KR20200029657A - Farming automation system using crop image big data - Google Patents

Abstract

The present invention relates to a farming automation system using crop image big data, a farming automation facility comprising a growth sensing module that transmits temperature, humidity, sunshine, and weather sensor data and a camera module that takes and transmits an image of a crop; A data collection module for continuously collecting and storing various sensor data and crop images output from the agricultural automation facility is formed, and the data collected in the data collection module is compared with the built-in crop growth database to be converted into big data by artificial intelligence. A smart farming automation server including an artificial intelligence module that is formed and stored; When the image of a crop is collected from the data collection module as included in the smart farming automation server, the artificial intelligence module is driven to output a growth management control signal of the crop, and comprehensive control information is produced with each collected sensor data. Consisting of an integrated control module that transmits to a worker's computer or a smart phone; when an agricultural automation facility is operated while taking an image of the crop and transmitting it, the algorithm of the artificial intelligence module is driven to drive the temperature, humidity, carbon dioxide, geothermal temperature of each crop, It is characterized in that soil moisture, sunshine, control, fertilization, crop moisture supply, oxygen supply, and sunshine are optimally derived so that the crop growth environment is automatically managed, and big data is updated by the learning algorithm of the AI module.

Description

Farming automation system using crop image big data}

The present invention relates to a farm automation system using crop image big data, IPC SECTION G physics; G06 Arithmetic logic calculation, calculation, coefficient; G06Q Data processing systems or methods particularly suitable for management, commercial, financial, managerial, supervisory or predictive purposes; Systems or methods particularly suitable for management, commercial, financial, management, supervisory or predictive purposes, not elsewhere classified; G06Q 50/02 It belongs to agriculture, fishing and mining.

In recent years, cultivation of facilities has been particularly popular due to the rapid changes in the agricultural environment.

Facility cultivation is an intensive production method of science and technology and capital.It provides comprehensive knowledge of plant physiology to maintain optimal conditions for crop production, physics and engineering for construction and environmental control, and chemistry of nutrient solution and soil. It is required, and a lot of capital is required to equip not only the construction of facilities, but also the equipment for environmental control and vitalization of work.

Environmental management of greenhouses must be done comprehensively from the viewpoint of greenhouse management, and the synthesis as such appears in the form of plant factories.

The plant factory is a symbolic existence of the agricultural production system. As a next-generation agricultural production system that deploys mass production systems of plants that apply biological knowledge elucidated in the field of biotechnology, great expectations are coming to the future society.

The production technology in the so-called artificial plant factory, which creates the optimum conditions for a high-tech greenhouse environment and introduces automation, robotization, and information, is an important requirement for new agricultural production in the coming future.

In the case of controlling the environment in a facility with a complex environmental control system that has been researched and developed to date, it has been common to individually control individual factors such as temperature, humidity, soil moisture, carbon dioxide gas, and light intensity.

In addition, the conventional greenhouse focused on room temperature management of crops, but in the future, it will focus more on the management of temperature, humidity, carbon dioxide, geothermal temperature, soil moisture, sunshine, control, fertilization, etc. by each growth stage of the crop. Research and development on intelligent greenhouse systems are underway in the sense that the role for control must be fulfilled, and that agricultural productivity is maximized through automation and unmanned, the final goal of facility cultivation.

The intelligent greenhouse system is composed of a greenhouse body, a power unit, a control module unit, a monitoring unit, a soil unit, and various sensors and controllers for complex environmental control. You can drive according to your requirements.

In addition, real-time monitoring of sensing data such as indoor temperature, humidity, outdoor temperature, humidity, geothermal temperature, CO ₂ concentration, soil moisture, light quantity, wind speed, rainfall, etc., automatically controls various controllers according to the growth environment of each stage of crop growth. , We are trying to automate environmental management in facility cultivation.

Since these efforts and researches on automation will lead to the expansion and development of next-generation agricultural management, it can be considered that it is the basis for high income, high quality, and multiple expansion.

(A Study of the Growth Environment Control of the Crops using an Intelligent Greenhouse System Chang-Su Kim, Graduate School of Industry, Ulsan University, Mechatronics, cited August 8, 2006)

In such cultivation of facilities, the development of automation software and automation systems that control them must be performed in parallel.

