KR102122131B1 - A livestock weighing system and a livestock weighing method using the same - Google Patents

Abstract

The present invention relates to a livestock weight measurement system and a livestock weight measurement method using the same, according to the present invention, the manager terminal for receiving livestock bio-information from the manager; The 3D scanner unit that acquires multiple 3D images by scanning livestock, and the manager terminal and the 3D scanner unit are interlocked, and receives multiple 3D images from the 3D scanner unit to derive the weight of the livestock through multiple 3D images It can provide a livestock weighing system comprising a livestock weighing server.
The method for measuring livestock weight using the method includes: (a) receiving bioinformation of livestock from a manager; (b) obtaining a plurality of three-dimensional images by scanning the livestock through a 3D scanner unit, and (c) a livestock weight measuring method comprising the step of the livestock weight measurement server deriving the weight of the livestock through the plurality of three-dimensional images. Can provide

Description

A livestock weighing system and a livestock weighing method using the same}

The present invention relates to a livestock weighing system and a livestock weighing method using the same, and more specifically, it is possible to simply and quickly measure the weight of the livestock using a 3D image obtained by scanning the livestock, thereby providing excellent accuracy and reliability. The present invention relates to a livestock weighing system and a livestock weighing method using the same.

In the case of the livestock industry, regular weight management is necessary for individual specification management of domesticated animals.

In particular, in the case of a pig farm, the standard is very important at the time of shipment, and it causes a very large difference in the income of the farm depending on the shipment of pigs that meet the standard. Pigs are graded according to the quantitative criteria according to weight and fat thickness, and the fat distribution of pork and the qualitative criteria according to meat color, and pigs ranging from 115kg to 120kg are generally called standard money. It is very important to accurately measure your weight and select shipping pigs because you can receive a higher grade if you satisfy the standard money.

For this, the need for periodic weight measurement or monitoring is required in the field.

Currently, pigs' weight is measured by the thorax positioning method and the money mold.

The thoracic positioning method converts the weight by applying the value obtained by measuring the chest of a money with a tape measure to the weight calculation formula, and is used as an advantage that there is no need to install a special facility, but the measurement error is very large.

In addition, the money machine is equipped with an accessory device on a scale for measuring the weight of the pig, and then measures the weight of the pig directly in a closed space. There is a problem that it takes a certain amount of time to stagnate within the sentence, and it takes a considerable amount of time to measure the weight of one animal in this process, which takes about 10 minutes or more on a per-worker basis. Trouble and maintenance were difficult.

In addition, there is a shortage of manpower due to the decrease and aging of the farming population, and measures for this are necessary.

Therefore, it is possible to easily and accurately measure the weight of the pig to continuously manage the weight of the pig and to reduce the labor force of the farm, thereby requiring technology to accurately predict the weight of the pig at the time of shipment.

In order to solve the above problems, the present invention can easily and quickly measure the weight of a livestock using a 3D image obtained by scanning a livestock, and thus the livestock weighing system excellent in accuracy and reliability and livestock weighing using the same In providing a way.

In order to solve the above problems, the livestock weight measuring system according to an embodiment of the present invention is a manager terminal that receives livestock biometric information from a manager; The 3D scanner unit that acquires multiple 3D images by scanning livestock, and the manager terminal and the 3D scanner unit are interlocked, and a number of 3D images are received from the 3D scanner unit to derive the weight of the livestock through the multiple 3D images It can provide a livestock weighing system comprising a livestock weighing server.

Here, the livestock weighing server, a pre-processing unit for extracting and optimizing a point cloud from a plurality of 3D images, and generating a single 3D image; A 3D construction unit for constructing 3D model data by forming a 3D isosurface using the point cloud of the single 3D image; A volume estimator for estimating the volume or length from the 3D model data and a weight measuring unit for converting the estimated volume or length into weight.

In addition, the pre-processing unit extracting the point cloud from each of the plurality of three-dimensional images; Characterized in that the step of removing the noise point (Noisy point) and overlap point (Overlap point) in the extracted point cloud and the step of generating a single three-dimensional image by matching the plurality of three-dimensional images.

In addition, the 3D construction unit constructs 3D model data by implementing a 3D isosurface with the points of the single 3D image through Poisson surface reconstruction and Marching cubes algorithms. It is characterized by.

In addition, the method for measuring livestock weight using a livestock weighing system according to an embodiment of the present invention includes a method for measuring livestock weight using a livestock weighing system, comprising: (a) receiving bioinformation of livestock from a manager; (b) obtaining a plurality of three-dimensional images by scanning the livestock through a 3D scanner unit, and (c) a livestock weight measurement method comprising the step of the livestock weight measurement server deriving the weight of the livestock through the plurality of three-dimensional images. Can provide

At this time, the step (c) is a pre-processing step of extracting and optimizing a point cloud from a plurality of 3D images and generating a single 3D image; Forming a 3D isosurface using the point cloud of the single 3D image to construct 3D model data; And estimating the volume or length in the 3D model data and converting the estimated volume or length into weight.

In addition, the pre-processing step of generating the single three-dimensional image may include extracting point clouds from the plurality of three-dimensional images, respectively; The method may include removing a noise point and an overlap point from the extracted cloud, and generating a single 3D image by matching the plurality of 3D images.

