KR20210036625A - Smart farm livestock management system using code and management method thereof - Google Patents

Abstract

The present invention relates to a smart farm livestock management system and management method using a code, capable of creating an optimal growth environment by accurately recognizing cattle or pigs by using a QR code or a barcode, etc. as an individual information recognition code and managing the growth status of each population. The system comprises: a code application means for applying an individual information recognition code to livestock; a photographing means for photographing the individual information recognition code; an analysis means for analyzing the individual information recognition code photographed by the photographing means; and a management means for managing the livestock according to the individual information recognition code analyzed by the analysis means. The code application means implants a plurality of QR codes or barcodes on the back, flanks, etc. of piglets during the first vaccination after piglets are born. The photographing means provides a configuration that is a camera installed on the ceiling of a pig pen to photograph the QR code or barcode, which is the individual information recognition code. Therefore, it is possible to create an optimal growth environment for livestock.

Description

Smart farm livestock management system using code and management method thereof.

The present invention relates to a system and method for managing livestock such as cattle and pigs by using a code, and in particular, by using a QR code or barcode as an individual information recognition code, a cow or pig individual is accurately recognized, and the development status of each individual group It relates to a smart farm livestock management system and management method using a code that can create an optimal growth environment by managing the.

With the recent rapid development of IT technology, research in a new field called computer and electronics in agriculture is being conducted. The application of such convergence technology is more necessary because domestic farms are breeding densely, but IT technology is not appropriately applied to farms due to the small size of pig farming farms.

In general, pigs are raised while moving along the piglets, middle pigs, and finishing pigs according to their growth stage and size.There are about 150-200 piglets and about 50 finishing pigs in one pig room. The technology to manage the number and history (development status) of the pigs raised is important.

Currently, various methods such as RFID and image capture are used as a system that can accurately recognize the number of pigs in pig houses, but RFID is cumbersome to transplant for each individual (average of 3,000 heads per farm), and image extraction When pigs are overlapped, there is a problem that they are not properly recognized, and since the number of shipments is manually calculated when pigs are shipped, it is difficult to confirm the exact number of shipments.

In addition, in order to individually manage about 20 pigs in the pen, it must first be possible to individually identify the pigs. That is, there is a need for an image processing technology capable of classifying each pig into individual pigs from images acquired from a camera installed on the ceiling of a dense pig room. However, when two or more pigs of similar size or color are in close proximity, a problem has arisen that cannot be separated by conventional image processing technology, and for 24-hour automatic monitoring of individual pigs, it is necessary to quickly and accurately solve the problem of segmenting adjacent pigs.

An example of a technique for solving this problem is disclosed in the following patent documents.

For example, in Patent Document 1 below, as shown in FIG. 1, an image acquisition unit 11 for acquiring an image photographed by a camera 3 of a plurality of objects 2 in a direction from the ceiling to the floor of the barn (1). ), a storage unit 12 in which the acquired image is stored, an image processing unit 13 that removes noise from an image stored in the storage unit and generates a binarized image frame by binarizing the image from which the noise has been removed, and the Disclosed is an object detection apparatus 10 in a livestock house including an object detection unit 14 that searches for adjacent objects among moving objects by comparing the binarized image frames, and detects the adjacent objects individually. .

In addition, in Patent Document 2 below, the step of obtaining an image by photographing a group of objects through a camera, extracting the center point using the current frame of the obtained image and the previous frame in which the outline of the adjacent object is divided, based on the extracted center point. Generating time series data representing the distance to the outline, mapping time series data of the current frame and time series data of the previous frame using DTW (Dynamic Time Warping), mapping color information labeled on the outline of the previous frame A method of dividing an outline for separating adjacent pigs is disclosed in the steps of matching the outline of the current frame and determining the boundary line based on a point where colors are distinguished from the outline of the current frame.