Accordingly, recently, as the Internet, short-range wireless communication technology, image processing and sensing technology have been developed, ICT (Information & Communication Technology) complex technology has been applied to crop cultivation and growth monitoring systems.

In addition, a greenhouse monitoring and management system has been developed to manage the growth environment of crops, and not only monitoring the growth environment of crops, but also measuring the growth change of crops and observing the growth process to analyze the state of crops.

However, the conventional crop image processing technology depends on a limited amount of data, so the amount of data is limited to analyze by image processing, and the recognition and measurement method of each crop institution is a method of directly measuring the crop organ using the pots of individual crops. , There is a problem that requires a separate conveyor system or an imaging device and a measuring device for image acquisition.

In addition, in order to acquire images by crop organs, consistent images have to be collected, but since the height changes with the growth of crops, there is a problem that requires constant management of the photographing apparatus.

In addition, in image recognition after photographing, when crops densely clustered are clustered, if crops overlap, it is difficult to accurately recognize crop organs, and current measured crop data is collected to predict crop growth or predict crop conditions. There is a great need for a diagnostic system.

As a prior art for the automation of agricultural crops, the production management system and method of the agricultural products are disclosed in published patent publication No. 10-2002-0062091, and the comprehensive management system of environmentally friendly agricultural products and processed products is disclosed in published patent publication No. 10-2009-0001527. It was released.

However, the above techniques are made by collecting information of agricultural producers, and there is a problem that it is difficult to collect and manage information about each agricultural producer.

Accordingly, as a technique for performing collective management of agricultural products through joint production, a collective management, processing, sales method and system of agricultural products through a wired or wireless network is disclosed in Patent No. 10-1474603.

However, the conventional technology including the above technology may cause a problem in that the cultivation time (temperature) for the cultivar is not appropriate even if the type of agricultural product (variety) required in the market is identified. Problems that do not occur at the right time may occur.

In the system and method for diagnosing crop growth through video image registration number 10-1832724, a technique for collecting crop images and performing growth diagnosis is introduced. According to this, the collection amount of crop images is limited and it is difficult to operate the comparison algorithm. Therefore, there is a problem in that accurate image data is required each time.

(0001) Published Patent Publication No. 10-2002-0062091 Production management system and method of agricultural products (0002) Patent Publication No. 10-2009-0001527 discloses a comprehensive management system for eco-friendly agricultural products and processed products. (0003) Collective management, processing, sales method and system of agricultural products through wired / wireless network in Patent No. 10-1474603 [0004] Registration No. 10-1832724 diagnostic system and method for growing crops through video images

The present invention is to solve the above problems, the growth environment monitoring system of a crop that can be applied to a greenhouse monitoring and management system that can manage the growth environment of a crop is measured as an image image and a growth process is measured. Analyzing the state of crops by observing them, and making this into big data using artificial intelligence algorithms, to provide 'a farming automation system using crop image big data' that enables various image analysis and accurately recognizes images by crop institutions. The purpose.

In order to achieve the above object, the present invention includes a farming automation facility comprising a growth sensing module for transmitting temperature, humidity, sunshine, and weather sensor data and a camera module for taking and transmitting an image of a crop; A data collection module for continuously collecting and storing various sensor data and crop images output from the agricultural automation facility is formed, and the data collected in the data collection module is compared with the built-in crop growth database to be converted into big data by artificial intelligence. A smart farming automation server including an artificial intelligence module that is formed and stored; When the image of a crop is collected from the data collection module as included in the smart farming automation server, the artificial intelligence module is driven to output a growth management control signal of the crop, and comprehensive control information is produced with each collected sensor data. Consisting of an integrated control module that transmits to a worker's computer or a smart phone; when an agricultural automation facility is operated while taking an image of the crop and transmitting it, the algorithm of the artificial intelligence module is driven to drive the temperature, humidity, carbon dioxide, geothermal temperature of each crop, Crop image characterized by optimizing derivation of soil moisture, sunshine control, fertilization, crop moisture supply, oxygen supply, and sunshine, so that the crop growth environment is automatically managed, and updating the big data with the learning algorithm of the AI module The agricultural automation system using big data is the technical point.

The artificial intelligence module of the present invention is preferably a farming automation system using crop image big data, characterized in that big data is formed using a convolutional neural network (CNN) algorithm.

