KR102525537B1 - Server, method and program for providing quality control services for agricultural products based on image analysis of agricultural products - Google Patents

Abstract

The present invention relates to a server for providing a quality management service of an agricultural product by analyzing an agricultural product image based on artificial intelligence which analyzes an image of a first agricultural product based on an artificial intelligence model to generate first quality information indicating quality of the first agricultural product included in the image of the first agricultural product when the at least one image of the first agricultural product is received from a user terminal.

Description

Server, method and program for providing quality control services for agricultural products based on image analysis of agricultural products}

The present disclosure relates to a server that provides quality control services for agricultural products, and more particularly, to a server that provides quality control services for agricultural products by analyzing images of agricultural products based on artificial intelligence.

Recently, due to the development of artificial intelligence technology, artificial intelligence technology is being applied to many fields.

If such artificial intelligence technology is applied in the cultivation process of agricultural products, it is expected that it will be easy to control the quality of agricultural products.

In particular, it is expected that the quality of agricultural products will be remarkably improved by providing customized information to farmers during the cultivation process, as it can manage the cultivation process, which farmers had to solve on their own, until the process of harvesting agricultural products.

Despite these advantages, there is currently no such technology published.

Republic of Korea Patent Publication No. 10-2021-0116965, (2021.09.28)

Embodiments disclosed in the present disclosure are aimed at providing a quality control service of agricultural products based on image analysis of agricultural products.

The problems to be solved by the present disclosure are not limited to the problems mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the description below.

A server for providing a quality management service of agricultural products based on image analysis of agricultural products according to an embodiment of the present disclosure for solving the above problems includes a communication unit communicating with a user terminal; A storage unit storing an artificial intelligence model in which an image-based agricultural product quality calculation method has been learned; When the artificial intelligence model is learned based on a learning dataset in which each of a plurality of images of agricultural products is labeled with quality information of the corresponding agricultural product, and at least one image of the first agricultural product is received from the user terminal, the image of the first agricultural product And a processor for analyzing based on the artificial intelligence model to generate first quality information representing the quality of the first agricultural product included in the first agricultural product image, wherein the processor, based on the first agricultural product image, Defect information and first color information of the first agricultural product may be calculated, and the first quality information may be generated based on the calculated defect information and the first color information.

In addition, the processor derives a photographing condition including at least one of a photographing distance, a photographing angle, and the number of photographed images of the first agricultural product according to the type of the first agricultural product, and transmits the derived photographing condition to the user terminal. can provide

In addition, when information on the type and cultivation amount of agricultural products grown by the user is received, the processor calculates the number of samples necessary for generating the first quality information based on the type and cultivation amount of the agricultural products, The calculated number of samples may be provided to the user terminal to request image capture of the first agricultural product to proceed.

In addition, the communication unit receives the value of the first sugar content directly measured through the user's sugar content meter for the at least one first agricultural product, the type of the first agricultural product, the received value of the first sugar content, A sugar level of the first agricultural product may be calculated based on the seasonal information of the first agricultural product and the calculated color information.

In addition, the communication unit receives second quality information generated by a manager directly checking at least one second agricultural product at the farmhouse of the user, and the processor performs the second quality information on the basis of the received second quality information. The first quality information may be corrected.

In addition, the second quality information includes a second sugar content value directly measured for a predetermined number of agricultural products at the user's farm site, and the processor determines the first sugar content value and the second sugar content value. If the value differs by more than a threshold value, a correction coefficient for the user's sugar content meter may be calculated based on the value of the second sugar content.

In addition, the storage unit stores cultivation period information in which the total cultivation period until cultivation is classified into a plurality of cultivation periods for each type of agricultural product, and the processor, when the first agricultural product is an agricultural product in cultivation, the first agricultural product Calculate the color uniformity of the agricultural products, calculate the expected harvest time of the agricultural products based on the type of the first agricultural products, the calculated first color information, and the calculated color uniformity, and calculate the expected harvest time It may be provided to the user terminal.

In addition, the storage unit stores at least one content for cultivating agricultural products necessary for cultivating agricultural products for each of the plurality of growing periods, and the processor includes the type of the first agricultural product, the calculated first color information, and the calculated content. Based on the determined color uniformity, determining the cultivation period corresponding to the first agricultural product, and extracting at least one agricultural product cultivation content necessary for the cultivation of the agricultural product according to the calculated expected harvest time and the determined cultivation period, The extracted content for cultivating at least one agricultural product may be provided to the user terminal.

In addition, a method for providing a quality control service of agricultural products based on image analysis of agricultural products according to an embodiment of the present disclosure for solving the above problems is a method performed by a server, and a first agricultural product is photographed from a user terminal. Receiving at least one image of a first agricultural product; Analyzing the first agricultural product image based on an artificial intelligence model to generate first quality information indicating the quality of the first agricultural product, wherein the generating agricultural product information includes the first agricultural product image based on the first agricultural product image. 1 calculating defect information and first color information of agricultural products; and generating the first quality information based on the calculated defect information and the first color information, wherein the artificial intelligence model uses a learning dataset in which quality information of agricultural products is labeled in an image for training data. learned based on

In addition to this, a computer program stored in a computer readable recording medium for execution to implement the present disclosure may be further provided.

In addition to this, a computer readable recording medium recording a computer program for executing a method for implementing the present disclosure may be further provided.

According to the above-described problem solving means of the present disclosure, there is an effect of providing a quality control service of agricultural products based on image analysis of agricultural products.

The effects of the present disclosure are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the description below.

