KR102607895B1 - Smart Cattle Management System - Google Patents

Abstract

The present invention relates to a smart cattle management system, and more specifically, to a smart cattle management system that can automatically manage the history of cattle, such as the age, gender, breed, and disease history of each cattle individual, through automatic recognition of cattle. . According to one aspect of the present invention, a smart cattle management system for identifying cattle individuals is provided, the smart cattle management system comprising: a cattle identification device; A user terminal connected to the cattle identification device; And a control server that is connected to the user terminal through a network and receives cattle identification data transmitted from the user terminal and provides cattle entity data to the user terminal in response, and the cattle identification device includes an RFID tag provided to the cattle entity. RFID reader for reading; Includes a camera for vision-based recognition.

Description

Smart Cattle Management System

The present invention relates to a smart cattle management system, and more specifically, to a smart cattle management system that can automatically manage the history of cattle, such as the age, gender, breed, and disease history of each cattle individual, through automatic recognition of cattle. .

Recently, computer reporting has been mandatory at the distribution stage of the livestock product traceability system, and the livestock product traceability system records and manages history information from the breeding stage of cattle and pigs, etc. to slaughter, packaging, and sales stages, thereby improving the efficiency of livestock quarantine and distribution of domestic livestock products. Safety is ensured, the distribution process is disclosed, and anyone can check the breeding site, slaughterhouse, etc.

In addition, if a company subject to electronic reporting packages or trades livestock products subject to traceability management, it is required to register them electronically through the traceability management system, and meat sales establishments are required to indicate the traceability (bundle) number when selling livestock products subject to traceability management. In addition, the importance of traceability management of livestock products has been highlighted as fines are imposed if electronic reporting is not made or the history number is not displayed.

Conventionally, a tag device has been used to manage cattle. When using a conventional tag device, it is possible to identify cattle, but there is a problem in that it takes a lot of time because the history management of each cattle individual is performed manually.

Republic of Korea Patent Publication 10-2017-0116509 (October 19, 2017)

The purpose of the present invention is to provide a smart cattle management system that can accurately identify cattle individuals and automatically manage the history of cattle individuals.

According to one aspect of the present invention to solve the above-described problem, a smart cattle management system for identifying cattle individuals is provided, the smart cattle management system comprising: a cattle identification device; A user terminal connected to the cattle identification device; And a control server that is connected to the user terminal through a network and receives cattle identification data transmitted from the user terminal and provides cattle entity data to the user terminal in response, and the cattle identification device includes an RFID tag provided to the cattle entity. RFID reader for reading; Includes a camera for vision-based recognition.

In the above-described aspect, the user terminal includes a vision recognition module for determining a cattle entity from an image of the cattle transmitted from a vision recognition camera,

The vision recognition module includes an image receiving unit for receiving an image transmitted from Camella; a feature point extraction unit for detecting feature points of the axis from the image received by the image receiver; and a cattle individual determination unit for determining the cattle individual based on the feature points extracted from the feature point extraction unit.

In addition, in the above-described aspect, the control unit of the user terminal first recognizes the coupon information of the cattle from the RFID tag using an RFID reader for the identified cattle entity, and when the coupon information of the cattle is recognized from the RFID tag information, it sends the coupon information of the cattle to the management server. It is configured to call cattle data corresponding to coupon information.

In addition, in the above-described aspect, the control unit of the user terminal first recognizes the tag information of the cattle from the RFID tag using an RFID reader for the identified cattle entity, and then further performs vision recognition on the cattle entity using the vision recognition module. It is configured to determine whether the cattle corresponding to the coupon information recognized through RFID and the characteristic points of the cattle recognized through the vision recognition module match each other.

In addition, in the above-described aspect, the feature point extraction unit is configured to extract feature points of the cattle from the image of the cattle and compare the extracted feature points of the cattle with feature points of the cattle stored in the management server to determine the authenticity of the cattle, and the feature points of the cattle are Chukwu's inscription; and the shape of the facial whorls of cattle; It includes at least one of the shape of the cattle's facial eyebrows, the cattle's coloring, and the nose pattern.