In addition, the AI module analyzes the captured crop image to recognize the crop organs including leaves, flower pots, nodes, and fruits, and uses the recognized crop organ data to determine the size of the leaves, the length of the growth points, and the thickness of the growth points. , It is desirable to be a farming automation system using crop image big data, characterized in that growth data including fruit size, plant water, and fruit number are derived and stored, and the derived data for each institution is included in the big data. .

In addition, the data collection module of the smart farming automation server receives and stores crop images and measured plant-specific data taken from the smart terminal so that crop growth data and crop recognition image information big data for each farm is built and crop images. It is desirable to be a crop growth diagnosis system characterized by improving accuracy to become learning data of the AI module.

According to the present invention, the growth environment monitoring system of a crop that can be applied to a greenhouse monitoring and management system capable of managing the growth environment of a crop is measured by an image image and the growth process is observed to observe the state of the crop. By analyzing and making this into big data using artificial intelligence algorithm, the image analysis is variously and accurately carried out, and the advantage of providing 'a farming automation system using crop image big data' that enables the quality of crops to be judged by the image of each crop institution is provided. have.

1 is a structural diagram of the present invention
Figure 2 is a database generation diagram of the artificial intelligence module of the present invention
Figure 3 is an exemplary diagram of crop image inference by using the big data of the artificial intelligence module of the present invention
Figure 4 is an illustration of a crop database classification image of the present invention

Hereinafter, the present invention will be described with reference to the drawings, and when it is determined that a detailed description of known technologies or configurations related to the present invention may unnecessarily obscure the subject matter of the present invention, the detailed description will be omitted. will be.

In addition, terms to be described later are terms defined in consideration of functions in the present invention, which may vary depending on a user's or operator's intention or practice, and thus the definition should be made based on the contents of the present specification describing the present invention.

The following Figure 1 is a structural diagram of the present invention, Figure 2 is a database generation diagram of the artificial intelligence module of the present invention, Figure 3 is an exemplary diagram of crop image inference by utilizing the big data of the artificial intelligence module of the present invention, Figure 4 Is an exemplary view of the crop database classification image of the present invention.

The present invention relates to a farming automation system using crop image big data, as shown in FIG. 1, a farming automation facility 10 comprising a growth sensing module 100 and a camera module 120; It is composed of a smart farm automation server 20 including a data collection module 210, a crop growth database 220, an artificial intelligence module 230, and an integrated control module 200.

The farming automation facility 10 of the present invention includes a growth sensing module 100 that transmits temperature, humidity, sunshine, and weather sensor data, and a camera module 120 that takes and transmits an image of a crop.

In general, the farming automation facility 10 is also called facility cultivation, and it is equipped with facilities for maintaining optimal conditions for the production of crops, with an intensive production method of science and technology, and automation for environmental control and vitalization of work. It is equipped with a device.

The automation device is a complex environmental control system that has been researched and developed to date, and is equipped with various sensors that perform environmental control in the facility, and individually controls individual factors such as temperature, humidity, soil moisture, carbon dioxide gas, and light intensity.

In the present invention, the farming automation facility 10, unlike the conventional crop room temperature management, is a complex environmental control device that controls and controls the temperature, humidity, carbon dioxide, geothermal temperature, soil moisture, sunshine, control, fertilization, etc. for each future growth stage of the crop. It is to construct.

To this end, the agricultural automation facility 10 comprises a greenhouse house, a power unit, a control module unit, a monitoring unit, various sensors and controllers for complex environmental control with the soil unit, and can be switched manually or automatically. It is designed to enable automatic operation according to external control commands.

The growth sensing module 100 provided in the agricultural automation facility 10 of the present invention provides real-time sensing data such as indoor temperature, humidity, outdoor temperature, humidity, geothermal temperature, CO ₂ concentration, soil moisture, light quantity, wind speed, and rainfall. As a set of various sensors to be monitored, the smart farming automation server 20, which will be described later, uses the information collected by the growth sensing module to produce a command to automatically control various controllers according to the growth environment of each crop growth stage. .

The camera module 120 provided in the farming automation facility 10 of the present invention always shoots the growth state of the crop during the automatic operation of the farming automation facility 10 and transmits it to the smart farming automation server 20 described later. .