1 is a schematic diagram of a quality management service providing system for agricultural products based on image analysis of agricultural products according to an embodiment of the present disclosure.
2 is a diagram illustrating a platform for providing a quality control service of agricultural products based on image analysis of agricultural products according to an embodiment of the present disclosure.
3 is a block diagram of a service providing server for quality management of agricultural products based on image analysis of agricultural products according to an embodiment of the present disclosure.
4 is a flowchart of a method for providing quality control services for agricultural products based on image analysis of agricultural products according to an embodiment of the present disclosure.
5 is a diagram illustrating a request for input of the type, cultivation amount, and cultivation period of agricultural products being cultivated by the user.
6 is a diagram illustrating providing a photographing method according to the information input of FIG. 5 .
7 is a diagram illustrating a request for uploading a photographed image of agricultural products.
8 is a diagram illustrating that the entire area of agricultural products is divided into a plurality of areas in order to calculate uniformity among color information of agricultural products.
9 is a diagram illustrating that the artificial intelligence model determines the current cultivation period of agricultural products according to image analysis of agricultural products, and extracts the expected harvest time and cultivation content necessary for cultivating agricultural products accordingly.
10 and 11 are diagrams illustrating an automatic photographing device for agricultural products.

Like reference numbers designate like elements throughout this disclosure. The present disclosure does not describe all elements of the embodiments, and general content or overlapping content between the embodiments in the technical field to which the present disclosure belongs is omitted. The term 'unit, module, member, or block' used in the specification may be implemented as software or hardware, and according to embodiments, a plurality of 'units, modules, members, or blocks' may be implemented as one component, It is also possible that one 'part, module, member, block' includes a plurality of components.

Throughout the specification, when a part is said to be "connected" to another part, this includes not only the case of being directly connected but also the case of being indirectly connected, and indirect connection includes being connected through a wireless communication network. do.

In addition, when a part "includes" a certain component, it means that it may further include other components without excluding other components unless otherwise stated.

Throughout the specification, when a member is said to be located “on” another member, this includes not only a case where a member is in contact with another member, but also a case where another member exists between the two members.

Terms such as first and second are used to distinguish one component from another, and the components are not limited by the aforementioned terms.

Expressions in the singular number include plural expressions unless the context clearly dictates otherwise.

In each step, the identification code is used for convenience of description, and the identification code does not explain the order of each step, and each step may be performed in a different order from the specified order unless a specific order is clearly described in context. there is.

Hereinafter, the working principle and embodiments of the present disclosure will be described with reference to the accompanying drawings.

In this specification, the 'quality management service providing server 100 of agricultural products according to the present disclosure' includes all of various devices capable of providing results to users by performing calculation processing. For example, the agricultural quality control service providing server 100 according to the present disclosure may include a computer, a server device, and a portable terminal 200, or may be in any one form.

Here, the computer may include, for example, a laptop computer, a desktop computer, a laptop computer, a tablet PC, a slate PC, and the like equipped with a web browser.

The server 100 device is a server that processes information by communicating with an external device, and includes an application server, a computing server, a database server, a file server 100, a game server, a mail server, a proxy server, and a web server. can do.

The portable terminal 200 is, for example, a wireless communication device that ensures portability and mobility, and includes a Personal Communication System (PCS), a Global System for Mobile communications (GSM), a Personal Digital Cellular (PDC), and a Personal Handyphone System (PHS). ), PDA (Personal Digital Assistant), IMT (International Mobile Telecommunication)-2000, CDMA (Code Division Multiple Access)-2000, W-CDMA (W-Code Division Multiple Access), WiBro (Wireless Broadband Internet) Terminal (200) All kinds of handheld-based wireless communication devices such as smart phones, watches, rings, bracelets, anklets, necklaces, glasses, contact lenses, or head-mounted devices ( HMD) and the like.

Functions related to artificial intelligence according to the present disclosure are operated through the processor 110 and the storage unit 130. The processor 110 may include one or a plurality of processors 110 . At this time, one or a plurality of processors 110 may include a general-purpose processor 110 such as a CPU, AP, or Digital Signal Processor (DSP), a graphics-only processor 110 such as a GPU, a Vision Processing Unit (VPU), or an artificial processor such as an NPU. It may be an intelligence-only processor (110). One or more processors 110 control input data to be processed according to predefined operation rules or artificial intelligence models stored in the storage unit 130 . Alternatively, when one or more processors 110 are dedicated AI processors 110, the AI dedicated processor 110 may be designed as a hardware structure specialized for processing a specific AI model.

A predefined action rule or an artificial intelligence model is characterized in that it is created through learning. Here, being made through learning means that a basic artificial intelligence model is learned using a plurality of learning data by a learning algorithm, so that a predefined action rule or artificial intelligence model set to perform a desired characteristic (or purpose) is created. means burden. Such learning may be performed in the device itself in which artificial intelligence according to the present disclosure is performed, or through a separate server and/or system. Examples of the learning algorithm include supervised learning, unsupervised learning, semi-supervised learning, or reinforcement learning, but are not limited to the above examples.

An artificial intelligence model may be composed of a plurality of neural network layers. Each of the plurality of neural network layers has a plurality of weight values, and a neural network operation is performed through an operation between an operation result of a previous layer and a plurality of weight values. A plurality of weights possessed by a plurality of neural network layers may be optimized by a learning result of an artificial intelligence model. For example, a plurality of weights may be updated so that a loss value or a cost value obtained from an artificial intelligence model is reduced or minimized during a learning process. The artificial neural network may include a deep neural network (DNN), for example, a Convolutional Neural Network (CNN), a Deep Neural Network (DNN), a Recurrent Neural Network (RNN), a Restricted Boltzmann Machine (RBM), A deep belief network (DBN), a bidirectional recurrent deep neural network (BRDNN), or deep Q-networks, but is not limited to the above examples.

According to an exemplary embodiment of the present disclosure, the processor 110 may implement artificial intelligence. Artificial intelligence refers to a machine learning method based on an artificial neural network in which a machine learns by mimicking a human's biological neuron. The methodology of artificial intelligence includes supervised learning in which input data and output data are provided together as training data according to the learning method, so that the answer (output data) of the problem (input data) is determined, and only input data is provided without output data. In unsupervised learning, where the answer (output data) of the problem (input data) is not determined, and whenever an action is taken in the current state, a reward is given in the external environment. , it can be classified as reinforcement learning in which learning proceeds in the direction of maximizing this reward. In addition, the methodology of artificial intelligence may be classified according to the architecture, which is the structure of the learning model. The architecture of widely used deep learning technology is Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN). , Transformers, and generative adversarial networks (GANs).

