How can we implement artificial intelligence into smart farm?

As interest in agriculture has increased over the past few years, interest in the Smart farm has also increased naturally.

People have variously tried to apply IoT to agriculture. As a result, the advancement in IoT technology for agriculture happens.

So, I would like to suggest some ideas on how it can be evolved to meet these developed growths and what solutions will be available.

※ The following contents are extracted and reconstructed from ITFIND which is an IT knowledge portal.

“The abroad case and utilization plan of the agricultural artificial intelligence”

Developed countries such as the United States are actively adopting artificial intelligence in three areas; agriculture robots, monitoring of crop and soil, and predictive analysis, to increase productivity and protect the environment in terms of climate change, population growth, and food security. Artificial intelligence has been introduced to enable more intelligent tasks to the areas with a simple process of repetition of statistics and machines in the past.

– Agriculture Robot: We develop autonomous robots that handle essential agricultural tasks such as harvesting crops more and faster than human workers. For example, weeding robots use herbicides only in the right place to reduce the environmental pollution and costs. Today, more than 250 weeds are known to be resistant to herbicides. As a result, the annual production loss is estimated at about $43 billion. Blue River Technology in the United States said that they reduced the amount of herbicide spraying by 80% and cost by more than 90% using the See Spray Robot which accurately sprays herbicides on the weeds using visual recognition system.

see spray

Picture1. See Spray Robot from Blue River Technology

– Monitoring of crop and soil: This is a technology to measure and monitor the health of crops and soil by processing data taken by farmers or unmanned aerial vehicle with software using computer visual recognition and in-depth study algorithm. PEAT in Germany insists that a program called Plantix can find the potential defects of soil and subalimentation. They also insist that they have developed a technology that can distinguish pests or diseases in advance by comparing the data of leaves photograph with various data collected by country.

Peat-Mobile

 

Picture2. Plantix from PEAT

– Predictive Analysis: Machine study models based on complex information are being developed to track and predict the various environmental effects of crop yield such as meteorological changes. aWhere in the United States provides service that predicts the presence of diseases and pests in weather information. The weather forecasts are provided locally in a lot of detail suitable for agriculture on this site. Through this information, farmers, crop consultants, and researchers work together to provide high-quality information on agriculture. It provides access to more than a billion agricultural data including temperature, rainfall, wind speed, and solar radiation energy.

vws

Picture3. Agricultural Weather Service from aWhere

When people have a variety of illnesses, they take the right treatments to cure them. For example, people take cold medicine when they have a cold and people take digestive medicine when they have a trouble with their digestion. The same thing goes with plants. A single crop can have a variety of diseases and you should block the disease by providing solutions for each situation or pests that you have now.

식물관련 질병

Picture4. Cases of apple-related diseases

As you can see from the picture above, the single crop(apple) can have a variety of diseases and it doesn’t make sense to solve this with a single insect repellent. However, AI can act as a plant’s physician and allow you to take the best action for your current crop situation.

 

병변사례

Picture5. Cases of the collection of lesion images suitable for artificial intelligence study

Currently, it is very difficult to study artificial intelligence related to agricultural AI & IoT. AI should teach us using the standardization pictures like the picture above. This is the reason why we are in collaboration with many platform companies. These materials will be processed through the process as shown in the picture below.

 

농업 빅데이터 처리과정

 

Picture6. An intelligent process of agricultural Big Data

 

인공지능 학습용 데이터 자동추출

Picture7. Extraction of data for automated artificial intelligence learning

Conclusion.

Beyond basic functions such as supplying water simply by checking the temperature and humidity, opening the vinyl house with the motor, or turning a fan, the current agriculture is evolving into the agriculture with AI applied. The era, that ‘plant’s regular physician’ cares for crops more than the relief of farmers, is coming. I think it is worth to consider review the application of a project and the awareness of the proportion that AI will occupy in IoT.