In addition, the step of constructing the 3D model data is realized by implementing a 3D isosurface with points of the single 3D image through Poisson surface reconstruction and Marching cubes algorithms. It is characterized by constructing dimensional model data.

The livestock weighing system according to the embodiment of the present invention and the livestock weighing method using the same implement a smart scale that can simply and quickly measure the weight of livestock using a 3D image obtained by scanning livestock. Thus, it is possible to provide a livestock weighing system having excellent accuracy and reliability and a livestock weighing method using the same.

Therefore, there is no need for additional equipment to measure the weight of the livestock, and it is possible to reduce the cost of breeding by continuously controlling the feed through the continuous weight management of the livestock, and to accurately predict the time of shipment, thereby increasing the profits of the farmers.

In addition, there is no hassle of stagnating for a certain period of time by inducing livestock to measure weight, thereby solving problems caused by lack of manpower, aging of the manpower, and enlargement of the scale.

1 is a block diagram showing a livestock weight measuring system according to an embodiment of the present invention.
Figure 2 is a block diagram showing a livestock weighing server of the livestock weighing system according to an embodiment of the present invention.
3 (a) and (b) is a conceptual diagram showing the estimated body length and chest when estimating the length in the livestock weighing system according to an embodiment of the present invention.
Figure 4 (a) and (b) is an exemplary view showing the circumference of the fine spacing and A used in estimating the volume in the livestock weighing system according to an embodiment of the present invention.
Figure 5 is a perspective view showing a 3D scanner unit of a livestock weighing system according to an embodiment of the present invention.
FIG. 6 is a partial perspective view showing a part of the 3D scanner of FIG. 5 projected;
Figure 7 is a perspective view showing the food supply of Figure 5;
8 is an exemplary view of using the food supply unit of FIG. 6.
9 is a flowchart schematically illustrating a method for measuring livestock weight using a livestock weight measurement system according to an embodiment of the present invention.
10 is a flowchart sequentially showing step S300 of FIG. 9.
11 is a flowchart sequentially showing step S310 of FIG. 10.

Hereinafter, the description of the present invention with reference to the drawings is not limited to a specific embodiment, and various conversions may be applied and various embodiments may be provided. In addition, it should be understood that the contents described below include all transformations, equivalents, and substitutes included in the spirit and scope of the present invention.

In the following description, terms such as first and second are terms used to describe various components, and are not limited in meaning to themselves, and are used only to distinguish one component from other components.

The same reference numerals used throughout this specification denote the same components.

The singular expression used in the present invention includes the plural expression unless the context clearly indicates otherwise. In addition, the terms "include", "have" or "have" described below are intended to designate the existence of features, numbers, steps, actions, components, parts or combinations thereof described in the specification. It should be interpreted and understood to not preclude the existence or addition possibility of one or more other features, numbers, steps, actions, components, parts or combinations thereof.

Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by a person skilled in the art to which the present invention pertains. Terms such as those defined in a commonly used dictionary should be interpreted as having meanings consistent with meanings in the context of related technologies, and should not be interpreted as ideal or excessively formal meanings unless explicitly defined in the present application. Does not.

In addition, in the description with reference to the accompanying drawings, the same reference numerals are assigned to the same components regardless of reference numerals, and redundant descriptions thereof will be omitted. In the description of the present invention, when it is determined that a detailed description of known technologies related to the present invention may unnecessarily obscure the subject matter of the present invention, the detailed description will be omitted.

Hereinafter, embodiments of the present invention will be described in detail with reference to FIGS. 1 to 11.

1 is a block diagram showing a livestock weight measuring system according to an embodiment of the present invention.

2 is a block diagram showing a livestock weighing server of the livestock weighing system according to an embodiment of the present invention.

3 (a) and (b) is a conceptual diagram showing the body length and chest estimated when estimating the length in the livestock weighing system according to an embodiment of the present invention.

4 (a) and (b) is an exemplary view showing the circumference of the micro-interval and A used when estimating the volume in the livestock weighing system according to an embodiment of the present invention.

5 is a perspective view showing a 3D scanner unit of a livestock weighing system according to an embodiment of the present invention.

6 is a partial perspective view showing a state in which a part of the 3D scanner of FIG. 5 is partially projected.

7 is a perspective view showing a food supply unit of FIG. 5.

8 is an exemplary view of using the food supply unit of FIG. 7.

Referring to Figure 1, the livestock weighing system according to an embodiment of the present invention may include a manager terminal 10, a 3D scanner unit 20 and a livestock weighing server 30.

First, the manager terminal 10 includes a livestock weighing application (or hair work app) implemented from the livestock weighing server 30, and may be a manager's mobile terminal such as a farmer's breeder, and a PC, tablet other than the mobile terminal , PDA (Personal Digital Assistant) may also be applied.

Here, the livestock weighing application that enables the use of the livestock weighing system means a general application based on Android and iOS. In addition, the livestock weight measurement application may be provided as a general application or a web service based application according to the manager terminal 10 or wired/wireless service type. As a method of providing, each terminal may access and download the livestock weighing server 30 or download and install it through an online application market (eg, an Android market, an Apple store, or an online market of a communication company).

Accordingly, the manager terminal 10 may receive livestock biometric information from a manager through a livestock weighing application and transmit it to the livestock weighing server 30.