In addition, in the above-described aspect, the smart cattle management system includes a sick cattle detection module for detecting the health status of the cattle; An integrated database storing historical data of cattle identified through the cattle identification device; a lesion data collection module that collects lesion data of cattle from the history data of the cattle; It includes a diagnosis and prescription module that diagnoses and prescribes the current condition of the cattle through lesion data and the cattle's history data, and the diagnosis and prescription module includes the main symptoms, secondary symptoms, and lesion location obtained from the lesion data, and the integrated It is configured to search the disease library based on age/sex, breed/pedigree, season/climate/region, and disease duration obtained from cattle history data in the database and diagnose diseases in the order of matching from the disease library.

According to an embodiment of the present invention, it is possible to provide a smart cattle management system that can identify cattle more accurately through vision-based individual recognition of cattle and individual recognition of cattle through RFID tags and manage the cattle through this. In addition, according to an embodiment of the present invention, data stored in the database through cattle individual recognition is combined with equipment such as ICT equipment, for example, estrus detector, body temperature, automatic feed feeder, automatic milking machine, etc., to enable more efficient management of cattle. In addition, the advantage of enabling continuous disease and health management for cattle individuals can be obtained.

In addition, accurate individual identification without individual identification errors is possible in livestock wholesale markets, slaughterhouses, etc., and by providing disease/prescription history in the integrated DB, the withdrawal period of drugs used in calf/cattle trade or cattle history and beef slaughter information is accurately reflected. We can provide safety history data on livestock products.

1 is a diagram showing an example of a cattle individual recognition system according to the present invention;
Figure 2 is a diagram showing the inscription characteristics of cattle used for vision recognition of cattle individuals;
Figure 3 is a diagram showing whorls (fixed features) among the facial features of cattle used for vision recognition of cattle;
Figure 4 is a diagram showing the hair color and nose skin color characteristics of cattle used for vision recognition of cattle;
Figure 5 is a diagram showing horns (variable features) among the facial features of cattle used for vision recognition of cattle;
6 is a diagram showing an example of a user terminal;
Figure 7 is a diagram showing an example of a vision recognition module provided in a user terminal;
Figure 8 is a flowchart showing a process of identifying a cattle entity using an RFID module and a vision recognition module in the control unit of the user terminal;
Figure 9 is a diagram showing a disease diagnosis and prescription algorithm using integrated data.

The advantages and features of the present invention and methods for achieving them will become clear by referring to the embodiments described in detail below along with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below and will be implemented in various different forms.

The examples herein are provided to make the disclosure of the present invention complete and to fully inform those skilled in the art of the scope of the invention. And the present invention is only defined by the scope of the claims. Accordingly, in some embodiments, well-known components, well-known operations and well-known techniques are not specifically described in order to avoid ambiguous interpretation of the present invention.

Like reference numerals refer to like elements throughout the specification. Also, the terms used (mentioned) in this specification are for describing embodiments and are not intended to limit the present invention. As used herein, singular forms also include plural forms, unless specifically stated otherwise in the context. Additionally, components and operations referred to as 'including (or, including)' do not exclude the presence or addition of one or more other components and operations.

Unless otherwise defined, all terms (including technical and scientific terms) used in this specification may be used with meanings that can be commonly understood by those skilled in the art to which the present invention pertains. Additionally, terms defined in commonly used dictionaries are not interpreted ideally or excessively unless they are defined.

Hereinafter, embodiments of the present invention will be described with reference to the attached drawings. 1 is a diagram showing an example of a smart cattle management system according to an embodiment of the present invention. As shown in Figure 1, the smart cattle management system according to the present invention includes a cattle identification device 10; A user terminal (20) connected to the cattle identification device (20); And a control server 30 that is connected to the user terminal 20 through a network, receives cattle identification data transmitted from the user terminal 20, and provides cattle individual data to the user terminal 20 in response. .