The device may include an artificial intelligence model. The artificial intelligence model may be one artificial intelligence model or may be implemented as a plurality of artificial intelligence models. Artificial intelligence models may be composed of neural networks (or artificial neural networks), and may include statistical learning algorithms that mimic biological neurons in machine learning and cognitive science. A neural network may refer to an overall model having a problem-solving ability by changing synaptic coupling strength through learning of artificial neurons (nodes) formed in a network by synaptic coupling. Neurons in a neural network may contain a combination of weights or biases. A neural network may include one or more layers composed of one or more neurons or nodes. Illustratively, the device may include an input layer, a hidden layer, and an output layer. A neural network constituting the device can infer a result (output) to be predicted from an arbitrary input (input) by changing the weight of a neuron through learning.

The processor 110 generates a neural network, trains or learns the neural network, performs an operation based on received input data, and generates an information signal based on the result of the operation. Neural network models include GoogleNet, AlexNet, VGG Network, etc., CNN (Convolution Neural Network), R-CNN (Region with Convolution Neural Network), RPN (Region Proposal Network) ), Recurrent Neural Network (RNN), Stacking-based deep Neural Network (S-DNN), State-Space Dynamic Neural Network (S-SDNN), Deconvolution Network, Deep Belief Network (DBN), Restricted Boltzmann Machine (RBM), It may include, but is not limited to, various types of models such as Fully Convolutional Network, Long Short-Term Memory (LSTM) Network, Classification Network, etc. The processor 110 is one for performing calculations according to the models of the neural network. It may include the above processor 110. For example, the neural network may include a deep neural network.

Neural networks include CNN (Convolutional Neural Network), RNN (Recurrent Neural Network), perceptron, multilayer perceptron, FF (Feed Forward), RBF (Radial Basis Network), DFF (Deep Feed Forward), LSTM (Long Short Term Memory), GRU (Gated Recurrent Unit), AE (Auto Encoder), VAE (Variational Auto Encoder), DAE (Denoising Auto Encoder), SAE (Sparse Auto Encoder), MC (Markov Chain), HN (Hopfield Network), BM(Boltzmann Machine), RBM(Restricted Boltzmann Machine), DBN(Depp Belief Network), DCN(Deep Convolutional Network), DN(Deconvolutional Network), DCIGN(Deep Convolutional Inverse Graphics Network), GAN(Generative Adversarial Network) ), LSM (Liquid State Machine), ELM (Extreme Learning Machine), ESN (Echo State Network), DRN (Deep Residual Network), DNC (Differentiable Neural Computer), NTM (Neural Turning Machine), CN (Capsule Network), It will be appreciated by those skilled in the art that it may include any neural network, including but not limited to Kohonen Network (KN) and Attention Network (AN).

According to an exemplary embodiment of the present disclosure, the processor 110 may include a Convolution Neural Network (CNN), a Region with Convolution Neural Network (R-CNN), a Region Proposal Network (RPN), such as Google Net, Alex Net, VGG Network, and the like. Recurrent Neural Network (RNN), Stacking-based deep Neural Network (S-DNN), State-Space Dynamic Neural Network (S-SDNN), Deconvolution Network, Deep Belief Network (DBN), Restricted Boltzmann Machine (RBM), Fully Convolutional Network, Long Short-Term Memory (LSTM) Network, Classification Network, Generative Modeling, eXplainable AI, Continual AI, Representation Learning, AI for Material Design, BERT for natural language processing, SP-BERT, MRC/QA, Text Analysis, Dialog System, GPT-3, GPT-4, Visual Analytics for vision processing, Visual Understanding, Video Synthesis, ResNet Data intelligence for Anomaly Detection, Prediction, Time-Series Forecasting, Optimization, Recommendation, Data Creation, etc. Algorithms may be used, but are not limited thereto. Hereinafter, embodiments of the present disclosure will be described in detail with reference to the accompanying drawings.

1 is a schematic diagram of a quality management service providing system 10 for agricultural products based on image analysis of agricultural products according to an embodiment of the present disclosure.

2 is a diagram illustrating a platform for providing a quality control service of agricultural products based on image analysis of agricultural products according to an embodiment of the present disclosure.

Referring to FIG. 1 , a system 10 for providing quality management services of agricultural products based on image analysis of agricultural products according to an embodiment of the present disclosure includes a server 100 , a user terminal 200 and a manager terminal 300 .

The server 100 is a subject that provides a quality management service for agricultural products, a user may mean a person who grows agricultural products in a farmhouse, and a manager may mean a person who visits a farmhouse and directly checks agricultural products.

Referring to FIG. 2 , a user who grows agricultural products in a farmhouse takes pictures of agricultural products being grown through a terminal 200 and transmits them to the server 100. It is possible to calculate an expected harvest time, extract cultivation contents for agricultural products, and provide them to the user.

In addition, a user who grows agricultural products in a farmhouse takes pictures of agricultural products that have been harvested after cultivation is completed through the terminal 200 and transmits them to the server 100, and the server 100 analyzes the received agricultural products images to obtain defect information of the agricultural products. , color information and quality information can be generated.

Through such a platform, the server 100 for providing quality management service of agricultural products according to an embodiment of the present disclosure provides various useful information so that the quality of agricultural products under cultivation can be improved, and after harvesting, by generating quality information, It will be possible to connect consumers who want agricultural products of the corresponding quality.

In the above, the quality control service providing system 10 of agricultural products based on image analysis of agricultural products according to an embodiment of the present disclosure has been schematically described through the schematic diagrams of FIGS. 1 and 2, and below, with reference to other drawings, in more detail. let me explain

3 is a block diagram of a server 100 for providing quality management service of agricultural products based on image analysis of agricultural products according to an embodiment of the present disclosure.

4 is a flowchart of a method for providing quality control services for agricultural products based on image analysis of agricultural products according to an embodiment of the present disclosure.

Referring to FIG. 3 , the server 100 for providing quality control services of agricultural products based on image analysis of agricultural products according to an embodiment of the present disclosure includes a processor 110 , a communication unit 120 and a storage unit 130 .

However, in some embodiments, the server 100 may include fewer or more components than those shown in FIG. 3 .