The smart farming automation server 20 of the present invention is formed with a data collection module 210 that continuously collects and stores various sensor data and crop images output from the farming automation facility 10, and the data collection module 210 It includes an artificial intelligence module 230 that compares the collected data and the built-in crop growth database 220 to form and store it as big data in an artificial intelligence method.

The data collection module 210 is connected to the farm automation facility 10 through the Internet and receives and stores various sensor data and crop images.

The data collected by the data collection module 210 is used as a crop growth database 220.

The artificial intelligence module 230 of the present invention is a processor for forming big data using the crop growth database 220. In the present invention, the collected database is used by using a convolutional neural network (CNN) algorithm. Form big data.

Big data is a technology that extracts value and analyzes results from data that includes large (several terabytes) of structured data even beyond the capabilities of existing database management tools, or even unstructured data sets that are not in the form of databases.

In the present invention, as the storage amount of the existing crop image database increases, it is possible to analyze growth information by collecting atypical aggregate signals.

Therefore, the use of big data in the crop image can produce predictable information of the growth state even when the crop image is only a part of the crop, for example, a part of the leaf.

The larger the data amount of the crop image according to the present invention, the more accurate growth information can be extracted with a random crop image.

For example, as illustrated in FIG. 3, the artificial intelligence module 230 of the present invention can infer the flat image of the Angelica leaf by using the big data even if the Angelica leaf is not photographed in the correct plane.

Convolutional Neural Network (CNN) is a computer model or computer equipment designed to demonstrate or simulate the ability of the brain to recognize and identify multilayer perceptrons designed to use minimal preprocessing. ).

CNN consists of one or several convolutional layers and common artificial neural network layers on top of it, and additionally uses weights and pooling layers.

Thanks to this structure, CNN can fully utilize the input data of the two-dimensional structure.

Compared to other deep learning structures, CNN shows good performance in both video and audio fields.

CNNs can also be trained through standard reverse delivery.

According to the use of CNN in the present invention, even as an inaccurate crop image as described above, the larger the amount of data, the more accurate the crop image can be inferred.

CNNs are easier to train than other feedforward artificial neural network techniques and have the advantage of using fewer parameters.

In recent deep learning, a Convolutional Deep Belief Network (CDBN) was developed.It is structurally very similar to the existing CNN, so you can use the two-dimensional structure of the picture well, and at the same time, Deep Belief Network , DBN). CDBN provides a generic structure that can be used for a variety of image and signal processing techniques and is used in several benchmark results for standard image data such as CIFAR.

In the present invention, the smart farming automation server 20 includes an integrated control module 200 as shown in FIG. 1, wherein the integrated control module 200 collects crop images from the data collection module 210. When the artificial intelligence module 230 is driven, the crop growth control signal is output, and comprehensive control information is produced with each collected sensor data to be transmitted to the worker computer 30 or the smart phone 40.

According to the present invention configured as described above, when an agricultural automation facility 10 is driven to take and transmit an image of a crop, the algorithm of the artificial intelligence module 230 is driven to temperature, humidity, carbon dioxide, and temperature of each crop growth stage. , Soil moisture, sunshine amount, control, fertilization, crop moisture supply, oxygen supply, sunshine yield.

Accordingly, the optimal crop growth environment is automatically managed, and the big data is updated by the learning algorithm of the artificial intelligence module 230.

The optimal crop growth environment is bi-directionally communicated to the worker computer 30 or the smart phone 40 through the integrated control module 200.

In the present invention, the artificial intelligence module 230 analyzes the crop image photographed as shown in FIG. 2 to recognize a crop organ including leaves, flower pots, nodes, and fruits, and uses the recognized crop organ data. From the photographed image, the growth data including the leaf size, the length of the growth point, the thickness of the growth point, the size of the fruit, the number of plants, and the number of fruits is derived and stored.

The crop image is inferred as an accurate crop image screen by the CNN algorithm and managed as big data even though an incorrect image or a part of the image is collected in the camera module 120 as shown in FIG. 3.

In addition, the algorithm of the artificial intelligence module 230 is driven to evaluate the growth status of the crop in the crop image, and the temperature, humidity, carbon dioxide, geothermal temperature, soil moisture, sunshine amount, control, and fertilization step by step suitable for the current growth status of the crop. , Analyze the moisture supply amount, oxygen supply amount, and sunshine amount of the crop, derive an optimized environment, and output a control signal for growth management of the crop driving the agricultural automation facility 10.