Here, the bio-information of the livestock is the bio-information of the livestock to be weighed, and may include one or more of a livestock classification number, species, sex, and months, and preferably all.

In addition, the manager terminal 10 may receive a plurality of three-dimensional images obtained from the 3D scanner unit 20 from the livestock weighing server 30 and provide it to the manager.

In addition, the manager terminal 10 can receive the livestock data generated from the livestock weight measurement server 30 and provide it to the manager, whereby the manager can check in real time the weight of the livestock inputting the biometric information.

The 3D scanner unit 20 acquires a plurality of three-dimensional images by scanning livestock, and the 3D scanner unit 20 can be used separately, but is installed in the manager terminal 10 and directly measures the livestock weight measurement server 30 ).

Here, the 3D image includes the 2D image and the depth information, and the plurality of 3D images may be 3D images of the front, rear, left, and rear sides of the livestock, and are not limited thereto and scanned in various directions It may be a three-dimensional image.

The livestock weighing server 30 is implemented based on standard data generated using pre-collected livestock data, and is provided as a livestock weighing application to the manager terminal 10 so that the manager can use the livestock weighing system. do.

Here, the standard data is data standardized for each month by analyzing livestock data collected in advance by 3D scanning and measuring livestock at a livestock farm, and may include 3D model data for each month and weight accordingly. Livestock data may include livestock bio-information, 3D model data, and weight.

The livestock weight measurement server 30 interlocks the manager terminal 10 and the 3D scanner unit 20, receives a plurality of 3D images from the 3D scanner unit 20, and weighs the livestock through a plurality of 3D images. Can be derived.

To this end, as shown in FIG. 2, the livestock weight measurement server 30 may include a pre-processing unit 31, a 3D construction unit 32, a volume estimation unit 33, and a weight measurement unit 34.

The pre-processing unit 31 may extract and optimize points from a plurality of 3D images, and generate a single 3D image.

In detail, the pre-processing unit 31 extracts a point cloud, removes noise points and overlap points, and matches to generate a single 3D image. By performing, it is possible to preprocess a plurality of 3D images.

In the step of extracting fortune-telling, partial fortune-telling may be extracted from a plurality of 3D images. That is, in order to extract the body shape (shape) of a livestock from a number of 3D images, 3D point cloud is extracted.

In the step of removing the noise point and the overlapping point, a noise point and an overlap point may be removed from the extracted point cloud. This is to optimize the quality and accuracy of 3D model data to be built in the future.

Specifically, in the step of removing noise points and overlapping points, the pre-processing unit 31 may remove the noise points by using standard deviations and average values of all point clouds, and the noise points to be removed through Equations 1 and 2 Can be extracted.

[Equation 1]

R=

Here, t _{α /2} is a threshold of whether or not to be included in the body shape, t is the cloud point, n-2 is the degree of freedom, n is the size of the sample.

At this time, the size of the sample means the total number of collected clouds.

[Equation 2]

δ=│(X-mean(X))/s│

Here, X is a data value, mean (X) is an average value, and s is a standard deviation.

After calculating R (removal area) and δ through Equations 1 and 2 as described above, when δ> R, it may be determined that the noise point to be removed, and when δ ≤ R, it may be determined that it is not the noise point to be removed. Thereafter, only noise points corresponding to the object to be removed are removed.

In addition, since the pre-processing unit 31 is based on standard data, when removing noise points and overlapping points, it can be removed using standard data.

That is, it is possible to improve accuracy and reliability by determining and removing noise points and overlap points based on standard data, which is information on the body shape of a livestock, rather than simply extracting and removing noise points and overlap points from an image.

The step of matching and generating a single three-dimensional image may generate a single three-dimensional image by matching a plurality of optimized three-dimensional images.

The 3D construction unit 32 may construct a 3D model data by forming a 3D isosurface using a point cloud of a single 3D image.

Specifically, the 3D construction unit 32 uses a Poisson surface reconstruction and a Marching cubes algorithm to implement a three-dimensional isosurface with the fortune of a single three-dimensional image, a three-dimensional model Build data.

That is, the 3D construction unit 32 is capable of constructing a surface by implementing point clouds into Delaunay triangles or squares.

In more detail, triangles can be formed by connecting point clouds extracted from the 3D space (S), and the poles can be formed by finding and connecting the most distant vertices to each patch element, S. have. At this time, the set of formed poles is called P.

Then, the triangle is composed of a combination of a patch element (S) and a pole (pole), and after removing all of the triangles connected to the pole (pole), the remaining triangles can be connected to form a surface.

In addition, the 3D construction unit 32 may increase the accuracy by filling the data of the insufficient area because the error rate of the measured weight may increase due to the formation of the empty area when there is an area where there is insufficient data for extracting fortune.

Before implementing this, the shape of the livestock was checked for the characteristic of maintaining the symmetry on the basis of the equiaxed axis, and the data of the insufficient area was restored using the symmetry characteristic.

To this end, the 3D construction unit 32 may recover the blank area by filling the insufficient area with data using the Principle Component Analysis (PCA) technique.

Specifically, the 3D construction unit 32 calculates the center point (centroid) and the eigen vector of each cloud point, and then transforms the cloud point into a base point (P _b =(0,0,0)) as a target, thereby transforming You can create a new point cloud by flipping the cloud point about the vertical axis. Here, the vertical axis corresponds to the axis that forgets the tail from the head.

