The central rural work conference has released important signals. Ten ways for AI technology to help agriculture can be expected in the future

The central rural work conference was held in 2021 year 12 month 25 to 26 In Beijing . The meeting analyzed the current “ Agriculture, rural areas and farmers ” Situation and tasks facing the work , Research deployment 2022 year “ Agriculture, rural areas and farmers ” Work , Release two important signals ： The supply of primary products will be more secure ; The integration of urban and rural development within the county has accelerated .

The topic of agriculture is always new . At present , The rapid development of agriculture must be inseparable from the strong support of science and technology . Under the power of science and technology , China's agricultural development is experiencing a new stage of transformation from traditional agriculture to digital agriculture , Developing digital agriculture is 18 The demand for fine management of 100 million mu of farmland , It is also the basis for the implementation of the Rural Revitalization Strategy . Major Internet enterprises and industry leaders continue to hold heavy positions in agriculture , When the technical means of smart agriculture begin to move towards large-scale application , It's time to focus on smart agriculture “ Difficult and correct ” On my way .

according to BI Intelligence Research Forecast report issued , To 2025 In, the world is networking intelligent agricultural technologies and systems （ Including artificial intelligence and machine learning ） Expenditure in this area is expected to triple , achieve 153 Billion dollars .

according to Markets&Markets Published data , Only in agriculture AI Technology and solutions expenditures are expected to be borne by 2020 Year of 10 Billion dollars to 2026 Year of 40 Billion dollars , Compound annual growth rate （CAGR） by 25.5%.

PwC points out , Agriculture based on Internet of things （IoTAg） Monitoring has become the fastest growing technology field in the field of networked intelligent agriculture , Total market value to 2025 It is expected to grow to 45 Billion dollars . 

AI、 machine learning （ML） And IOT sensors , It can provide rich real-time data for the algorithm , So as to improve the efficiency of agricultural production 、 Increase crop yields and reduce food production costs . According to the forecast data of the United Nations on population and hunger , To 2050 year , The global population will increase further 20 Billion , Agricultural productivity needs to be improved 60% To provide enough food . According to the data released by the Bureau of Economic Research of the United States Department of Agriculture , In the United States alone , planting 、 The total market value of processing and food distribution business is as high as 1.7 Trillions of dollars . To 2050 year , Artificial intelligence and machine learning are likely to become the core of new technologies , Help us deal with 20 The expected food demand brought about by the billion new population .

01 AI+ Technical difficulties in agricultural application

Look at the links before, during and after agricultural production , There are different application demand scenarios and professional directions . such as ： Upstream prenatal focus on Biotechnology , Information oriented cultivation 、 Biological breeding 、 Pesticides, fertilizers, etc ; Midstream production is concentrated at the level of digital agriculture , Big data for agriculture 、 Farmland monitoring 、 Agricultural UAV, etc ; And the quality inspection of downstream postpartum at the consumption level 、 Processing of agricultural products 、 Cold chain logistics, etc .

In the long run AI+ In the follow-up report and interview of Agriculture , The technical difficulties of insiders are nothing more than the following two points ：

1、 Agricultural data collection is difficult ;

2、 Each agricultural customer scenario is different , Need customized training . At the same time, agricultural customers and R & D should jointly define the corresponding industry standards .

Of course , This is just a general overview of the problems encountered in most smart agricultural applications . In the actual business promotion , It may also lead to problems to be solved . Previously, a domestic plant factory was in the process of business promotion , Encountered a shortage of agricultural talents 、 Manual confirmation of production is inefficient 、 Unable to promote and apply on a large scale .

some AI The plant plant plant is optimized and upgraded on the development platform , And pest monitoring model 、 Product development such as growth influencing factor model , Digitize the experience of agricultural experts 、 Commercialization , Turn the actions frequently confirmed by experts into machine recognition , It greatly improves the work efficiency of experts ; Through efficient and accurate machine identification , It has the effect of improving product quality and output . Reduce the output of defective products , Increase production 15%, Production materials ( seeds 、 stroma 、 nutrient solution ) Cost reduction 15%.

02「 Agriculture 」—— One of the most promising artificial intelligence and machine learning application scenarios

imagine , In these large farming areas, which usually take hundreds of acres as the basic planning unit , At least 40 Need synchronous tracking 、 Highlight and monitor the basic process . In depth analysis of weather changes 、 Seasonal differences in sunlight 、 Grasp the migration mode of birds and insects 、 Understand the use needs of special fertilizers 、 Choose the right pesticides for the crops 、 Supervise the planting period and irrigation period, etc , For machine learning, it is a major problem that is expected to be solved and of great practical significance . today , Crop production is increasingly dependent on excellent data collection and analysis capabilities .