Hereinafter, growth information usable for big data as a crop of interest of the present invention will be described.

Example 1: Angelica

Although Angelica has direct cultivation and seedling transplantation, if it is harvested in the same year by direct directing, the quality of the meat is excellent, but the yield is low and the content of medicinal ingredients is low. In the second year, when the direct wave, the flower stand rises, and the root becomes woody and cannot be used as a medicinal material. And the method of cultivating seedling transplantation is advantageous for effort and land use.

Sowing time: Sowing in late autumn or early spring after the ground is loose before the ground freezes, sowing in autumn is good. Fall sowing is from late October to early November, and spring sowing is from late March to early April or mid-April. In spring sowing, seeds must be buried immediately after harvesting seeds in the fall, so germination inhibitory substances in the seed can be removed and germination is good.

Seedling and sowing of seedlings: 1.2 ~ 1.5m wide dukes are made of seedlings or loams that are neither fertile nor barren, and scattered in autumn or lined at intervals of 5 ~ 10cm to be transplanted in the following spring. The smaller the number of seedlings for transplantation, the lower the selection rate. For the production of medium seedlings or seedlings for transplantation, the amount of seeding is about 50g per 1m2, and about 750 ~ 950g of seed is needed to produce seedlings to be cultivated in an area of 10a. 15 ~ 19m2 (about 4.5 ~ 5.7pyeong), and 20m2 (about 6pyeong) of area is required for line mulching.

Very planting: from late March to late April. Early planting has good roots, while late planting has less flower stalks. The set meal is good on a cloudy day without wind. Covered with wet sawdust and planted while planting, the drying rate of the seedlings is low and the rate of survival increases.

Diseases: In the dwarf ear, a mycobacterium disease usually occurs, but when it rains a lot, it can occur when the water does not drain well, so it is good to repair the drainage. Mottled disease occurs on the leaves, and it occurs in the rainy season in summer when the temperature is high and humid. Symptoms of the disease appear as brown spots at first, and when developed, enlarge to brown to dark brown irregular lesions, and the inside of the lesions may tear or puncture. The disease is severely pearled, and registered noticed drugs such as azoxystrobin hydrating agent are sprayed at the beginning of the onset. In addition, pay attention to the brown spots disease that occurs when there is a lot of rain or when the humid period continues for a long time, and the stem rot disease that occurs in May to August.

Insects: The main insects include spotted mite, which occurs frequently during creeping, and root knot nematodes, which reduce the quality and reduce the quantity by making humps on the roots and sucking the juice. To control mottled mite, use registered and registered drugs such as azocyclotin hydrating agents, but use different types of drugs alternately to prevent resistance to drugs, and to control nematodes, crop and crop with crops.

When the crop image of the Angelica thus constructed is collected, the artificial intelligence module classifies the Angelica from the leaves and outputs the growth information database.

The artificial intelligence module 230 outputs key characteristics, efficacy, planting time, pests, and pest information of Angelica and produces breeding information in comparison with the collected Angelica image, and transmits it to a farm automation facility to perform automatic farming.

At this time, since the crop image is automatically photographed and transmitted, an intact leaf of the Angelica can not be automatically transmitted, but since the comparison data is formed of big data, an accurate crop image can be inferred and compared.

In addition, when it is determined that the Angelica is the leaf image, the big data may infer and output the growth state of the root with the leaf image.

Example 2: Astragalus

Astragalus is cold and can be grown anywhere in the country, but the southern coastal regions with heavy rainfall and heavy weather are not suitable, and growing in a relatively cool mid-northern mountainous region, that is, in summer where the temperature is not too high and where there is a large daily difference is the root growth. Good and good quality. In particular, it is suitable for cultivation of soil with deep soil, good water drainage, and a lot of humus because the roots extend deeply as rectilinearity.

  In soils with poor drainage, the roots tend to rot when rainy summer comes, and in sandy soils, the growth of foliage is good, but the roots do not go down straight and many roots are formed, making it difficult to harvest good quality. This excess soil should also be avoided. In the plains where the temperature is high in summer, the root disease (root rot) is so severe that it is difficult to produce roots for 2-3 years. The breeding of astragalus mainly uses the seed propagation method.