On the other hand, Patent Document 3 below includes the step of obtaining a depth information image by photographing an object in space through a camera, and correcting the acquired depth information image, and an image and an object of the space containing the object in the corrected depth information image. Performing a difference image technique on an image of an unrecognized space, removing noise using morphological calculation and connection component analysis on an image that has been subjected to the difference image technique, and an object using SLIC algorithm and region expansion technique. Disclosed is a method of detecting an object including the step of detecting.

In addition, the following Patent Document 4 includes comparative information for analyzing image information such as a plurality of photographing units that generate image information of livestock raised in a livestock, the shape of livestock according to breed, behavioral characteristics, and weight of livestock according to breed and size. A database in which is stored, an image analysis unit that comprehensively analyzes a plurality of image information collected in real time based on the comparison information to generate information on the number of animals and weight information of each individual in the livestock, the number of individuals and the weight information A classification unit that classifies livestock by grade based on and calculates the number of livestock that can be shipped, the number of livestock information, the weight information, and one or more terminal devices capable of confirming the number of livestock classified by grade and the number of livestock that can be shipped. It is disclosed for a smart farm livestock management system based on machine learning.

Korean Registered Patent Publication No. 10-1720708 (registered on Mar. 22, 2017) Republic of Korea Patent Publication No. 10-1899241 (registered on September 10, 2018) Republic of Korea Patent Publication No. 10-1915893 (registered on October 31, 2018) Republic of Korea Patent Publication No. 10-2014353 (registered on August 20, 2019)

In the technology disclosed in the patent document as described above, when the pig manager in charge of feed supply and livestock environmental management is changed, the growth of pigs may be remarkably changed, and thus there is a problem that it is not possible to maintain consistency with respect to the growth of pigs.

An object of the present invention has been made to solve the above-described problems, and even when a pig manager in charge of feeding and housekeeping environmental management is changed, it is possible to consistently manage the growth of pigs by maintaining consistency with respect to growth. It is to provide a smart farm livestock management system and management method using code.

Another object of the present invention is to provide a smart farm livestock management system and management method using a code capable of executing history management through feed feeding and pig house management according to the individual development status of pigs.

Another object of the present invention is to provide a smart farm livestock management system and management method using codes that can be used for calculating asset value by accurately calculating the number of pigs per pig house after building the system and systematically managing each size until shipment. To provide.

In order to achieve the above object, the smart farm livestock management system using the code according to the present invention includes a code applying means for applying an object information recognition code to livestock, a photographing means for photographing the object information recognition code, and a photograph taken by the photographing means. An analysis means for analyzing the individual information recognition code, and a management means for managing the livestock according to the individual information recognition code analyzed by the analysis means, and the code application means is a QR code or barcode at the time of initial vaccination due to the birth of a piglet. A plurality of the piglets are implanted on the back, side, etc., and the photographing means is a camera installed on the ceiling of the pig room to photograph a QR code or barcode, which is the individual information recognition code.

In addition, in order to achieve the above object, the smart farm livestock management method using the code according to the present invention includes the steps of (a) applying a QR code or barcode as an individual information recognition code to livestock, (b) applying in the step (a). Photographing the QR code or barcode of the livestock, (c) analyzing the QR code or barcode photographed in the step (b), (d) according to the QR code or barcode analyzed by the step (c) Including the step of managing the history information of the livestock, and since the area of the QR code or the barcode increases according to the growth of the livestock, the reference data for accurately converting the area information into weight information is set in the step (c). It is done.

As described above, according to the smart farm livestock management system and management method using the code according to the present invention, the code applying means for applying the object information recognition code to the livestock and the analysis for analyzing the object information recognition code photographed by the photographing means By providing the means, the effect of creating an optimum growing environment for livestock is obtained.

In addition, according to the smart farm livestock management system and management method using the code according to the present invention, the size of the QR code is extracted using a QR code recognition camera installed in the upper part of the pighouse divided by the size of the pig, and from this, the size of the QR code is extracted. By establishing a recognizable system, the number of pigs per pig house can be accurately calculated, used for asset value calculation, and systematically used for size management until shipment.