The cattle identification device 10 includes an RFID reader 11 and a camera 12 for vision-based recognition. The RFID reader 11 and the camera 12 may be provided independently as separate devices from the user terminal 20 and used while attached to the user terminal, or may be formed integrally with the user terminal 20.

The RFID reader 11 functions to recognize the RFID tag provided for each cattle individual in the form of a cattle electronic tag or microchip. Radio Frequency Identification (hereinafter referred to as 'RFID') is a technology that attaches an electronic tag and uses wireless communication technology to derive or record the information contained in the tag in a non-contact manner. It is a technology that can detect and process information about animals, people, etc. and surrounding situation information in real time.

RFID tags are divided into passive tags, which operate by receiving energy from a reader's radio signal without an internal power source, and active tags, which have a built-in power source. In an embodiment of the present invention, either an active or passive tag may be used as the RFID tag, and the present invention is not limited thereto.

In addition, in the present invention, a configuration using an RFID tag and a reader is used as a first means for cattle individual identification, but the present invention is not limited thereto and includes optical identification devices such as barcodes and barcode readers, or QR codes and QR code readers. Can be used to identify cattle individuals.

The cattle identification device 10 further includes a vision recognition camera 12 for vision-based cattle individual recognition. The camera 12 may be a camera provided to the user terminal 20 itself, or a high-resolution camera means for more precise vision recognition may be provided separately, and the image taken from the camera 12 may be transmitted wirelessly or wired. It may also be configured to be transmitted to the user terminal 10 through a communication interface.

A vision recognition system including a camera 12 operates to identify fixed and variable features appearing on the cow's face. Here, fixed characteristics refer to characteristics that appear in cattle from birth, and these fixed characteristics are maintained without change during the cattle's growth process. The variable characteristics of a cow mean that the characteristics can change as the cow grows. Fixed features of a cow may include the cow's facial whorls, eyebrows, inscriptions, and coat pattern features, and variable features may include the cow's horns and sign number.

As shown in Figure 2, the cow's inscription can be used as a fixed feature, and because the cow's inscription is unique to each individual like a human fingerprint, it enables individual identification of the cow. This inscription feature can be commonly used for both Korean beef and dairy cows.

Figure 2 is a diagram for explaining the features according to the facial whorls among the fixed features appearing on the facial area of a cow. The facial whorl shape used as a fixed feature of the eyebrows can be divided into facial whorls, such as face-on, face-centered, face-down, face-pair, and face-shaped. In addition, depending on the presence or absence of whorls at the ends of the eyebrows, fixed features such as left-side, right-side, positive-side, and mid-face are formed. It can be used. This feature can be commonly used in both Korean beef and dairy cows.

In addition, in addition to the inscription on the cow as described above, in Figure 3, the feature according to the facial whorl among the fixed features appearing on the facial part of the cow can be used. The shape of the whorls between the eyes (noodle lines), which are used as a fixation feature of the eyelids, can be divided into the shape of the whorls such as face-wise, face-middle, face-wise, face-paired, and face-like. In addition, depending on the presence or absence of whorls (mis-lines) at the end of the eyebrows, there are left, right, and both ends. , fixed features such as pending may be used. This feature can be commonly used in both Korean beef and dairy cows.

Figure 4 is a diagram showing the color pattern of a dairy cow (Holstein). As shown in Figure 5, in the case of dairy cows, each individual cow has unique hair patterns and nose skin color characteristics. Therefore, if the cattle raised on a farm are Holstein dairy cattle, the coat pattern and nose skin color can be used as fixed characteristics and as additional characteristics. This feature can only be used on Holstein dairy cows.

In addition to the fixed characteristics described above, in the present invention, variable characteristics of cattle can be used for individual identification of cattle. As a variable feature, a variable feature according to the direction of the cow's horn may be used as shown in FIG. 5. Depending on the direction of the cow's horn, features such as upward angle, downward angle, forward angle, incised angle, angled angle, and non-angled angle can be used. The direction of the horn can be changed in size or shape through surgery such as beveling surgery in cattle, and the size and shape of the growing cattle can be changed. In some cases, there may be changes in size and shape as the horns grow. The variable features of cattle can be used together with fixed features to improve vision recognition performance, and these variable features are common to both Korean cattle and dairy cows.