At this time, when the point cloud is transformed to the base point (P _b =(0,0,0)), the extracted eigen vector may be used to transform the eigen vector in the direction of the reference point.

Thereafter, the 3D construction unit 32 may finally recover the insufficient blank area by filling in the data using the data of the adjacent area if there is insufficient empty area even after restoring the empty area through the above process.

Accordingly, a fortune cloud is generated in a region where data is insufficient, and the blank region is restored, so that 3D model data close to the shape of the livestock can be constructed.

The volume estimator 33 may estimate the volume or length from the 3D model data to derive the weight of the livestock.

Here, when estimating the length, the volume estimator 33 may estimate the thorax and the body length as shown in FIG. 3, which refers to the length of the chest and back between the two axillas of the livestock, and the body length is the neck behind the ear. It means the length from to just before the tail, that is, the length of the body of the livestock.

Specifically, when estimating the thoracic of the length, the volume estimator 33 assumes that the curve that induces and rotates the center line from the head is the thoracic, and sets a number of points along the surface to extract the curve that rotates. You can estimate the thoracic by linking it.

At this time, it is expressed on the correct surface based on the last point, and the average value of all points within a certain interval can be applied to minimize the error between the points.

By applying the average value in this way, it is possible to estimate the thorax forming a smooth curve and reduce the error rate.

The weight measuring unit 34 may convert the estimated volume or length into weight to derive weight. Here, the method for deriving weight according to the estimated volume or length of the weight measuring unit 34 may be configured differently.

First, as an example, when estimating the volume in the volume estimator 34, the volume estimator 33 divides the volume into micro-intervals and obtains the volume as the sum of the micro-volumes for the section divided through Equation (3). , To derive the weight by converting the volume. The microvolume is the volume of each of the divided sections.

[Equation 3]

Volume =

Here, P _i is the circumference of the divided section, and t is the thickness of the divided section.

That is, when the estimated volume is divided into fine intervals as shown in FIG. 4(a), a divided cross section is formed, and one divided cross section (A) is shown in FIG. 4(b). It has a circumference. By multiplying the circumference of the divided cross-section and the fine spacing (thickness), the micro-volume of the divided cross-section is obtained, and then the total volume of the livestock is obtained as the sum of the micro-volumes, and then the volume can be converted into weight through a relational expression.

At this time, the relational expression that can derive the volume by weight is a statistical process of standard data to establish the relationship between weights by volume.

In addition, in this example, when estimating the length in the volume estimator 34, the weight estimator 33 may calculate and derive the weight through a relationship established using the length of the chest and the body length.

At this time, the established relational formula was established by formulating the relation of weight according to volume by statistical processing of standard data, and was established using Y=aX+b and the determination coefficient (R ² ).

Here, X was set as the chest as an independent variable, and Y was set as the body length as the dependent variable, and a relational expression was established to derive body weight as the chest and body length.

This relational expression can be established as, for example, Equation 4 below.

[Equation 4]

Weight = (chest constant x chest) + (length constant x length)

Here, the thoracic constant and the body length constant may be updated accordingly when the standard data is updated with constants derived by statistical processing of the standard data.

In this way, the weight measuring unit 34 can derive weight by calculating the weight of the livestock by substituting the thorax and the body length into the relationship established as in Equation 4 above.

Such a relational expression can be updated accordingly when standard data is updated as the livestock data measured and generated through the livestock weighing system is stored, thereby gradually improving accuracy and reliability.

With the above configuration, the weight measurement unit 34 can derive the weight of the livestock.

In addition, the livestock weight measurement server 30 may further include a transmission unit 35.

The transmission unit 35 may generate the received biometric information of the livestock, the generated 3D model data, and the derived weight as livestock data and transmit the generated livestock data to the manager terminal 10.

In addition, the livestock weighing system of the present invention may further include a control unit 40.

The control unit 40 may be linked to the livestock weight measurement server 30 to receive livestock data, store livestock data, and provide it to an administrator.

To this end, the control unit 40 includes a livestock weighing application (or hair work app) implemented from the livestock weighing server 30, and may be a PC of a manager such as a farmer's breeder, a mobile terminal other than a PC, a tablet, PDA (Personal Digital Assistant) may also be applied.

Accordingly, the control unit 40 may receive livestock data from the livestock weighing server 30 through a livestock weighing application, and an administrator can easily manage the weight of the livestock by receiving livestock data.

In addition, the control unit 40 includes a database (DB) to store the received livestock data, the database (DB) can be stored by sorting the received livestock data according to the number of months. In addition, the database DB can store standard data and can be updated by received livestock data.

Through such a control unit 40, the manager can monitor the weight of livestock in real time, determine the feed amount according to the weight, and determine the shipping time according to the standard.

Through this livestock weighing system, the livestock can be easily weighed within about 10 seconds.

In addition, it is possible to accelerate the delivery of high-quality products by monitoring the growth status through continuous weight management of livestock, thereby reducing the number of days of breeding and the cost of rearing.

In addition, it is possible to accurately predict the time of shipment, thereby increasing the profits of livestock farmers.