1 is a block diagram for explaining the configuration of a conventional apparatus for detecting an object in a livestock house;
Figure 2 is a block diagram for explaining the configuration of a smart farm livestock management system using the code according to the present invention,
Figure 3 is a process chart for explaining the process of managing the smart farm livestock using the code according to the present invention,
4 is a view showing a state in which a QR code is applied to piglets in a smart farm livestock management system using a code according to the present invention;
Figure 5 is a view showing the QR code of the finishing pig in the smart farm livestock management system using the code according to the present invention.

The above and other objects and new features of the present invention will become more apparent from the description of the present specification and the accompanying drawings.

The term "smart farm" used in the present invention properly maintains and manages the growing environment of livestock using technologies such as Internet of Things, big data, and artificial intelligence, and grows remotely with a PC or smartphone. It means a smart house that can be automatically managed for each stage.

Hereinafter, embodiments according to the present invention will be described with reference to the drawings.

2 is a block diagram illustrating the configuration of a smart farm livestock management system using a code according to the present invention.

Smart farm livestock management system 100 using a code according to the present invention, as shown in Fig. 2, a code applying means 110 for applying an object information recognition code to livestock, a photographing means for photographing the object information recognition code (120), analysis means (130) for analyzing the object information recognition code photographed by the photographing means, and a management means (140) for managing the livestock according to the object information recognition code analyzed by the analysis means. .

The code applying means 110 allows a plurality of QR codes or barcodes to be implanted on the back and sides of the piglets at the time of initial vaccination after the piglets are born. Such transplantation can be carried out with a QR code or barcode at the birth of a piglet by a conventional transplantation method.

The photographing means 120 may include a configuration such as a camera installed on the ceiling of a money room to photograph a QR code or a barcode, which is the object information recognition code. In addition, the photographing means 120 may include a lighting unit that applies ultraviolet rays or the like, since it may stress the pig when the flash is used.

The analysis means 130 recognizes the QR code or barcode that is deformed according to the growth of the pig with respect to the QR code or barcode photographed by the photographing means 120. To this end, the analysis means 130 may be executed by programming to process the QR code or barcode photographed by the photographing means 120.

In other words, when a piglet is born and is initially vaccinated, if multiple QR codes are implanted on the pig's back and sides, the area of the QR code increases as the pig grows. However, the color of the QR code may be lightened, but there is no problem in recognizing the QR code with the camera.

A camera installed on the ceiling of a pig room recognizes the area information of the QR code and converts it into weight information, but the reference information that is converted into a weight value to increase the reliability of the converted value is obtained by collecting reference data through machine learning such as machine learning techniques. do. Also, area information can be obtained multiple times a day.

For example, if you enter birth information (year, month, day, hour), location information, etc. in the QR code in advance, you can directly recognize the object. In addition, a developmental status management system is established using a QR code and a separate management APP.

Since the basic information of the QR code cannot be updated unless a separate QR code is implanted, a separate input can be performed for some biometric information.

In other words, the volume, weight, growth process by date, environment information of the pig house, and the amount of pig house feeding can be automatically entered, but the vaccination date must be manually entered. In this case, the date of birth information previously entered in the QR code is used. Alarm services for individual vaccinations can be provided automatically.

When constructing an entity recognition system using a QR code in the analysis means 130, the checklist may be executed according to the following criteria.

That is, since the size of the pig room varies depending on the size of the farm or pig, the analysis means 130 defines a QR code recognition distance in advance, and when the pig room is dark at night or when the pig room is dark, it is difficult to recognize the QR code according to the performance of the camera. It can be used for temporary lighting such as a flash, but if the flash is used, it can stress the pig, so it can be checked by applying ultraviolet rays or the like.

While growing from piglets to finishing pigs, the QR code generated by the code applying means 110 may contain contaminants or the QR code may not be seen due to pig hair, so the analysis means 130 checks whether the QR code is well recognized. do.