Additionally, the coupon number can be used as a variable characteristic of cattle. The coupon number refers to the standard coupon and ranch coupon number, and may not be recognized or may change when the coupon is removed or replaced.

Figure 6 is a diagram showing an example of a dedicated user terminal 20 designed to be used in a smart cattle management system. As shown in Figure 6, the dedicated user terminal 20 includes an RFID reader 11 capable of reading the RFID tag provided on the cattle and a vision recognition camera capable of identifying the individual cattle through the fixed and variable characteristics of the cattle described above. (12), the vision recognition camera 12 is connected to the vision-based identification module 100 composed of dedicated vision recognition software installed on the user terminal 20, and the image captured through the camera 12 is It is transmitted to the based identification module 100, and the vision-based identification module 100 analyzes the transmitted image and performs identification of the cattle entity.

The vision-based identification module 100 recognizes one or more of the following characteristics to identify the object and performs the following operations.

① Inscription (nose print)

② Color/pattern (coat pattern)

③ Recognize pre-registered individual features through facial photography: Fixed features (noodle lines, hair lines, etc.) / variable features (horn shape, facial tag number)

Figure 7 is a block diagram schematically showing the internal configuration of the vision-based identification module 100 as described above.

The image receiver 100 may receive a facial image or an image including the face of the target cattle from the camera 12. This corresponds to a component that receives the image to be judged in order to judge the cattle entity through vision recognition. In general, cattle include Korean cattle and dairy cows, and in this case, the image receiver refers to an image including the entire image of Korean cattle and dairy cows or at least the face of the cattle.

The feature point extraction unit 120 extracts facial fixed feature points from the received image as described above, for example, a cow's inscription image, a cow's whorl feature point between the eyes (face-up, face-down, face-middle, face-pair, face-grain), and eyebrow whorl feature points (left-hand, right-hand). , sheep rice, unfinished business) images, dairy cow hair color and nose skin color images, cow horn direction images, and facial mark numbers are extracted. Since the hair color and nose skin color feature points correspond to dairy cows, the feature point extraction unit 120 may be configured to first determine whether the cow is a Korean cow or a dairy cow, and then extract the corresponding feature points to identify the cow individual.

The cattle entity determination unit 130 compares the extracted feature point image with the feature point image stored in the database of the server 30 and identifies the object of the target cattle.

Additionally, the feature point extraction unit 120 may include the preprocessing unit 121 on the received image to perform preprocessing on the cattle image for accurate identification of the cattle cattle. The preprocessor may set an image corresponding to a region of interest (ROI) among the facial images input from the image receiver 110 and then perform preprocessing on the image. This is to extract facial data corresponding to the region of interest related to facial features among facial images. It lowers the system load that occurs when extracting feature points for the entire image and improves vision-based processing speed.

For this purpose, the pre-processing unit includes images of the cow's inscriptions, which are the areas necessary for identification among the facial parts as described above, images of the facial lines (on the face, under the face, mid-face, pair of faces, and face lines) and tail lines (left tail, right tail, double tail, misgyeol), hair color and nose of the dairy cow. By operating to extract only the skin color image, cow horn direction image, and ticket number image, recognition speed and accuracy can be increased.

To extract such feature point images, the preprocessor may use an adaptive thresholding algorithm to extract only images corresponding to the region of interest.

Subsequently, when the preprocessed region of interest image is extracted from the preprocessor, filtering is performed on the extracted region of interest to obtain a filter image. This refers to cow characteristic points, such as the cow's inscription, face line (face shape, face bottom, face weight, face pair, face texture), tail line (left tail, right tail, double tail, misjyeol), cow's hair color and nose skin color, cow's horn direction image, and mark number. By performing filtering processing, the weight value for the feature point can be increased.