Meanwhile, referring to FIGS. 5 to 6, the 3D scanner unit 20 of the livestock weighing system according to an embodiment of the present invention includes a camera 21 capable of photographing livestock, and a livestock is installed by installing the camera 21. A cage 22 may be further included to efficiently acquire a 3D image.

When provided as the camera 21 as described above, since a 2D image is obtained, a process of converting a 2D image to a 3D image may be added, which may be repeatedly performed. Accordingly, a plurality of 3D images may be used by the livestock weight measurement server 30.

Further, the camera 21 may also be provided as a scanner.

The cage 22 may be provided with a camera 21, and may be formed of a rectangular parallelepiped frame with all surfaces opened so that livestock can be accommodated therein. At this time, all the surfaces are formed to be open because it is preferable to be designed so as not to overlap with the body of the housed animal because the accuracy of the three-dimensional image obtained by scanning may be degraded when there is a side that overlaps the livestock.

Here, the frame is preferably formed of an aluminum material, but is not limited thereto.

Specifically, the edge 22 may include a lower horizontal frame 220, a vertical frame 221, and an upper horizontal frame 222, and may further include an up and down adjustment unit 223.

The lower horizontal frame 220 is formed in an empty bar shape and is provided with four pieces, and may have a rectangular shape. The lower horizontal frame 220 is a portion supported on the ground.

The vertical frame 221 is formed in the shape of an empty bar, and is provided with four, vertically installed with respect to the ground, and connected vertically to the lower horizontal frame 220, each corner in the lower horizontal frame 220 of a rectangular shape Located in the four lower horizontal frame 220 can be connected.

The upper horizontal frame 222 is formed in an empty bar shape and is provided with four, and may be vertically connected between the four vertical frames 221. Accordingly, the upper horizontal frame 222 may have a rectangular shape.

A plurality of cameras 21 may be installed on the lower horizontal frame 220 and the upper horizontal frame 222 in the frame 22 configured as described above, but is not limited thereto.

In addition, the upper horizontal frame 222 may be moved up and down in the vertical frame 221, which adjusts the position of the camera 21 installed in the upper horizontal frame 222 according to the size of the livestock to move the livestock at a desired angle. It allows scanning.

The up and down adjustment unit 223 configured for this may include a rail unit 2230 and a fixing unit 2231.

The rail part 2230 is formed on each of the four vertical frames 221, and may be formed on all surfaces on which the upper horizontal frame 222 is connected in one vertical frame 221.

In addition, the rail portion 2230 forms a length vertically on the vertical frame 221, and is fastened to the fixing portion 2231 and allows the fixing portion 2231 to move up and down.

The rail part 2230 may include a rail 2230a. The rail 2230a is formed on both sides of the rail part 2230, and the concave part and the convex part may be alternately formed. In addition, the rail 2230a may be formed in a wavy shape by forming a gap between concave portions and a gap between convex portions.

Here, the rail 2230a may be formed symmetrically around the rail portion 2230.

Accordingly, the locking projections 2223c of the fixing part 2231 may move along the rail 2230a, and when the user presses the pressing part 2231b and applies force to the upper side or the lower side, it moves along the convex portion, and then moves to the concave portion. Upon reaching, the locking projection 2223c may be fixed by the lower convex portion.

The fixing portions 2231 are respectively formed inside the four upper horizontal frames 222, and both ends are inserted into the rail portions 2230 formed in the vertical frames 221 to move up and down along the rail portions 2230.

The fixing portion 2231 may include a connecting bar 2231a, a pressing portion 2231b, a locking projection 2223c, and an elastic member 2223d.

The connecting bar 2231a is formed to be long in the longitudinal direction of the upper horizontal frame 222, and two may be formed symmetrically.

One or more of each of the pressing portions 2231b may be formed on both sides around the width of the connecting bar 2231a, and may be symmetrically formed on both sides. In addition, the pressing portion 2231b may be formed to be exposed to the outside through both sides (left and right or front and rear) of the upper horizontal frame 222 so that the user can press it. At this time, a pressing hole is formed on both sides (left, right, or front and rear) of the upper horizontal frame 222 so that the pressing portion 2231b can be inserted therethrough.

The locking projections 2223c may be formed to be vertically projected to the outer ends of the two connecting bars 2231a, respectively. The engaging projection 2223c is inserted into the rail portion 2230 and installed in contact with the rail 2230a, and thus can move up and down, and is fixed to fix the position of the upper horizontal frame 222.

The locking protrusion 2223c may be formed to have a curvature on one side to facilitate moving up and down along the rail portion 2230.

The elastic member 2223d is provided between the connecting bar 2231a and the locking projection 2223c, and can support the connecting bar 2231a and the locking projection 2223c, so that the fixing portion 2231 is attached to the rail portion 2230. It can be fixed, it can also move up and down along the rail portion 2230.

That is, when the locking protrusion 2223c is pressed by the convex portion of the rail 2230a while the locking protrusion 2231c moves up and down, the elastic member 2231d contracts while the elastic member 2231d contracts and moves the locking protrusion 2223c. It may be easier, and when it is located in the concave portion, it is restored to its original state and pushes the locking projection 2223c so that the locking projection 2223c is fixed by the rail 2230a.

With this configuration, the height of the upper horizontal frame 222 in the rim 22 can be adjusted, so that the position of the camera 21 can be adjusted according to various sizes of livestock.