Varieties: There is a “Pungseong Hwanggi” developed by a research institute affiliated with the Rural Development Administration.

Chaejong: The seeds of Hwanggi should be collected from healthy abandonment of 2 ~ 3 year olds. In particular, the old seeds have poor germination rate, poor growth condition, and death, so be careful, but the seed color is blackish brown and shiny, and the seeds with faithful seed weight (1L weight of more than 750g) are selected. Since the flowering period lasts more than 60 days, it is difficult to catch the cultivation time. In the early period of mid-October to mid-October, before the frost falls, the stems are cut at a height of 20 cm, tied up into small columns, dried, and then selected. When purchasing commercially available seeds, choose one that is black, shiny, heavy, and faithful.

-Jeongseon: The seeds should be carefully selected so that weed seeds such as soybean or new ginseng do not mix.

-Cultivation: Sowing in the first year (early to late April)-> Yeotum, Kim Mae-gi (late April to mid-May)-> Chongqing, Laughing (mid-June to mid-July)-> Baeto (late July) ~ Mid-August)-> Harvest of the same year (mid-November ~), and in the second year, laughter and chongqing (mid-March to mid-April)-> Red Soak (mid-May to early June)-> Laughing, Chongqing (From early June to late June)-> Soaking (mid-July to early August)-> Chaejong (early to mid-October)-> Harvest (from late October to mid-November).

Sowing: In the central and southern plains, early to mid-April (adjusted by region to avoid damaging damage). If the sowing period is delayed, the growth will fall and the quantity will decrease, so be careful not to be too late.

-Sowing method: Before sowing, fertilize the entire floor and make a dug of 90 ~ 120cm. When harvesting for the same year, sow at intervals of 15 cm between goals and 10 cm between aerations, or 2 to 3 tablets at 10 cm intervals. When harvesting more than 2 years, small bones are made at intervals of 30 cm, soaked into 10 cm rksruir between aerations, or blasted at intervals of 10 cm. After sowing, the cover is 0.5 ~ 1cm thick, but using a man-made planter and a multi-purpose manure planter with a tractor, it is convenient and can save manpower, and the seeding depth is uniform, so that germination is even.

Seed Requirement: At the time of harvesting this year, it takes about 3.6L / 10a for sowing and 2.4L / 10a for crushing. When harvesting for more than two years, 1.8L / 10a is required.

Fertilization method

① Since it is a legume food (it has nitrogen fixation ability), it should be given more compost, phosphoric acid, and potassium fertilizer than nitrogen, and in acidic soil, it should be sufficiently neutralized and planted.

② Usually, 6kg of nitrogenous fertilizer per 10a, 8kg of phosphate fertilizer, 9kg of potassium fertilizer, and 1,000kg of compost are given as bases.

③ If the stems of the late autumn are dried yellow, leave about 10cm above the ground to harvest the next year, and then cut and overwinter.

④ In the early spring, fertilizers such as compost should be given about 30% more than the base manure for good growth for the second year.

Main field management: Germination takes place in about 10 days after sowing, but only the excessively cut areas are removed. Raising a bit of astragalus is good because the side roots are less likely to occur. For the grinding work, make the gap between the aerations to 10 cm, leaving only one aeration and scoop it up. Kim Mae-gi sometimes sprays the soil after covering the yellow seedlings.

① Bopa: Hwanggi is a straight-rooted crop that can't be transplanted well, and it needs to harvest straight roots. Normally, a seed that does not germinate after 20 days is sown.

② Sunjiri (wetting): If the growth of the upper part is too good, there is a fear of overcoming (falling). The first year is cut before mid-July, and the second year is cut in quarters in late June and late July, respectively. If it is cut too much, there is a problem with growth, and the quantity decreases.

③ Drainage: In the rainy season during summer, if the water level rises or becomes over-humidified, the roots will rot. Usually, the depth of the drainage channel should be as high as 40cm or over and 80cm or less.

Disease pests and preventive measures

Diseases: The main diseases are old-growth disease, wearing disease, and powdery mildew. As a severe disease during the summer rainy season, 3 items other than dimethomorph hydrating agent are added to Nokyun disease, 2 items other than azocystrobin liquid hydrating agent to powdery mildew disease, and hymexizol metaraxyl liquid is registered and registered to warehouse workers.