For example, in the case of a cow's inscription, it is configured to implement optimized image processing according to the thickness of the inscription line. To this end, the filter unit 122 may apply a Gabor filter to the region of interest image preprocessed by the preprocessor 121 to output a filter image. The Gabor filter can output continuous data of the user's desired thickness and direction through the directionality of waves, and this can help extract the lines of the cow's inscription more effectively by changing the parameters of the Gabor filter.

The feature output unit 123 may extract a feature point image by extracting feature points from the filtered image output from the feature filter unit 122. The feature point output unit 123 may apply a shift algorithm to the filter image output from the filter unit 122 to extract feature points. The reason is that the shift algorithm can extract more stable feature points from the cattle inscription. At this time, the extracted feature points are at least one of the cow's inscription, face line (face shape, face bottom, face middle, face pair, face texture), tail line (left end, right end, double end, miseol), cow's hair color and nose skin color, cow's horn direction image, and mark number. Includes more.

The cattle entity determination unit 130 is configured to determine the cattle entity by comparing feature points extracted from the feature output unit 123 and images of a plurality of cattle stored in the database unit of the server 30. This makes it possible to determine the individuality of the corresponding cattle by comparing the characteristic points of the cow's face, the cow's inscription feature, and the cow's variable feature points extracted from the vision-based identification module 100 as described above with the feature points of the cattle individuals stored in the database.

The cattle entity determination unit 130 is configured to determine the cattle entity through image comparison and feature point matching. Specifically, the determination of the cattle entity through image comparison is based on the Euclidean distance (L2-Norm) between the plurality of feature points of the cattle extracted from the feature output unit 123 and the feature points of the image of each cattle entity stored in the database. Each can be calculated and compared.

For feature point matching, the feature point of the image with the smallest Euclidean distance value calculated by the image comparison unit 410 among the feature points of each of the plurality of axle images stored in the database may be selected as the matching feature point. This can be said to be a configuration that utilizes the characteristic of the Euclidean distance value that the Euclidean distance value of comparison objects (cattle) showing similar patterns in the present invention is small. Meanwhile, in addition to this, feature point matching may be configured to select additional matching feature points in descending order of Euclidean distance values calculated from image comparison in addition to the selected matching feature points, and then determine the matching reliability of the selected matching feature points.

In the present invention, an object discrimination module using an artificial neural network, for example, a convolutional neural network (CNN), may be used for image comparison and feature point matching. Artificial neural network is an information processing system that models the operating principles of biological neurons and the connection relationships between neurons. It is an information processing system in which multiple neurons, called nodes or processing elements, are connected in the form of a layer structure. Artificial neural network can refer to an overall model in which artificial neurons (nodes), which form a network through the combination of synapses, change the strength of the synapse connection through learning and have problem-solving capabilities.

The term artificial neural network may be used interchangeably with the term neural network, and the artificial neural network may include a plurality of layers, and each of the layers may include a plurality of neurons. Additionally, artificial neural networks may include synapses that connect neurons. Artificial neural networks generally use the following three factors: (1) connection patterns between neurons in different layers, (2) learning process to update connection weights, and (3) output values from the weighted sum of inputs received from the previous layer. It can be defined by the activation function it creates. Artificial neural network models as described above are, for example, CNN (Convolutional Neural Network), DNN (Deep Neural Network), RNN (Recurrent Neural Network), BRDNN (Bidirectional Recurrent Deep Neural Network), and MLP (Multilayer Perceptron). Network models may be included, but the present invention is not limited thereto.

Figure 8 is a flowchart showing the cattle object recognition process in the control unit of the user terminal using the RFID recognition module and vision-based recognition module as described above. As shown in FIG. 8, in step S110, the user first contacts the RFID tag attached to the cattle tag using the dedicated user terminal 20 to perform RFID identification. If the RFID identification result is successful in step S130, the step may proceed to S150 to transmit the coupon number to the management server, but in order to confirm the tagged cattle and its authenticity (for example, if the tag is If the tag is incorrectly attached to another cattle) or tag identification is impossible, the tag proceeds to step S140 and vision-based recognition can be performed.