The processor 110 performs the above-described operation using a storage unit 130 that stores data for an algorithm or a program that reproduces the algorithm for controlling the operation of the components in the present device, and the data stored in the storage unit 130. It may be implemented with at least one processor 110 that performs. In this case, the storage unit 130 and the processor 110 may be implemented as separate chips. Alternatively, the storage unit 130 and the processor 110 may be implemented as a single chip.

In addition, the processor 110 may control any one or a combination of a plurality of components described above in order to implement various embodiments according to the present disclosure described in the following drawings on the present device.

In addition to operations related to the application program, the processor 110 may control general operations of the present device. The processor 110 may provide or process appropriate information or functions to the user by processing signals, data, information, etc. input or output through the components described above or by running an application program stored in the storage unit 130. .

In addition, the processor 110 may control at least some of the components of the present device in order to drive an application program stored in the storage unit 130 . Furthermore, the processor 110 may combine and operate at least two or more of the components included in the device to drive the application program.

The communication unit 120 communicates with the user terminal 200 and the manager terminal 300 .

In an embodiment of the present disclosure, the server 100 may provide a quality control service for agricultural products through a service application or the web, and may provide a service to a user accessing the server 100 through the communication unit 120.

The server 100 may provide a guide for capturing an image of agricultural products through a service application, and may provide cultivation contents customized to agricultural products being grown by the user. In this case, grown content may be viewed through a service application, or grown content may be provided by operating an external application through a link address.

In some embodiments, the terminal 200 and the automatic photographing device 400 for agricultural products are connected by a short-range communication method, and the service application controls the automatic photographing apparatus 400 for agricultural products by analyzing the agricultural products recognized on the photographing screen. An automatic shooting function may be provided.

The communication unit 120 may include one or more modules that connect the agricultural product quality control service providing server 100 to one or more networks.

The communication unit 120 may include one or more components enabling communication with an external device, and for example, at least one of a broadcast reception module, a wired communication module, a wireless communication module, a short-distance communication module, and a location information module. can include

Wired communication modules include not only various wired communication modules such as Local Area Network (LAN) modules, Wide Area Network (WAN) modules, or Value Added Network (VAN) modules, but also USB (Universal Serial Bus) ), high definition multimedia interface (HDMI), digital visual interface (DVI), recommended standard 232 (RS-232), power line communication, or plain old telephone service (POTS).

In addition to the WiFi module and the WiBro module, wireless communication modules include global system for mobile communication (GSM), code division multiple access (CDMA), wideband code division multiple access (WCDMA), and universal mobile telecommunications system (UMTS). ), time division multiple access (TDMA), long term evolution (LTE), and a wireless communication module supporting various wireless communication schemes such as 4G, 5G, and 6G.

The wireless communication module may include a wireless communication interface including an antenna and a transmitter for transmitting signals. In addition, the wireless communication module may further include a signal conversion module that modulates a digital control signal output from the processor 110 through a wireless communication interface into an analog type wireless signal under the control of the processor 110 .

The short-range communication module is for short-range communication, and includes Bluetooth®, Radio Frequency Identification (RFID), Infrared Data Association (IrDA), Ultra-Wideband (UWB), ZigBee, NFC ( Near Field Communication), Wi-Fi (Wireless-Fidelity), Wi-Fi Direct, and wireless USB (Wireless Universal Serial Bus) technologies may be used to support short-distance communication.

The storage unit 130 stores an artificial intelligence model in which an image-based agricultural product quality calculation method is learned.

The artificial intelligence model is trained based on a training dataset in which each of a plurality of images of agricultural products is labeled with the quality information of the corresponding agricultural products.

A CNN model can be applied as an artificial intelligence model, and a bounding box of an agricultural product image can be formed during CNN learning.

The processor 110 may set criteria for forming a bounding box in an image of agricultural products for each type of agricultural products, and may perform image recognition through a bounding box having a predetermined ratio for recognition accuracy and processing.

For example, the size of the bounding box for each type of agricultural product may be set as in the following equation.

H < S1/S2 < K, (H, K: minimum and maximum range of bounding boxes for each type of agricultural product)

A detailed description of the artificial intelligence model will be described later.

The storage unit 130 may store data supporting various functions of the device. The storage unit 130 may store a plurality of application programs (applications) running in the device, data for operation of the device, and commands. At least some of these application programs may exist for basic functions of the device. Meanwhile, the application program may be stored in the storage unit 130, installed in the device, and driven by the processor 110 to perform an operation (or function).

The storage unit 130 may store data supporting various functions of the device and programs for operation of the processor 110, and input/output data (eg, music files, still images, moving pictures, etc.) ) may be stored, and a plurality of application programs (application programs or applications) running in the device, data for operation of the device, and commands may be stored. At least some of these application programs may be downloaded from an external server through wireless communication.

The storage unit 130 may be a flash memory type, a hard disk type, a solid state disk type, a silicon disk drive type, or a multimedia card micro type. card micro type), card-type memory (for example, SD or XD storage unit 130, etc.), RAM (random access memory; RAM), SRAM (static random access memory), ROM (read-only memory; ROM) , electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), a magnetic memory, a magnetic disk, and an optical disk. In addition, the storage unit 130 is separated from the present device, but may be a database connected by wire or wirelessly.

In addition, the storage unit 130 may include a plurality of processes for the agricultural product quality management service providing server 100 .

In an embodiment of the present disclosure, the artificial intelligence model may identify at least one agricultural product in the image and determine the type of the identified agricultural product.

When the type of agricultural products is identified, the artificial intelligence model can generate defect information by analyzing images of agricultural products.

In an embodiment of the present disclosure, the defect information may mean the degree of defects included in agricultural products, and the degree of defects may be 0 when there are no defects in the image of agricultural products.

In the embodiments of the present disclosure, the defects of agricultural products are not limited to specific defects, and any defects that can reduce the quality of agricultural products, such as poor coloring, scratches, bruises, heat, disease, insect damage, nicks, etc., can be applied. do.

In addition, the artificial intelligence model can calculate/generate color information about agricultural products by analyzing images of agricultural products.

