Shrimp that utilize artificial intelligence to assess orchard conditions. These robots have been tested in both apple and almond orchards. The robots use artificial intelligence to assess canopy volume, which directly relates to yield and compare this data to historical data to estimate flower and fruit density. Once the tree has fruit, the robots can do a prompt count to anticipate harvest yield.
Strawberry harvesting robot: Labor shortage spurs farmers to use robots for handling delicate tasks in the fresh produce industry. An automated harvester wheeled through rows of strawberry plants can reduce the cost of picking, which otherwise is a labor-intensive operation.
A lettuce-thinning robot: It is used for thinning and weeding of lettuce to increase yield. Its vision system scrutinizes each plant and then applies advanced artificial intelligence algorithms that make plant-by-plant decisions to optimize yield and then eliminate unwanted plants according to its programming.

This is how robotics help in digital farming.

Expert Systems Implemented at CLAES:

Agricultural production has evolved into a complex business requiring the accumulation and integration of knowledge and information from many diverse sources. To remain competitive, the modern farmer often relies on agricultural specialists and advisors to provide information for decision-making. Unfortunately, agricultural specialist assistance is not always available when the farmer needs it. To alleviate this problem, expert systems were identified as a powerful tool with extensive potential in agriculture. An Expert System (ES), also called a Knowledge-Based System (KBS), is a computer program designed to simulate the problem-solving behavior of an expert in a narrow domain or discipline. In agriculture, expert systems unite the accumulated expertise of individual disciplines, viz plant pathology, entomology, horticulture, and agricultural meteorology, into a framework that best addresses the specific on-site needs of farmers.

Expert systems combine the experimental and experiential knowledge with the reflexive reasoning skills of a multitude of specialists to aid farmers in making the best decisions for their crops. Some of the expert systems developed for the specific crops are; Cuptex: An Expert System for Cucumber Crop Production, Citex: An Expert System for Orange Production Tomatex: An Expert System for Tomatoes, and Limex: A Multimedia Expert System for Lime Production, etc.

Though Artificial Intelligence offers vast opportunities for application in horticulture, there still exists a lack of familiarity with high-tech machine learning solutions on farms across most parts of the world. Exposure of farming to external factors like weather conditions, soil conditions and presence of pests is quite a lot.

Conclusion

Artificial Intelligence in farming not only helps growers to automate their farming but also shifts to precise cultivation for higher crop yield and better quality while utilizing fewer resources. Labour is a major cost for horticulture. Intensive horticultural crops require a much more skilled labor force than broad-scale farming. Almost 50% of production costs account for hired labor engaged for different operations. AI can reduce the cost of cultivation by regulating the use of labor, efficient use of fertilizers and pesticides, and reducing crop losses by harvesting at a proper time. AI can help farmers to increase the capacity of production and reduce the cost of production. The application of AI in all application domains will also bring an ideal shift in the way we do research and development in horticulture now. Artificial Intelligence-based products or services like training data for agriculture, drone, and automated machine making will get technological advancements in the future and will provide more useful applications to this sector, for increasing efficiency.

References

Bhavana, H and Bhagwan, A.2021. Review on: Role of robotics in horticulture. Journal of Pharmacognosy and Phytochemistry Sp 10(1): 306-309
Chandani, D. 2019. Cardamom grading using Artificial Intelligence. Indian Farming. 69(03): 46-48
Dharmaraj,V and Vijayanand,C.2018. “Artificial Intelligence (AI) in Agriculture”, International Journal of Current Microbiology and Applied Sciences 7 (12): ISSN: 2319-7706
Dushyant D.Champaneri.2021. Artificial intelligence: An Intelligent Path to improve Indian Agriculture. Just Agriculture; 1(10):2582-8223.

https://innovationorigins.com/en/how-are-robots-doing-in-greenhouse-agriculture/
https://www.researchgate.net/publication/221928727 Postharvest Technologies of Fresh Horticulture Produce
https://www.researchgate.net/publication/339923291 Identification classification grading of fruits using machine learning computer intelligence: a review
Jan Bontsema, Jochen Hemming, Erik Pekkeriet, Ola Ringdahl.2015. CROPS: Clever Robots for Crops. Eng. Technol., 1-11.
Liakos, K., Busato, P., Moshou, D., Pearson, S., Bochtis, D.2018. Machine Learning in Agriculture: A Review. Sensors 18 (8), 2674.
Manaware, D. 2020. Artificial Intelligence: A New Way to Improve Indian Agriculture. International journal of Current Microbiology and Applied Sciences 9(03):1095-1102.doi: https://doi.org/10.20546/ijcmas.2020.903.128
Saxena, A.; Suna, T. and Saha, D. (2020). “Souvenir: 19 National Convention”. Application of Artificial Intelligence in Indian Agriculture. RCA Alumni Association, Udaipur, India
Ya Xiong, Yuanyue Ge, Lars Grimstad, Pål J. 2020. An autonomous strawberry-harvesting robot: Design, development, integration, and field evaluation. J Field Robotics; 37:202-224.