In vision-based recognition, inscription identification, fixed feature point identification, and variable feature point identification may be performed sequentially as in steps S141 to S145. Vision-based recognition, which consists of three steps, can be configured to transmit a coupon number to the management server if any one of the three steps is successful when RFID tag identification is successful, and only if all of the three steps are successful will the coupon number be sent to the management server. It may also be configured to transmit a number.

If RFID tag identification fails, it may be configured to transmit the coupon number to the management server if any one of the three steps is successful. However, for more reliable cattle identification, the coupon number will be sent to the management server only if all of the three steps are successful. It may also be configured to transmit.

If tag identification is successful and any of the three vision-based recognition steps is not successful, the user terminal 20 determines that the tag attached to the recognized cattle and the biometric information (vision information) of the actual recognized cattle are different from each other. It is determined that this is the case, and an error message is transmitted along with the vision recognition result to the server 30 as in step S152.

When the server 30 receives the above-described error message, it searches the database using the vision recognition results of the corresponding cattle transmitted together, such as inscription identification results, hair color, pattern, nose skin color identification results, and facial feature point identification results. The identification data of the cattle corresponding to the vision recognition result is searched, and the RFID coupon number of the identified cattle is matched again with the vision recognition result and stored on the server.

Meanwhile, if the coupon number is recognized through tag identification, and the vision recognition result matches the recognized coupon number, the user terminal 20 transmits the coupon number to the management server 30, and the server 30 sends the received coupon number. In response, the corresponding cattle data is called and transmitted to the user terminal, or the data transmitted from the user is updated and stored in the server.

The cattle data stored on the server includes historical data such as the cattle's date of birth and breeding address, pedigree data including pedigree information of the mother cow/sister cow, slaughter data such as meat quality and meat grade data of the mother cow/sire cow, and dairy cow data. It may include public data including oil test data such as flow rate, oil quality, oil composition, etc.

In addition, cattle data stored on the server includes medical data, and medical data may include all health-related data such as breeding treatment data, infectious disease data, and metabolic disease data. In the case of disease data, data such as lesion location, repeated onset (parity, individual), disease duration, etc. can be additionally included.

Such data is stored in a database as shown in Figure 9, and the stored data is combined with equipment such as ICT equipment, for example, estrus detector (detection of activity level, rumination time, body temperature, etc.), automatic feeder, automatic milking machine, etc. This not only enables more efficient cattle management, but also provides the advantage of enabling continuous disease and health management for cattle individuals.

More specifically, Figure 9 is a diagram showing a system for comprehensive disease diagnosis and prescription. As shown in Figure 9, the diagnosis and prescription system includes a patient detection module, a data collection module, an integrated database, and a diagnosis and prescription module.

The diseased cattle detection module is provided to detect cattle suspected of having a disease. The cattle detection module utilizes data from estrus detectors, automatic feeders, automatic milking machines, and other ICT equipment for cattle. The estrus detector is configured to detect the cattle's activity level, rumination time, and body temperature, and the automatic feeder is provided to detect the feed intake by the cattle. Automatic milking machines are configured to detect the amount of milk being milked, flow rate, milk composition, blood milk, and somatic cells.

The patient detection module is configured to automatically detect cattle suspected of being diseased among the information stored in the integrated database. For example, among the data collected from the estrus detector, the data from the sick cattle detection module, in which the activity level and rumination time have decreased while the body temperature continues to remain higher than usual, is compared with data collected in the past to automatically indicate that the cattle individual is currently in a disease state. It can be detected with In addition, through the data collected from the automatic feeder, it is possible to detect abnormalities in cases where the amount of feed consumed by the cattle has drastically decreased compared to past data.

The data collected from the sick cattle detection module is managed through the cattle identification device and stored in an integrated database, and the integrated database includes the data collected from the sick cattle detection module, including the cattle's age and gender, breed/pedigree, and season/climate. /Cattle history information, such as local data and lesion history, is stored and managed for each cattle individual.