In the process of selecting agricultural products, the external color of agricultural products is a very important condition in selecting agricultural products. In particular, looking at the color of the fruit, you can infer the nutrients in the fruit. For example, fruits such as grapes, cherries, strawberries, and apples contain a large amount of anthocyanin, which is colored red, purple, or blue depending on the concentration of hydrogen ions.

In addition, depending on the type of agricultural product, the degree of uniformity (uniformity) of the color of the entire area inside and outside of one agricultural product may be directly related to the quality of the agricultural product. At this time, the quality may include the taste and nutrients of the agricultural product.

The artificial intelligence model generates color information and calculates/creates the quality of agricultural products using uniformity will be described in more detail with reference to other drawings below.

Referring to FIG. 4 , a process of a method for providing a quality management service of agricultural products based on image analysis of agricultural products according to an embodiment of the present disclosure will be described.

The server 100 receives information about the type and cultivation amount of agricultural products from the user terminal 200 through the communication unit 120 . (S100)

The processor 110 calculates the number of samples required to generate the first quality information. (S200)

5 is a diagram illustrating a request for input of the type, cultivation amount, and cultivation period of agricultural products being cultivated by a user.

6 is a diagram illustrating providing a photographing method according to the information input of FIG. 5 .

In an embodiment of the present disclosure, in order for the server 100 to provide a quality control service for agricultural products, the user transmits images of agricultural products grown or harvested to the server 100, and images of agricultural products received from the server 100 It is necessary to go through the process of analysis based on artificial intelligence models.

At this time, the server 100 performs S100 and S200 because it is more effective to request the number of samples required according to the user's cultivation amount and cultivation scale than to analyze the entire agricultural product during cultivation or harvest by the user.

In an embodiment of the present disclosure, a sample required for generating first quality information is referred to as a first agricultural product.

In one embodiment, the server 100 requests an input of at least one of the type, amount, and period of agricultural products being cultivated by the user.

In this case, as the cultivation amount, any numerical value capable of estimating the cultivation amount, such as seeding amount, yield, or cultivation area, can be applied.

The cultivation period means the period from sowing to the present, and the artificial intelligence model can be used in the process of determining/calculating the current cultivation period by analyzing the image of the first agricultural product, but information on the cultivation period is not necessarily required.

Referring to FIG. 6 , the processor 110 derives a photographing condition including at least one of a photographing distance, a photographing angle, and the number of photographed images of the first agricultural product according to the type of the first agricultural product.

The processor 110 provides the derived shooting conditions to the user terminal 200 .

The processor 110 calculates the number of samples (first agricultural products) necessary for generating the first quality information in S200, and the number of images taken in S300 indicates how many images should be taken for one first agricultural product. it means.

In order to accurately analyze agricultural products, this process is required because images of the entire area of agricultural products are required, and depending on the size of agricultural products, the shooting distance, shooting angle, and number of photographed images are required.

In one embodiment, the processor 110 may calculate the number of photographed images required for each type of agricultural product based on the following equation.

(L/N) x P

L: total surface area of the first agricultural product

N: standard area according to the type of first agricultural product

P: Weight of image importance by type of agricultural products

In some embodiments, the processor 110 may request a photographing such that the entire area of the agricultural product is included at a specific distance from the agricultural product, and a photographing angle may not be separately requested.

In some embodiments, the processor 110 may request a photographed image in which all agricultural products are photographed without deriving the number of photographed images.

As an embodiment, the processor 110 may derive a shooting condition when information on the type of agricultural products is received from the user terminal 200 . However, it is not limited thereto, and the processor 110 may recognize the type of agricultural products from the real-time photographed image of the user terminal 200 when the service application is operated, and derive a shooting condition based on the type and size of the recognized agricultural products. there is.

As an embodiment, the processor 110 may modify the shooting conditions even after the shooting conditions are provided to the user terminal 200 . For example, if the size of agricultural products is estimated based on the type of agricultural products and the required shooting distance is calculated, but as a result of analyzing the real-time photographed image of the user terminal 200, the size of the actual agricultural products is different, the processor 110 operates in real time. The shooting conditions can be modified based on the size of the produce in the shooting video.

The processor 110 receives at least one first agricultural product image in which the first agricultural product is photographed from the user terminal 200 . (S300)

7 is a diagram illustrating a request for uploading a photographed image of agricultural products.

The processor 110 calculates defect information and first color information of the first agricultural product based on the image of the first agricultural product. (S400)

The processor 110 generates first quality information representing the quality of the first agricultural product based on the defect information and the first color information of the first agricultural product. (S500)

In an embodiment of the present disclosure, quality information may be generated by including the size of agricultural products or considering the size of agricultural products.

The defect information may consist of the degree of defect or defect score for the first agricultural product, but may also be configured to include various types of information about the defect.

For example, the defect information may include at least one of a defect type, a defect size, a defect score according to a severity of the defect, and a defect size relative to the total area of the agricultural product.

The manager may visit the farmhouse of the user and directly check at least one second agricultural product to generate the second quality information.

As an embodiment, the server 100 may receive at least one of second color information and second sugar content values that a manager directly checks/measures for at least one second agricultural product at the user's farm site.

The processor 110 may generate second quality information indicating the quality of at least one second agricultural product based on at least one of the second color information and the second sugar content value received from the manager terminal 300. .

Since the first quality information generated based on the data directly measured by the user and the second quality information generated based on the data directly measured by the administrator may be different, the server 100 operating the platform may be different from each other at least once. Alternatively, a manager may be periodically dispatched to generate second quality information.

In one embodiment, the processor 110 may calculate a correction coefficient for the agricultural product of the user based on at least one second quality information.

Such a correction factor may be for standardizing quality measurement of agricultural products grown by users using platform services.

The processor 110 may generate final quality information by applying the correction coefficient to the first quality information generated after the correction coefficient is calculated.

In one embodiment, the second quality information may include a value of the second sugar content directly measured for at least one second agricultural product at the user's farm site.