In addition, the cage 22 may include a food supply unit 23 for supplying food when scanning livestock.

The food supply unit 23 may feed the livestock to be scanned to improve the quality of the 3D image obtained by minimizing the movement of the livestock during scanning, thereby increasing the accuracy of the derived weight.

That is, since the livestock weighing system of the present invention derives the weight of the livestock through the obtained 3D image, the quality of the 3D image affects the accuracy of the derived weight, thereby minimizing the movement of the livestock and accurately scanning the livestock. It is necessary to do.

Water, sugar water, ice cream, etc. may be used as the food supplied for this, but is not limited thereto.

Specifically, referring to FIGS. 7 and 8, the food supply unit 23 may include a food container 230, a supply pipe 231, a food plate 232, and a receiving container 233.

The food container 230 is a container that accommodates water, sugar water, ice cream, and the like, and may include an inlet that can be opened and closed to fill the food.

The supply pipe 231 may be connected to the food container 230 to supply food to the food plate 232. When it is recognized that livestock is accommodated, the food is supplied from the food container 230 of the food supply unit 23 to the supply pipe 231 and provided to the food plate 232.

The food plate 232 is formed on the lower side of the supply pipe 231 to receive food from the supply pipe 231. The food plate has a predetermined depth and is formed in a plate shape so that the animal can easily ingest the food, and the rear side is inclined upward so that the food is guided to the front where the animal is.

The receiving container 233 is a container that can accommodate the head of the livestock when the animal tries to consume the food. In order to ingest the food provided by the livestock, the food plate 232 requires receiving the head in the receiving container 233. It is configured to eat.

That is, it is to prevent the livestock from moving as much as possible by inducing the livestock to put its head into the container 233 through the feed.

The receiving container 233 may be formed such that the front and upper surfaces are opened so that the livestock can insert their heads.

In addition, the receiving container 233 is made of transparent tempered glass, acrylic, and the like, so that the head of the livestock can also be scanned when scanning the livestock.

In addition, the receiving container 233 may be formed with a plurality of breathing holes 2330 at the bottom, so that moisture due to breathing of livestock is not occupied therein.

By using the food supply unit 23 configured as described above when scanning livestock, it is possible to obtain an accurate three-dimensional image without shaking by minimizing the movement of the livestock.

In addition, the cage 22 further includes a livestock detection sensor, and it can be controlled to supply food from the food supply unit 23 when it is recognized that the livestock is accommodated in the cage 22 through the livestock detection sensor.

The configuration of the 3D scanner unit 20 is not necessary, but may be provided for the enhancement of the obtained 3D image.

Hereinafter, a method for measuring livestock weight using the livestock weighing system described above will be described in detail.

9 is a flowchart schematically illustrating a method for measuring livestock weight using a livestock weight measurement system according to an embodiment of the present invention.

10 is a flowchart sequentially showing step S300 of FIG. 9.

11 is a flowchart sequentially showing step S310 of FIG. 10.

Referring to FIG. 9, a method for measuring livestock weight using a livestock weighing system according to an embodiment of the present invention includes receiving bioinformation of livestock (S100), obtaining a plurality of three-dimensional images (S200), and multiple It may include the step of deriving the weight of the livestock through the three-dimensional image of (S300).

Specifically, in step S100 of receiving the biometric information, the biometric information of the livestock may be received from the manager to the manager terminal 10 and transmitted to the livestock weighing server 30.

In step S200 of acquiring a plurality of 3D images, when biometric information of livestock is input to the manager terminal 10, the 3D scanner unit 20 is operated to scan the livestock to obtain a plurality of 3D images. When a plurality of 3D images are acquired by the 3D scanner unit 20, the livestock weight measurement server 30 may be transmitted.

At this time, in order to obtain an accurate 3D image, step S200 may be performed in a state in which food is supplied to livestock through the food supply unit of the 3D scanner unit 20.

In the step of deriving the weight of the livestock through the plurality of 3D images (S300 ), the livestock weight measurement server 30 receiving the plurality of 3D images may derive the weight of the livestock through the plurality of 3D images.

To this end, step S300 is a pre-processing step of generating a single 3D image (S310), a step of building 3D model data (S320), a step of estimating the volume or length (S330) and converting the volume or length into weight Step S340 may be included.

The pre-processing step (S310) of generating a single 3D image may extract and optimize point clouds from a plurality of 3D images, and generate a single 3D image to preprocess.

In step S310, specifically, extracting a point cloud (S311), removing a noise point (Noisy point) and an overlapping point (Overlap point) (S312), and matching to generate a single 3D image (S313). It can contain.

In step S311 of extracting the fortune cloud, the fortune cloud may be extracted from a plurality of 3D images. That is, in order to extract the body shape (shape) of a livestock from a number of 3D images, 3D point cloud is extracted.

In operation S312 of removing noise points and overlapping points, a noise point and an overlap point can be removed from the extracted cloud. This is to optimize the quality and accuracy of 3D model data to be built in the future.

Specifically, in the step of removing noise points and overlapping points, noise points can be removed using standard deviations and average values of all point clouds, and noise points to be removed can be extracted through Equations 1 and 2.

[Equation 1]

R=

Here, t _{α /2} is a threshold of whether or not to be included in the body shape, t is a point cloud, n-2 is the degree of freedom, and n is the size of the sample.