Because of that , a farmer 、 Cooperatives and agricultural development enterprises have decided to further adopt a data centric approach , And continue to introduce AI And machine learning elements to improve agricultural production and crop quality . Looking at the next few years , The following ten ways are expected to promote the further development of agriculture ：

First of all 、 Using a AI Monitoring system with machine learning , A real-time video source that tracks each crop field , This is used to identify animal or human violations and give an immediate alarm .

AI And machine learning can reduce the accidental destruction of crops by livestock or wild animals 、 Or the possibility of breaking into farms in remote areas . With AI With the rapid development of machine learning algorithms in the field of video analysis , Every agricultural production participant can take this opportunity to protect their fields and agricultural facilities .AI With machine learning, the video surveillance system can be easily extended to adapt to large-scale agricultural operations , Focus on the whole farm . Over time , We can program or train the monitoring system based on machine learning , Teach them to identify people and vehicles .

As a one-stop shop AI The leader of the platform , Shenyan technology has proved that these technologies can effectively protect remote facilities with practical actions 、 Optimize crop production and identify unexpected intruders in the field through machine learning . The following figure shows an example of real-time monitoring in the agricultural field ：

chart ： rely on AI Identify people and vehicles with machine learning algorithms , It can help the global agricultural enterprises to simplify the remote operation process .

second 、AI With machine learning —— Through UAV real-time sensor data and visual analysis data , Improve crop yield forecasts .

With the real-time video stream provided by intelligent sensors and the data captured by UAVs , Agricultural experts have access to new data sets that they have never had access to before . Now , Researchers can combine water 、 Sensor data such as fertilizer and natural nutrient levels analyze the changing growth patterns of each crop over time . Machine learning is responsible for integrating a large number of data sets , Take recommendations based on constraints to optimize crop yields . Shown below , by AI、 machine learning 、 Field sensor 、 Examples of scenes where infrared images are combined with real-time video analysis techniques , Farmers can gain new insights into improving crop health and yield per mu ：

chart ： The fact proved that , UAV has become a very reliable platform , Be able to collect information about specific fertilizers 、 Data on the impact of irrigation methods and pesticide treatment methods on the actual yield of crops .

Third 、 Yield mapping is an agricultural technology , By monitoring machine learning algorithms , Find patterns from large-scale datasets and understand the orthogonality between different patterns in real time , This brings immeasurable value to crop production planning .

today , We have been able to do this before the planting cycle begins , To roughly determine the potential yield of a particular field . By combining machine learning technology with 3D mapping 、 Combination of sensor data and field color data based on UAV , Agricultural experts can quickly predict the yield of a particular crop under potential soil conditions . These data sets captured by UAVs are accurate and reliable . Shown below , Results obtained for yield mapping analysis ：

chart ： With the support of supervised and unsupervised machine learning algorithms , Agricultural experts were able to determine how to maximize the yield of the field .

Fourth 、 The United Nations 、 International institutions and large-scale agricultural projects , They have combined UAV data with field sensors , To improve pest management .

By combining the infrared thermal imager data of UAV with the sensor that can monitor the relative health level of plants , The agricultural management team can AI Make prediction and identification before the occurrence of pests with the help of . at present , The United Nations will cooperate with PwC to assess the potential pest infestation in palm plantations in Asia , As shown in the figure below ：

chart ： The United Nations combines field sensors with UAV data , To tune machine learning algorithms 、 Help farmers get higher yields from plantations .

The fifth 、 Now , There is a serious shortage of agricultural workers , Make based on AI Intelligent tractor with machine learning 、 Agricultural robots and other intelligent machines , Become the first choice for agricultural planting in remote areas .

at present , Large agricultural enterprises cannot find enough employees , Can only rely on Robotics to collect crops from hundreds of acres , At the same time, it also gives a positive impetus to the security situation in remote areas . By programming autonomous robot devices , They can spread fertilizer for crops 、 This reduces operating costs and further increases field production . At present, the complexity of agricultural robots is rapidly increasing , The following figure shows the dashboard information of the robot during operation .

chart ： The fact proved that , Agricultural robotics can quickly capture valuable data , Use this to tune AI And machine learning algorithms , So as to further improve crop yield .

The sixth 、 By removing a series of traditional obstacles , Emerging technologies are expected to deliver more freshness to the market 、 Safer crops , And greatly improve the traceability of agricultural supply chain .

2020 The outbreak of the COVID-19 in accelerated the deployment of tracking and traceability functions in the agricultural supply chain ,2021 This trend will remain stable in . This well managed tracking system can provide greater visibility , Comprehensively improve the overall control over the supply chain , To effectively reduce inventory . The latest tracking system can even distinguish the batches of incoming goods 、 Belongs to the project and implements fine-grained records at the container level .