In an embodiment of the present disclosure, the sugar content value includes at least one of a first sugar content value and a second sugar content value, the first sugar content value is directly measured by the user using a sugar content meter owned by the user, and 2 The value of sugar content is directly measured by the administrator using a sugar content meter after visiting the user's farmhouse.

In one embodiment, the communication unit 120 may receive the value of the first sugar content directly measured through the user's sugar content meter for the at least one first agricultural product.

At this time, it is not necessary to measure the sugar content of the first agricultural product, but it may be recommended to measure the sugar content of the first agricultural product for consistency of quality information. In some embodiments, the processor 110 may request to measure the sugar content of a predetermined number of agricultural products among agricultural products grown by the user.

The processor 110 may calculate the sugar level of the first agricultural product based on the type of the first agricultural product, the received first sugar content value, seasonal information of the first agricultural product, and the calculated first color information.

In one embodiment, when the first agricultural product is a harvested agricultural product, the processor 110 predicts the first agricultural product based on the type of the first agricultural product, the received value of the first sugar content, and the calculated first color information. A sugar level can be calculated.

In one embodiment, when the first agricultural product is an agricultural product being cultivated, the processor 110 includes the type of the first agricultural product, the received value of the first sugar content, seasonal information of the first agricultural product, and up to date of the first agricultural product. An expected sugar level of the first agricultural product may be calculated based on at least one of the cultivation period and the calculated first color information.

If the first agricultural product is an agricultural product that has been harvested, the sugar content may be almost confirmed, so the process is performed as above. The expected sugar level is calculated by adding the cultivation period up to.

The processor 110 may calculate a correction coefficient for the user's sugar content meter based on the second sugar content value when the first sugar content value and the second sugar content value differ by more than a threshold value.

The processor 110 may correct the first sugar content value by applying the correction coefficient to the received first sugar content value after the correction coefficient for the user's sugar content meter is calculated.

Since the value of sugar content measured may vary depending on the type of sugar content meter, the server 100 may standardize sugar content measurement through the above embodiment.

In one embodiment, the processor 110, when information (eg, brand, product name, etc.) on the sugar content meter used by the user is received from the user terminal 200, the correction coefficient for the user's sugar content meter based on the received information can be calculated.

In one embodiment, the processor 110 calculates the expected sugar content level for the harvested first agricultural product after the expected sugar content level is calculated for the first agricultural product being grown, the weather conditions during the cultivation period and the amount of change in the expected sugar content level can be trained by inputting it into an artificial intelligence model.

As this learning is repeated, the artificial intelligence model can more accurately calculate the expected sugar level for agricultural products under cultivation. As an embodiment, the processor 110 may input expected weather conditions together when calculating the expected sugar level for agricultural products being grown.

In an embodiment of the present disclosure, the storage unit 130 may store color grades according to color information for each type of agricultural product.

The processor 110 may determine the color grade of the first agricultural product based on the calculated first color information, and generate first quality information based on at least one of defect information, color grade, and first sugar content value. can do.

In one embodiment, the processor 110 may receive acidity information on the first agricultural product from the user terminal 200, and the first agricultural product based on at least one of defect information, color grade, acidity information, and the value of the first sugar content. 1 Quality information can be created.

In one embodiment, the processor 110 may access the server 100 and request input of quality information desired to purchase from a buyer (consumer) who needs agricultural products.

In this case, the information received from the consumer may include the type of agricultural products demanded, the quality level of the agricultural products, the size of the agricultural products, and the like.

In one embodiment, after S500, the processor 110 may match the first quality information calculated in S400 with quality information set from consumers registered in the server 100 to match producers (users) and consumers (buyers). there is.

The processor 110 may provide an immediate purchase function to consumers (buyers) when the first agricultural products are harvested agricultural products, and may provide a purchase reservation function to consumers (buyers) when the first agricultural products are agricultural products being grown. .

When providing a purchase reservation function, the processor 110 may guide the estimated harvest time calculated for the first agricultural product.

8 is a diagram illustrating that the entire area of agricultural products is divided into a plurality of areas in order to calculate uniformity among color information of agricultural products.

The storage unit 130 stores the importance of color information for each type of agricultural product, and in an additional embodiment stores the importance of color uniformity for each type of agricultural product.

For example, sugar content weights may be set according to color information and color uniformity for each type of agricultural product, and the calculated sugar content level may be corrected.

For example, even if the color of agricultural products is not directly related to the taste of agricultural products, depending on the color, it may be aesthetically pleasing and satisfy consumer preferences, so at least some agricultural products depend on color information and color uniformity for each type of agricultural product. Quality information weights may be set accordingly.

For example, although the average color information of an apple mixed with very red and very yellow and a normal red apple can be calculated the same, the uniformity of color is significantly different, and the average color information with good uniformity in the eyes of consumers is Because you might prefer red apples.

For example, agricultural products such as apples, jujubes, tomatoes, and pumpkins may have a higher importance of color uniformity than other agricultural products.

The processor 110 may calculate an average color value for the entire surface of the first agricultural product based on the image of the first agricultural product, for example, may calculate the first average color value using RGB values.

The processor 110 may divide the entire surface of the first agricultural product into a plurality of regions based on the image of the first agricultural product, and may calculate a second average color value for each of the plurality of divided regions.

The processor 110 may calculate the color uniformity of the first agricultural product based on the distribution of the second average color value from the first average color value.

9 is a diagram illustrating that the artificial intelligence model determines the current cultivation period of agricultural products according to image analysis of agricultural products, and extracts the expected harvest time and cultivation content necessary for cultivating agricultural products accordingly.

Referring to FIG. 9 , the server 100 may calculate an expected harvest time by inputting an image of a first agricultural product into an artificial intelligence model, extract contents for cultivating agricultural products, and provide the content to the user.

In one embodiment, the processor 110 calculates the color uniformity of the first agricultural product when the first agricultural product is an agricultural product being cultivated, and the type of the first agricultural product, the calculated first color information, and the calculated color uniformity. Based on , it is possible to calculate the expected harvest time of agricultural products.

The artificial intelligence model is trained in advance by inputting images of agricultural products according to the cultivation period for each type of agricultural product. Therefore, the artificial intelligence model can calculate the expected harvest time when the type of agricultural products, color information, and color uniformity are input.