The lesion data collection module analyzes current data collected and stored through ICT equipment as described above by comparing it with past data, or detects abnormal values if the current data collected and stored through ICT equipment falls below or above the normal standard. If it is seen, notify the livestock owner, ranch manager, or veterinarian in charge and instruct them to accurately monitor the animal and enter the symptoms. The lesion data collected in this way is input into the diagnosis and prescription module.

In addition, the lesion data collection module can detect abnormal signs in cases where the amount of feed consumed by the cattle has drastically decreased compared to past data through data collected from the automatic feed feeder.

The diagnosis and prescription module transmits the data collected from the lesion data collection module, such as data containing symptoms/lesion locations, to the diagnosis and prescription module, which integrates the identification information of the cattle with suspected lesions. The database is searched to collect management history data of the corresponding cattle, and diagnosis and prescription are performed based on the collected information.

Data used in the diagnosis and prescription module include main symptoms, secondary symptoms, lesion location, age/sex of cattle, breed/pedigree, season/climate/region data, and disease duration data, and the diagnosis and prescription module includes the above-mentioned symptoms. Based on the items, the disease library is searched and automatic diagnosis and automatic prescription are performed.

However, if a diagnosis matching the aforementioned data items is not found by the disease library, the corresponding data is sent to the connected veterinarian's account on the cattle diagnosis platform, a detailed diagnosis is performed by the veterinarian, diagnosis and prescription are made, and the veterinarian is sent to the veterinarian to make a diagnosis and prescription. The results diagnosed by can be stored and managed for each cattle individual in an integrated database.

The device described above may be implemented with hardware components, software components, and/or the device described above may be implemented with hardware components, software components, and/or a combination of hardware components and software components. For example, devices and components described in the embodiments include, for example, a processor, a controller, an Arithmetic Logic Unit (ALU), a Digital Signal Processor, a microcomputer, and a Field Programmable Gate Array (FPGA). , may be implemented using one or more general-purpose computers or special-purpose computers, such as a Programmable Logic Unit (PLU), a microprocessor, or any other device that can execute and respond to commands. A processing device may execute an operating system (OS) and one or more software applications that run on the operating system. Additionally, a processing device may access, store, manipulate, process, and generate data in response to the execution of software. For ease of understanding, there are cases where a single processing device is described, but those skilled in the art will understand that the processing device includes multiple processing elements and/or multiple types of processing elements. It can be seen that it may include. For example, a processing device may include a plurality of processors or one processor and one controller. Additionally, other processing configurations, such as parallel processors, are also possible.

Software may include a computer program, code, commands, or a combination of one or more of these, and may configure a processing unit to operate as desired, or may be processed independently or collectively. You can command the device. Software and/or data may be used on any type of machine, component, physical device, virtual equipment, computer storage medium or device to be interpreted by or to provide instructions or data to a processing device. It can be embodied in . Software may be distributed over networked computer systems and stored or executed in a distributed manner. Software and data may be stored on one or more computer-readable recording media.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, etc., singly or in combination. Program instructions recorded on the medium may be specially designed and configured for the embodiment or may be known and available to those skilled in the art of computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic media such as floptical disks. -Includes optical media (Magneto-Optical Media) and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, flash memory, etc. Examples of program instructions include machine language code, such as that produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc.

As described above, although the embodiments have been described with limited examples and drawings, various modifications and variations can be made by those skilled in the art from the above description. For example, the described techniques are performed in a different order than the described method, and/or components of the described system, structure, device, circuit, etc. are combined or combined in a different form than the described method, or other components are used. Alternatively, appropriate results may be achieved even if substituted or substituted by an equivalent.

Therefore, other implementations, other embodiments, and equivalents of the claims should also be construed as falling within the scope of the claims described below.