In addition, since pigs move freely in the pig room, if the distance from the camera to the same object is different, the area information of the QR code may change, so it is necessary to correct the size of the QR code according to the distance.

In addition, the analysis means 130 must accurately represent the two-dimensional data as three-dimensional data when extracting the QR code. To this end, the number of QR codes and implantation positions are preset in the analysis means 130.

In order to apply the QR code as described above to the farm, the analysis means 130 executes the following settings as a preliminary preparation operation.

First, the code applying means 110 determines the size of the QR code implanted in the piglets to be recognized by the camera. That is, since the area of the QR code increases as the pig grows, reference data for accurately converting area information into weight information is set in the analysis means 130.

Meanwhile, the management means 140 determines the biometric data of the pig that can be implanted in the QR code, obtains additionally while converting the area information into weight information, and defines and applies biometric information values that can be automatically input.

Next, a smart farm livestock management method using the code according to the present invention will be described with reference to FIGS. 3 to 5.

3 is a process chart for explaining the process of managing livestock according to the present invention, and FIG. 4 is a view showing a state in which a QR code is applied to piglets in a smart farm livestock management system using a code according to the present invention. 5 is a view showing the QR code of the finishing pig in the smart farm livestock management system using the code according to the present invention.

In the livestock management method according to the present invention, a plurality of QR codes or barcodes are applied to the back and sides of the piglets as individual information recognition codes when livestock, for example, piglets are born and initially vaccinated (S10). Applying such a QR code or barcode can be performed in the same way as the process of implanting individual identification information into a normal piglet.

Next, the QR code or barcode of the livestock applied in step S10 is photographed during the growth process of piglets (S20). The photographing in step S20 may be performed at predetermined time intervals by the same configuration as the camera installed on the ceiling of the money room.

Subsequently, in step S20, the QR code or barcode photographed in the finishing pig as shown in FIG. 5 is recognized by the analysis means 130 as a QR code or barcode modified according to the growth of the pig. That is, the analysis means 130 analyzes the modified QR code or barcode to recognize the modified QR code or barcode compared to the QR code or barcode of piglets as shown in FIG. 4 (S30).

Thereafter, the history information of the livestock is managed according to the QR code or barcode analyzed by the step S30 (S40).

As described above, by analyzing the QR code or barcode photographed from the finishing pigs based on the QR code or barcode of the piglet by the analysis means 130, the development status of each pig population can be managed, so that the optimal pigs It can create a growing environment.

Although the invention made by the present inventor has been described in detail according to the above embodiment, the invention is not limited to the above embodiment, and it goes without saying that the invention can be changed in various ways without departing from the gist.

In other words. In the above embodiment, the growth environment of pigs has been described, but the present invention is not limited thereto, and can be applied to livestock such as cattle, deer, and horses that can grow by applying a QR code or barcode as an individual information recognition code. .

By using the smart farm livestock management system and management method using the code according to the present invention, it is possible to create an optimal growing environment for livestock.

100: livestock management system
110: code authorization means
120: shooting means
130: analysis means
140: management means

Claims (2)

Code approval means for applying the individual information recognition code to livestock
Photographing means for photographing the object information recognition code,
Analysis means for analyzing the object information recognition code photographed by the photographing means,
And a management means for managing the livestock according to the individual information recognition code analyzed by the analysis means,
The code applying means implants a plurality of QR codes or barcodes on the back and sides of the piglets at the time of the first vaccination after the piglets are born,
The photographing means is a smart farm livestock management system using a code, characterized in that the camera installed on the ceiling of the money room to photograph the QR code or barcode, which is the object information recognition code. (a) applying a QR code or barcode as an individual information recognition code to livestock,
(b) photographing the QR code or barcode of the livestock applied in step (a),
(c) analyzing the QR code or barcode photographed in step (b),
(d) managing the history information of the livestock according to the QR code or barcode analyzed by the step (c),
Since the area of the QR code or the barcode increases according to the growth of the livestock, in step (c), reference data for accurately converting the area information into weight information is set.

Images