In addition, the processor 110 may provide information by providing the calculated expected harvest time to the user terminal 200 .

In this process, the server 100 may use information about the cultivation period directly input by the user.

In one embodiment, the storage unit 130 stores at least one agricultural product cultivation content necessary for the cultivation of agricultural products for each cultivation period classified by type of agricultural products.

The processor 110 may determine a growing period corresponding to the first agricultural product based on the type of the first agricultural product, the calculated first color information, and the calculated color uniformity.

In an embodiment of the present disclosure, the cultivation period determined by the processor 110 may mean the actual cultivation period from the sowing of the first agricultural product until now, but the cultivation period date for standardization of agricultural products in the server 100 may be That is, the cultivation period determined/calculated by the processor 110 does not necessarily mean the actual number of cultivation days.

In an embodiment of the present disclosure, the cultivation period may mean a cultivation section, and the period from sowing to harvest is divided into a plurality of sections, and the processor 110 determines which cultivation section the first agricultural product currently corresponds to. can

As an embodiment, when the actual cultivation period is input from the user terminal 200, the processor 110 may calculate the immaturity of the first agricultural product based on the difference between the calculated cultivation period and the actual cultivation period.

The processor 110 extracts at least one content capable of resolving the immaturity level and provides it to the user terminal 200 when the actual cultivation period differs from the calculated cultivation period by more than a preset standard or the immaturity level exceeds a threshold value. and recalculate the expected harvest time.

The storage unit 130 stores at least one content for cultivating agricultural products for each type of agricultural product and for each growing period, and each content may have at least one tag stored therein for easy retrieval.

Required contents for cultivating agricultural products may differ according to the user's experience, career, skill, equipment, etc., and thus customized contents may be extracted and provided to the user by searching tags based on the user's information.

In one embodiment, the processor 110 may calculate a defect score for the first agricultural product using the following algorithm.

Defect score for produce = ( B x C / A ) x D

A: Total surface area of produce

B: Area of agricultural products by type of defect

C: Weight by type of defect in agricultural products

D: Defect Importance Weight by Type of Agricultural Product

In one embodiment, the severity may be different for each type of defect depending on the type of agricultural product.

In this case, if at least one specific agricultural product contains a certain percentage or more of specific defects relative to the total surface area, the defect score of the agricultural product may be treated as 0 points.

In one embodiment, the severity of the defect itself may be different depending on the type of agricultural product.

Certain agricultural products are sensitive to defects, and certain agricultural products may be less sensitive to defects than other agricultural products.

10 and 11 are diagrams illustrating an automatic photographing device 400 for agricultural products.

Referring to FIG. 10 , in an embodiment of the present disclosure, the agricultural product quality control service providing system 10 may further include an automatic agricultural product photographing device 400 capable of automatically photographing a first agricultural product image.

The automatic agricultural product photographing device 400 includes a processor for operation, a communication unit for communication with the terminal 200, a storage unit storing an operation algorithm of the automatic agricultural product photographing apparatus 400, a shelf 420 on which agricultural products are placed, and a shelf It may include a roller 430 for rotating agricultural products placed on 420, a cradle type arm 410 having one end coupled to the shelf 420 and the terminal 200 mounted at the other end.

A processor, a communication unit, and a storage unit of the automatic photographing device 400 for agricultural products may be provided inside the shelf.

When the terminal 200 and the automatic photographing device 400 for agricultural products are connected through short-range communication, the first agricultural products are placed on the shelf 420, and the service application of the terminal 200 is operated, the processor captures the screen of the terminal 200. The first agricultural product of the above can be recognized.

The processor recognizes the type and size of the first agricultural products on the shooting screen of the terminal 200, calculates the number of samples required for generating the first quality information, and shooting conditions for shooting the first agricultural products (shooting distance, shooting distance, shooting number of images, etc.) can be derived.

The processor may control the arm 410 so that the distance between the first produce on the shelf 420 and the camera of the terminal 200 is separated by the derived photographing distance.

When the arm 410 is adjusted by the shooting distance, the processor rotates the roller 430 at a preset speed to rotate the agricultural products on the shelf 420, and operates the camera of the terminal 200 to image the first agricultural products. can be obtained.

The processor may control the communication unit to transmit the photographed image of the first agricultural products to the server 100 .

In some embodiments, the processor may also adjust magnification, brightness, and color of the camera to standardize images of agricultural products.

When using such an automatic photographing device 400 for agricultural products, the user can take an image of the first agricultural product automatically and transmit it to the server 100 without having to read a separate shooting guide and shoot according to the shooting conditions will have an effect.

The method according to one embodiment of the present disclosure described above may be implemented as a program (or application) to be executed in combination with the server 100, which is hardware, and stored in a medium.

The above-described program is C, C++, JAVA, C, C++, JAVA, which the processor 110 (CPU) of the computer can read through the device interface of the computer so that the computer reads the program and executes the methods implemented as a program. It may include a code coded in a computer language such as machine language. Such codes may include functional codes related to functions defining necessary functions for executing the methods, and include control codes related to execution procedures necessary for the processor of the computer to execute the functions according to a predetermined procedure. can do. In addition, these codes may further include memory reference related code for determining where (address address) of the computer's internal or external memory the additional information or media required for the computer's processor to execute the functions should be referenced. there is. In addition, when the processor 110 of the computer needs to communicate with any other remote computer or server 100 in order to execute the functions, the code uses the communication module of the computer to Communication-related codes for how to communicate with other computers or servers, what information or media to transmit/receive during communication, and the like may be further included.

The storage medium is not a medium that stores data for a short moment, such as a register, cache, or memory, but a medium that stores data semi-permanently and is readable by a device. Specifically, examples of the storage medium include ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device, etc., but are not limited thereto. That is, the program may be stored in various recording media on various servers accessible by the computer or various recording media on the user's computer. In addition, the medium may be distributed to computer systems 10 connected through a network, and computer readable codes may be stored in a distributed manner.