Insect: The main insects are aphids and nightworms. Aphids occur from May to October, especially during the dry season. It is sprayed with the registered aphid agent for 3 to 4 items such as acetamiprid hydrating agent. Also, beware of nightworms, slugs, and other soil pests, which can lead to colonization and reduced marketability. It can be controlled with soil pesticides such as Etope granules and Tabor granules, but registration has not been announced.

Example 3: Jihwang

There are two methods of breeding rhizomes: seed propagation and seed roots, but seed propagation is used for breeding methods to improve new varieties, and is usually used as a root root method for cutting and cutting rootstock to a size of about 5 to 6 centimeters. After harvesting, pick a thick one for medicinal use, and pick a thin one and use it as a seed root.

For fine long roots, the thinness of about 6 mm is good for growth and quantity, and for long roots with a thickness of 1 cm or more, the number of flower stalks increases and the thickness is too thin, and the small roots have slow growth of roots, so the quantity decreases. After selecting the good seed root, dry it for about a day, dig a pit, store it, and dig it for the next year. The planting period is from the end of April to the beginning of May. The method of planting the roots of the lateral roots laterally sprouts and shortens the planting time. This planting is advantageous because of its high quality and quantity. The amount of longitudinal root per 10a is about 60kg.

Making the main field: As it is an herb that requires a lot of manure, in the autumn before planting, spray 2,000 tons of manure, 50 kg of manure, and 50 kg of organic fertilizer evenly in autumn, and then cultivate it 2-3 times, and then plant it and plant it at the time of planting. Should use The field is made of mounds of about 100 centimeters, well selected, and planted with 30 centimeters between goals and 10 centimeters between aerations has a large number of roots and a high product ratio. Above all, it should be well drained, so it is necessary to select the soil well and to increase the height of the weirs by at least 50 cm.

Planting the garden: Before planting, dig the roots, and if there are any rotten parts, cut them off and plant them about 6 to 9 centimeters long. The cut must be smeared and cut with a hand or bamboo knife. When planting, place one root or two small ones at intervals of 12 to 15 centimeters apart, cover the soil about 2 to 2.5 cm, press it with a hoe or the like, and cover it with straw. Currently, it is mainly grown in plastic coating.

Main management method: It is an herb that needs to be planted shallowly, so when weeds are overgrown, there is a high risk of harming the roots. Also, after weeding, spread grass or straw immediately to prevent weeds from growing. In addition, since the hypertrophy of the roots is slow when the flower stand is raised, remove the flower stand from time to time. The method that is currently used a lot is to cultivate it by covering it with black plastic.

Rainforest cultivation: Farming in some areas has good growth, a lot of quantity, and rainforest house cultivation is carried out to prevent bacterial diseases from moving through the soil by splashing on the back of the leaves during the rainy season in the middle of summer and preventing moisture damage. There are many cases. After the rainy season, the rain-covering vinyl is removed.

Yield: 600 ~ 1,200 kilo / 10a / 1 year as raw root, 200 ~ 300 kilo / 10a / 1 year as dry root

Prevention and control of pests and diseases

Disease: The main disease is root disease (root rot). It occurs in high temperature and humid conditions between the end of July and the beginning of September. It withered during the day and becomes alive at night, and dies after about a week. The bottled aeration is pulled out and burned, and the entire package is sprayed with Daisen M-45400. Nitrogenous fertilizers should not be overdosed, and rooting disease is more likely to occur in crop fields and humid packaging, so avoiding crops and taking care of drainage.

Insect: There is damage from blue worms and casterworm larvae, but the damage is not severe.

The present invention uses this growth information to produce an artificial intelligence control signal, wherein the artificial intelligence module 230 analyzes the captured crop image to recognize and recognize crop organs including leaves, flower pots, nodes, and fruits. Using the data of each crop organization, growth data including leaf size, growth point length, growth point thickness, fruit size, flower number, and fruit number are derived from the captured image and stored.

For example, when analyzing the growth image of Angelica as shown in FIG. 4, flower garden (A), leaf (B), stem or node (C), root (D), and root (E) as crop plants of interest You can sort and accumulate data by crop institution.