10: Cattle identification device 11: RFID reader
12: Camera 20: User terminal
30: Management server 100: Vision recognition module
110: Image receiving unit 120: Feature point extraction unit
130: Cattle entity determination unit 121: Feature preprocessing unit
122: feature filter unit 123: feature output unit

Claims (6)

In a smart cattle management system for identifying cattle individuals,
cattle identification device; and a user terminal connected to the cattle identification device; And a control server that is connected to the user terminal through a network, receives cattle identification data transmitted from the user terminal, and provides cattle entity data to the user terminal in response,
The cattle identification device includes an RFID reader for reading an RFID tag provided on a cattle entity; and a camera for vision-based recognition,
The user terminal includes a vision recognition module for determining a cattle object from an image of the cattle transmitted from a vision recognition camera,
The vision recognition module includes an image receiving unit for receiving an image transmitted from a camera; a feature point extraction unit for detecting feature points of the axis from the image received by the image receiver; And a cattle individual determination unit for determining the cattle individual based on the feature points extracted from the feature point extraction unit,
The control unit of the user terminal first recognizes the coupon information of the cattle from the RFID tag using an RFID reader for the identified cattle entity, and when the coupon information of the cattle is recognized from the RFID tag information, it responds to the coupon information of the cattle to the management server. configured to call cattle data, and the control unit of the user terminal is configured to perform identification of the cattle entity using a vision recognition module if the mark information of the cattle is not recognized through the RFID reader,
The feature point extraction unit extracts feature points of the cow from the photographed image of the cow, wherein the feature points of the cow include an inscription feature point, a fixed feature point, and a variable feature point,
The fixed feature points include the shape of the whorl of the cow's face; Includes at least one of the shape of the facial eyebrows of the cattle, the color of the cattle, and the nose pattern - The shape of the facial whorls of the cattle includes face shape, mid-face, sub-face, pair of faces, and face texture, and the shape of the facial eyebrows of the cattle includes the presence or absence of whorls at the ends of the eyebrows. Includes the left, right, left, and unfinished forms according to - ;
The variable feature points include the horn shape of the cattle and the sign number of the cattle, and - the specific points regarding the horn shape of the cattle include upward angle, downward angle, deflection angle, incised angle, angled angle, and oblique angle depending on the horn direction of the cattle. - ;
The cattle entity determination unit is configured to perform an inscription feature point identification operation, a fixed feature point identification operation, and a variable feature point identification operation for the cattle entity,
The cattle entity determination unit first determines whether the cattle recognized from the facial image or image including the face of the cattle received from the vision recognition module is a Korean cattle or a dairy cow, and then searches for a fixed feature point identification operation if the recognized cattle is a Holstein dairy cow. It is further configured to perform a fixed feature point identification operation by further including the pattern shape as a feature point,
The cattle entity determination unit is configured to sequentially perform an inscription feature point identification operation, a fixed feature point identification operation, and a variable feature point identification operation to identify the target cattle and notify the server only when all feature points match, and the inscription feature point identification operation , configured to transmit an error message to the server if any of the fixed feature point identification operation and the variable feature point identification operation do not match,
The cattle entity determination unit calculates the Euclidean distance value between the cattle feature point received from the image receiver and the feature point stored in the database of the server, selects matching feature points in descending order of the Euclidean distance value, and then determines the matching reliability of the selected matching feature point. configured to identify a cattle individual,
The smart cattle management system is,
A disease detection module for detecting the health status of cattle; An integrated database storing historical data of cattle identified through the cattle identification device; a lesion data collection module that collects lesion data of cattle from the history data of the cattle; It includes a diagnosis and prescription module that diagnoses the current condition of the cattle and prescribes a prescription based on the lesion data and the cattle's history data,
The diagnosis and prescription module is based on the main symptoms, secondary symptoms, and lesion location obtained from the lesion data, and the age/gender, breed/pedigree, season/climate/region, and disease duration obtained from the cattle history data in the integrated database. Characterized in that it is configured to search the disease library based on the disease library and diagnose the disease in the order of matching from the disease library.
Smart cattle management system. delete delete delete delete delete