At this time, the size of the sample means the total number of collected clouds.

[Equation 2]

δ=│(X-mean(X))/s│

Here, X is a data value, mean (X) is an average value, and s is a standard deviation.

After obtaining R and δ through Equations 1 and 2 as described above, when δ> R, it can be determined as a noise point to be removed, and when δ ≤ R, it can be determined that it is not a noise point to be removed. Thereafter, only noise points corresponding to the object to be removed are removed.

The step of matching and generating a single 3D image (S313) may generate a single 3D image by matching a plurality of optimized 3D images.

In the step of constructing 3D model data (S320 ), 3D model data may be constructed by forming a 3D isosurface using a cloud of a single 3D image.

In step S320, 3D model data is constructed by implementing a 3D isosurface with the cloud of the single 3D image through Poisson surface reconstruction and Marching cubes algorithms.

More specifically, in step S320, three-dimensional model data can be constructed by forming triangles by connecting point clouds extracted from the three-dimensional space (S). Finding the most distant vertex for each patch element S It can be connected to form a pole. At this time, the set of formed poles is called P.

Then, the triangle is composed of a combination of a patch element (S) and a pole (pole), and after removing all of the triangles connected to the pole (pole), the remaining triangles can be connected to form a surface.

In addition, in step S320, when there is an area in which there is insufficient data to extract fortune, when an empty area occurs, the empty area may be restored by filling the insufficient area with data using a Principle Component Analysis (PCA) technique.

Specifically, in step S320, after calculating the centroid and eigen vector of each point cloud, the point cloud is transformed to the base point (P _b =(0,0,0)), and the transformed point cloud is vertical axis. You can create a new point cloud by symmetry as a reference.

At this time, when the point cloud is transformed to the base point (P _b =(0,0,0)), the extracted eigen vector may be used to transform the eigen vector in the direction of the reference point.

Subsequently, in step S320, if there are insufficient blank areas even after restoring the blank areas in the same manner as described above, the insufficient spaces may be finally restored by filling data with data of adjacent areas.

Accordingly, a fortune cloud is generated in a region where data is insufficient, and the blank region is restored, so that 3D model data close to the shape of the livestock can be constructed.

Step S330 for estimating the volume or length may estimate the volume or length from the 3D model data. When estimating the length from the 3D model data, it is to estimate the chest and body length.

Specifically, in step S330, when estimating the thorax, it is assumed that the curve that induces and rotates the center line from the head is the thorax, and sets a number of points along the surface and connects them to extract the curve that rotates. Can be estimated.

In addition, step S330 is expressed on the correct surface based on the last point, and in order to minimize the distance difference between points, the average value of all points within a certain interval is derived and applied to the extracted rotating curve to finally estimate the bust can do.

By applying the average value in this way, it is possible to estimate the thorax forming a smooth curve and reduce the error rate.

Step S340 of converting the volume or length into weight may convert the estimated volume or length into weight.

In step S340, when estimating the volume in step S330, the volume is divided into micro-intervals, the volume is obtained as the sum of the micro-volumes for the cross-section divided through Equation 3, and the volume can be converted to derive weight. The microvolume is the volume of each of the divided sections.

[Equation 3]

Volume =

t is the thickness of the divided cross section.

That is, the volume can be obtained through the above equation, and then the volume can be converted into weight through the relational equation. At this time, the relational formula that can derive the volume by weight is a statistical relationship with standard data to establish the relation between weights by volume.

In addition, in step S340, when estimating the length in step S330, the weight may be derived by calculating the weight through a relational formula established using the length of the chest and the body length.

The relational formula established here is to establish the relation of weight according to volume by statistical processing of standard data, and is established using Y=aX+b and the determination coefficient (R ² ).

In the above formula, X is set as the independent variable as the thorax, and Y as the dependent variable as the body length, thereby establishing a relational expression for the bust and body length.

This relational expression can be established as, for example, Equation 4 below.

[Equation 4]

Weight = (chest constant x chest) + (length constant x length)

Here, the thoracic constant and the body length constant may be updated accordingly when the standard data is updated with constants derived by statistical processing of the standard data.

In this way, the weight measuring unit 34 can derive weight by calculating the weight of the livestock by substituting the thorax and the body length into the relationship established as in Equation 4 above.

Through this step S300, the weight of livestock can be derived.

In addition, the method for measuring livestock weight according to an embodiment of the present invention may further include transmitting livestock data after step S300.

In the step of transmitting livestock data, the livestock weight measurement server 30 may generate livestock data including livestock bio-information, three-dimensional model data, and weight, and transmit the livestock data to the manager terminal 10, and also the control unit 40 Can transmit.

As described above, the livestock weight measurement system and the livestock weight measurement method using the same are smart scales that can measure the weight of livestock simply and quickly using a 3D image obtained by scanning livestock. By implementing, it is possible to provide a livestock weighing system having excellent accuracy and reliability and a livestock weighing method using the same.

Therefore, there is no need for additional equipment to measure the weight of the livestock, and it is possible to reduce the cost of breeding by continuously controlling the feed through the continuous weight management of the livestock, and to accurately predict the time of shipment, thereby increasing the profits of the farmers.

In addition, there is no hassle of stagnating for a certain period of time by inducing livestock to measure weight, thereby solving problems caused by lack of manpower, aging of the manpower, and enlargement of the scale.