Steps of a method or algorithm described in connection with an embodiment of the present disclosure may be implemented directly in hardware, implemented in a software module executed by hardware, or a combination thereof. A software module may include random access memory (RAM), read only memory (ROM), erasable programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, hard disk, removable disk, CD-ROM, or It may reside in any form of computer readable recording medium well known in the art to which this disclosure pertains.

In the above, the embodiments of the present disclosure have been described with reference to the accompanying drawings, but those skilled in the art to which the present disclosure pertains can be implemented in other specific forms without changing the technical spirit or essential features of the present disclosure. you will be able to understand Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive.

10: Agricultural products quality management service provision system
100: server
110: processor
120: communication department
130: storage unit
200: user terminal
300: manager terminal
400: Automatic filming device for agricultural products
410 cancer
420: shelf
430: roller

Claims (10)

A communication unit that communicates with a terminal of a user who grows agricultural products in a farmhouse;
A storage unit storing an artificial intelligence model in which an image-based agricultural product quality calculation method has been learned; and
The artificial intelligence model is learned based on a learning dataset in which quality information of the corresponding agricultural product is labeled for each of a plurality of images of agricultural products for learning, and information about agricultural products grown or harvested by the user from the user terminal through the communication unit. When an agricultural product image is received, a processor for analyzing the agricultural product image based on the artificial intelligence model to generate first quality information indicating the quality of the user agricultural product included in the agricultural product image,
The first quality information includes defect information and color information of the agricultural product, wherein the defect information is information indicating defects of at least one of coloring defects, wounds, bruises, heat, disease, insect damage, and nicks of the agricultural products,
the processor,
Deriving the defect information and the color information of the user agricultural product based on the agricultural product image,
When first information on the type and amount of agricultural products grown or harvested by the user is received through the communication unit, the number of samples necessary for generating the first quality information is calculated based on the first information. do,
The second information on the calculated number of samples is transmitted to the user terminal through the communication unit, so that the user terminal photographs as many images of agricultural products as the number of samples based on the second information, and the captured images of agricultural products Request to transmit to the communication unit,
When the third information on the first sugar content value of the user agricultural product measured by the user's sugar content meter is received through the communication unit, the type of the user agricultural product, the first sugar content value, seasonal information of the user agricultural product, and Based on the color information, the sugar level of the user's agricultural products is calculated,
The fourth information on the second sugar content value of the user's agricultural products measured by the sugar content meter of the manager dispatched from the farmhouse of the user and the second quality information about the user's agricultural products generated by directly checking the manager When received through the communication unit, calculating a first correction coefficient for standardization of quality measurement of the user's agricultural products based on the second quality information,
Generating final quality information by applying the first correction coefficient to the first quality information generated after the first correction coefficient is calculated;
When the first sugar content value and the second sugar content value differ by more than a threshold value, calculating a second correction coefficient for the user's sugar content meter based on the second sugar content value,
correcting the first sugar content value by applying the second correction coefficient to the first sugar content value received after the second correction coefficient is calculated;
A server that provides quality control services for agricultural products based on image analysis of agricultural products.
delete delete delete delete delete According to claim 1,
The storage unit stores cultivation period information in which the total cultivation period until cultivation for each type of agricultural product is classified into a plurality of cultivation periods,
the processor,
If the user agricultural product is an agricultural product in cultivation, calculating the color uniformity for the user agricultural product,
Based on the type of user agricultural products, the color information, and the color uniformity, the expected harvest time of the user agricultural products is calculated,
Providing the calculated expected harvest time to the user terminal,
A server that provides quality control services for agricultural products based on image analysis of agricultural products.
According to claim 7,
The storage unit stores at least one content of cultivating agricultural products necessary for cultivating agricultural products for each of the plurality of growing periods,
the processor,
Based on the type of user agricultural product, the color information, and the color uniformity, determining the cultivation period corresponding to the user agricultural product,
Extracting at least one agricultural product cultivation content necessary for the cultivation of the user agricultural product according to the calculated expected harvest time and the determined cultivation period,
Providing the extracted at least one agricultural content to the user terminal,
A server that provides quality control services for agricultural products based on image analysis of agricultural products.
As a method performed by the server,
Receiving an agricultural product image for user agricultural products grown or harvested by the user from a terminal of a user who cultivates agricultural products in a farmhouse;
Analyzing the image of the agricultural product based on an artificial intelligence model to generate first quality information indicating the quality of the user agricultural product included in the image of the agricultural product,
The first quality information includes defect information and color information of the agricultural product, wherein the defect information is information indicating defects of at least one of coloring defects, wounds, bruises, heat, disease, insect damage, and nicks of the agricultural products,
The server,
Deriving the defect information and the color information of the user agricultural product based on the agricultural product image,
When first information about the type and cultivation amount of agricultural products grown or harvested by the user is received, based on the first information, the number of samples necessary for generating the first quality information is calculated,
Second information on the calculated number of samples is transmitted to the user terminal, and the user terminal photographs as many images of agricultural products as the number of samples based on the second information, and requests that the captured images of agricultural products be transmitted. do,
When the third information on the first sugar content value of the user agricultural product measured by the user's sugar content meter is received, the type of the user agricultural product, the first sugar content value, the seasonal information and the color information of the user agricultural product Calculate the sugar level of the user's agricultural products based on the
Fourth information on the second sugar content value of the user's agricultural products measured by the sugar content meter of the manager dispatched from the user's farmhouse site and second quality information about the user's agricultural products generated by directly checking by the manager are received If it is, calculating a first correction coefficient for standardizing the quality measurement of the user agricultural products based on the second quality information,
Generating final quality information by applying the first correction coefficient to the first quality information generated after the first correction coefficient is calculated;
When the first sugar content value and the second sugar content value differ by more than a threshold value, calculating a second correction coefficient for the user's sugar content meter based on the second sugar content value,
correcting the first sugar content value by applying the second correction coefficient to the first sugar content value received after the second correction coefficient is calculated;
A method of providing quality control services for agricultural products based on image analysis of agricultural products.
A computer-readable recording medium in which a program for executing the method of claim 9 is stored in combination with a computer, which is hardware.

Images