Besides , With RFID And the rapid popularization of IOT sensors throughout the manufacturing process , At present, most advanced tracking systems also begin to rely on advanced sensors to obtain more status information about each batch of goods . Wal Mart is promoting a pilot project , To study how to use RFID Simplify the tracking performance of the distribution center , And improve the efficiency to that of manual operation 16 times .

The seventh 、 With the help of AI Combine with machine learning to optimize the correct mixing ratio of biodegradable pesticides and use only when necessary , Thus reducing operating costs and increasing unit field output .

By combining intelligent sensors with UAV visual data flow , Agriculture AI The application can now detect the areas with the most serious diseases and pests in the planting area . On this basis , Then use supervised machine learning algorithm , Agricultural experts can determine the best combination of pesticides , Effective control of pest threat 、 Prevent its further spread and infect other healthy crops .

The eighth 、 Determine the total yield according to the unit yield of crops , So as to formulate a reasonable and effective crop pricing strategy .

Accurately grasp the harvest rate and quality level of crops , Help agricultural enterprises 、 Cooperatives and farmers better develop pricing strategies . Considering that the overall market demand for specific crops is basically constant , Each party can choose a fixed selling price according to the crop harvest 、 Strategies such as unified selling price and flexible selling price . These figures alone , It can eliminate millions of dollars of losses for agricultural enterprises every year .

The ninth 、AI It can help farmers find the leakage point in the irrigation system , Optimize system performance and measure how to adjust irrigation frequency to improve crop yield .

In many parts of North America , Water is one of the most scarce resources , It even directly determines the life trend of the whole community that lives on agriculture . Efficient use of water resources , Maybe it can turn a farm into a profit 、 Coming back to life . Through linear programming , We can quickly calculate the optimal amount of water required for a particular field or crop to reach the desired yield level . Supervised machine learning algorithms ensure that fields and crops get enough moisture to optimize yields , But not to waste this precious resource too much .

The first ten 、 Monitor and maintain the health of livestock —— Including life weight 、 Daily activity level and food intake —— Has become a AI And machine learning .

Ensure good care for livestock for a long time , We must keep abreast of the actual reactions of various types of livestock to the current diet and living conditions . utilize AI And machine learning technology , Agricultural experts can understand what determines the mood of cows , And improve the milk production of dairy cows through appropriate adjustment . For the animal husbandry industry dominated by cattle and other livestock , The intervention of emerging technologies has brought unprecedented new directions for ranchers to open up new profit margins .

Summary

at present , The deep integration of artificial intelligence and agriculture in China still faces multiple challenges , such as , Rural network infrastructure is weak 、 Applied to agricultural technology is still in the basic stage 、 The research and development of artificial intelligence agricultural robot is not mature , There will be more or less problems in the process of putting into use , This requires the relevant departments to start from the infrastructure 、 Technology supply 、 Industrial demand and other aspects , Comprehensively promote the deep integration of artificial intelligence and agriculture , Explore the effective path of high-quality development of modern agriculture . In terms of support , Focus on strengthening the construction of rural network infrastructure and agricultural information service platform ; In terms of technology supply , Continuously improve the supply level of artificial intelligence technology in the agricultural field ; In terms of industrial demand , Vigorously cultivate farmers' willingness and ability to apply artificial intelligence , Continuously carry out technical guidance and popularization of relevant knowledge .

Take Shenyan technology as an example , With its exquisite algorithm technology , adopt “ Soil testing 、 Intelligent fertilizer 、 New fertilizer 、 Crop yield estimation 、 Intelligent drip irrigation 、 Pest control 、 Digital service platform ” The development of the industrial chain has opened the door to the future , Towards agricultural digitalization 、 Intelligent 、 Unmanned .

We believe that , In the near future , With the continuous development of artificial intelligence technology , Its large-scale application in the field of agriculture will finally realize .

｜ About Deep extension technology ｜

Shenyan technology was founded in 2018 year , It's Shenlan technology （DeepBlue） Its subsidiaries , With “ Artificial intelligence enables enterprises and industries ” For the mission , Help partners reduce costs 、 Improve efficiency and explore more business opportunities , Further develop the market , Serving the people's livelihood . The company launched four platform products —— Deep extension intelligent data annotation platform 、 Deep extension AI Development platform 、 Deep extension automatic machine learning platform 、 Deep extension AI Open platform , It covers data annotation and processing , To model building , And then to the whole process service of industry applications and solutions , One stop help enterprises “AI” turn .