The embodiment of the present invention described above is not implemented only through an apparatus and/or method, and is implemented through a program for realizing a function corresponding to the configuration of the embodiment of the present invention, a recording medium in which the program is recorded, and the like. Alternatively, such an implementation can be easily implemented by those skilled in the art to which the present invention pertains from the description of the above-described embodiments.

Although the embodiments of the present invention have been described in detail above, the scope of the present invention is not limited thereto, and various modifications and improvements of those skilled in the art using the basic concept of the present invention defined in the following claims are also provided. It belongs to the scope of rights.

10: administrator terminal
20: 3D scanner unit
21: Camera
22: Street
220: lower horizontal frame
221: vertical frame
222: upper horizontal frame
223: up and down control unit
2230: rail part
2230a: Rail
2231: fixing part
2231a: Connection bar
2231b: Pressing part
2231c: Jam
2231d: Elastic member
23: food supply
230: Feeder
231: supply pipe
232: feeding plate
233: Receptacle
2330: breathing ball
30: livestock weighing server
31: pre-processing unit
32: 3D construction
33: volume estimation unit
34: weight measuring unit
35: transmission unit
40: control unit

Claims (6)

A manager terminal that receives livestock biometric information from a manager;
3D scanner unit that acquires a number of three-dimensional images by scanning livestock and
And a livestock weighing server that interlocks the manager terminal and the 3D scanner unit, receives a plurality of 3D images from the 3D scanner unit, and derives the weight of the livestock through the plurality of 3D images.
The livestock weighing server,
A pre-processing unit that extracts and optimizes point clouds from a plurality of 3D images and generates a single 3D image;
A 3D construction unit for constructing 3D model data by forming a 3D isosurface using the point cloud of the single 3D image;
A volume estimator for estimating a volume or a length from the 3D model data, and
It includes a weight measuring unit for converting the estimated volume or length into weight,
The pre-processing unit,
Extracting a point cloud; Removing a noise point and an overlap point; Matching to generate a single 3D image; performing,
In the step of removing the noise point and the overlapping point, when removing the noise point and the overlapping point, it is removed using standard data,
The standard data,
This is standardized data for each month by analyzing livestock data collected in advance by 3D scanning of livestock at a livestock farm, and includes 3D model data per month and weight accordingly.
The collected livestock data,
Including livestock bio-information, 3D model data and weight,
The pre-processing unit,
In the step of removing the noise point and the overlapping point, the noise point can be removed using standard deviations and average values of all point clouds. If R, it is determined as a noise point to be removed, and if δ ≤ R, it is determined that it is not a noise point to be removed.
[Equation 1]
R=

(Where, t _α/2 is the threshold of whether or not to be included in the body type, t is the cloud point, n-2 is the degree of freedom, n is the size of the sample)
[Equation 2]
δ=│(X-mean(X))/s│
(Where X is the data value, mean (X) is the average value, and s is the standard deviation)
delete According to claim 1,
The volume estimation unit,
When estimating the length from the three-dimensional model data, the bust and body length are estimated,
When estimating the thoracic of the length, it is assumed that the curve that rotates by deriving the center line from the head is the thoracic, and the thoracic is estimated by setting and connecting a number of points along the surface to extract the curve that rotates. Livestock weighing system characterized by applying the average value of all points within the interval.
In the method for measuring livestock weight using a livestock weighing system,
(A) receiving the biological information of the livestock from the manager;
(b) scanning a livestock through a 3D scanner to obtain a plurality of 3D images, and
(c) the livestock weight measurement server includes deriving the livestock weight through a plurality of 3D images,
Step (c) is,
A pre-processing step of extracting and optimizing point clouds from a plurality of 3D images and generating a single 3D image;
Forming a 3D isosurface using the point cloud of the single 3D image to construct 3D model data;
Estimating a volume or length from the 3D model data, and
And converting the estimated volume or length into weight,
The pre-processing step of generating the single three-dimensional image,
Extracting point clouds from the plurality of 3D images, respectively;
Removing noise points and overlap points from the extracted point cloud; and
And matching the plurality of 3D images to generate a single 3D image,
Removing the noise point (Noisy point) and the overlap point (Overlap point) from the extracted point cloud,
When removing noise points and overlapping points, remove them using standard data.
The standard data,
This is standardized data for each month by analyzing livestock data collected in advance by 3D scanning of livestock at a livestock farm, and includes 3D model data per month and weight accordingly.
The collected livestock data,
Including livestock bio-information, 3D model data and weight,
Removing the noise point (Noisy point) and the overlap point (Overlap point) from the extracted point cloud,
The noise point can be removed using the standard deviation and average values of all point clouds. R(removal area) and δ are obtained through Equations 1 and 2 below, and when δ> R, it is determined as the noise point to be removed, δ If ≤ R, it is determined that it is not a noise point to be removed, and a method for measuring livestock weight is characterized by extracting and removing a noise point to be removed.
[Equation 1]
R=

(Here, t _α/2 is the threshold of whether or not to be included in the body type, t is the point cloud, n-2 is the degree of freedom, and n is the size of the sample.)
[Equation 2]
δ=│(X-mean(X))/s│
(Where X is the data value, mean (X) is the average value, and s is the standard deviation)
delete delete

Images