On the other hand, in the present invention, since the integrated control module 200 and the worker computer 30 or the smart phone 40 are capable of two-way communication, the image accuracy and data of the crop are directly transmitted to the smart phone 40 to improve data accuracy. It can be enhanced or connected to a cloud server on the Internet to collect and store big data information.

Accordingly, the data collection module 210 of the smart farming automation server 20 receives and stores crop images and measured plant-specific data taken from the smart phone 40 to store crop growth data and crop recognition images for each farmer. Information big data can be built, and crop image accuracy can be improved to become learning data of the AI module.

Therefore, according to the present invention, it is possible to form a system for monitoring the growth environment of a crop that can be applied to a greenhouse monitoring and management system capable of managing the growth environment of a crop.

According to this, as shown in FIG. 3, the growth change amount of the crop can be measured by image image, the growth process is observed, the condition of the crop is analyzed, and the quality of the crop can be evaluated and output as excellent, excellent, average, and below. .

That is, as shown in FIG. 3, according to the present invention, an actual photographed image is big-data using an artificial intelligence algorithm, and this is fed back to analyze the image to derive an accurate image from the photographed image and derive the derived image. Again, it can be compared with big data to determine and output the current growth status.

Accordingly, according to the present invention, when the current growth state of the crop is judged as the best, excellent, average, or below, the temperature, humidity, carbon dioxide, geothermal temperature, soil moisture, sunshine, control, fertilization, etc. are controlled and managed according to the future growth stage. It is possible to construct a complex environmental control device that aims for quality growth.

The drawings shown for the purpose of explanation of the present invention can be seen that various combinations of forms are possible to realize the gist of the present invention as shown in the drawings as one embodiment in which the present invention is embodied.

Therefore, the present invention is not limited to the above-described embodiments, and various modifications can be made to anyone having ordinary knowledge in the field to which the present invention belongs without departing from the gist of the present invention as claimed in the claims below. It will be said that there is a technical spirit of the present invention to the extent possible.

10: Farm automation equipment
20: Smart Farm Automation Server
30: computer
40: smartphone
100: Growth sensing module
120: camera module
200: integrated control module
210: data collection module
220: crop growth database
230: artificial intelligence module

Claims (4)

A farming automation facility comprising a growth sensing module that transmits temperature, humidity, sunshine, and weather sensor data and a camera module that takes and transmits crop images;
A data collection module for continuously collecting and storing various sensor data and crop images output from the agricultural automation facility is formed, and the data collected in the data collection module is compared with the built-in crop growth database to be converted into big data by artificial intelligence. Smart farming automation server including the artificial intelligence module to form and store; and
When the image of a crop is collected from the data collection module as included in the smart farming automation server, the artificial intelligence module is driven to output a growth management control signal of the crop, and comprehensive control information is produced with each collected sensor data. Integrated control module for transmitting to the worker computer or smart phone;
Consists of
When the image of the crop is taken and transmitted while driving the agricultural automation equipment, the algorithm of the artificial intelligence module is operated to judge the quality of the crop, and the temperature, humidity, carbon dioxide gas, geothermal temperature, soil moisture of each crop growth stage according to the quality of the crop , Crop image control, fertilization, crop moisture supply, oxygen supply, and sunshine to optimize the crop growth environment to be automatically managed, and the artificial intelligence module learning algorithm to update the big data crop image characterized in that the big data Farm automation system using. The artificial intelligence module of claim 1
Farming automation system using crop image big data, characterized by forming big data using a convolutional neural network (CNN) algorithm. The method of claim 2
The AI module recognizes a crop organ including leaves, flower pots, nodes, and fruits by analyzing the photographed crop image, and uses the recognized crop organ data to determine the leaf size, growth point length, growth point thickness, and fruit in the captured image. Farming automation system using crop image big data, characterized by extracting and storing growth data including the size, plant water, and fruit number, and allowing the derived data for each institution to be included in the big data. The method of claim 3
The data collection module of the smart farming automation server
By receiving and storing the crop image and the measured crop institution-specific data from the smart terminal, the crop growth data and crop recognition image information big data for each farm are built, and the accuracy of the crop image is improved to improve the learning data of the AI module. Crop growth diagnostic system, characterized in that to be.

Images