Livestock population shipping information calculation device and method, and shipping information calculation system comprising shipping information calculation device

Abstract

Disclosed are a livestock population shipping information calculation device and method, and a shipping information calculation system comprising the shipping information calculation device. The disclosed system comprises: a surveillance camera for periodically capturing area images of a farm; and a shipping information calculation device for calculating shipping information on the basis of the periodic area images transmitted from the surveillance camera. Here, the shipping information calculation device calculates, for each of the periodic area images, a count value of two or more livestock entity shapes included in the respective area images, and if the respective count values of a plurality of first area images among the periodic area images are all less than a preset threshold count value, calculates, as a shipping time of the livestock population, the capturing time of a second area image captured a preset first number of captures prior to the plurality of first area images among the periodic area images.

Description

A device and method for producing shipment information on a livestock population, and a shipment information producing system including the shipment information producing device

[0001] Embodiments of the present invention relate to a device and method for calculating shipment information for a livestock population, and a shipment information calculating system including the shipment information calculating device.

[0002] Poultry farming is one of the fastest-growing livestock industries, and its high economic viability has led to its adoption in most countries around the world, even those in developing countries. In particular, the domestic poultry market accounts for approximately 30% of domestic livestock production.

[0003] To achieve high production efficiency, chickens are raised in close proximity within a single farm. Farm managers monitor the weight of each chicken by age and predict the presence of disease in order to manage the chicken population. Specifically, managers measure the weight of each chicken by age to monitor changes in weight, thereby improving productivity.

[0004] Traditionally, managers would manually hold chickens, place them on scales, and measure their weight. However, this method suffers from inaccurate weight measurements, as the chickens constantly move around on the scale. Furthermore, because managers must manually measure the chickens, they require significant labor.

[0005] To address these issues, an image-based chicken weight prediction method has been developed. Conventional image-based chicken weight prediction methods acquire images of chickens via a camera and input the area of ​​the chicken image into a preset algorithmic model to estimate the chicken's weight.

[0006] However, the weight of chickens derived using a single algorithm model also has the problem of inaccuracy, and ultimately, it is not possible to accurately identify changes in the weight of chickens.

[0007] In particular, inaccurate weight measurements at the time of shipment can result in significant losses for chicken farms and affiliated companies during production and distribution. Therefore, a technology that can easily and accurately measure chicken weight at shipment is needed.

[0008] The purpose of the present invention is to provide a livestock population shipment information calculation device and method that determines the shipment time of a livestock population based on area images periodically captured by a surveillance camera installed in a farm and accurately estimates the shipment weight of the livestock population at the determined shipment time, and a shipment information calculation system including the shipment information calculation device.

[0009] The objectives of the present invention are not limited to those mentioned above, and other objectives and advantages of the present invention not mentioned above can be understood through the following description and will be more clearly understood through the embodiments of the present invention. Furthermore, it will be readily apparent that the objectives and advantages of the present invention can be realized by the means and combinations thereof set forth in the claims.

[0010] A system for calculating shipment information for a livestock population raised on a farm according to one embodiment of the present invention includes a surveillance camera that periodically captures area images of the farm, and a shipment information calculating device that calculates the shipment information based on the periodic area images transmitted from the surveillance camera. At this time, the shipment information calculating device calculates, for each of the periodic area images, a count value of shapes of two or more livestock individuals included in the area images, and when the count values ​​of each of a plurality of first area images among the periodic area images are all less than a preset threshold count value, calculates the shooting time of a second area image that was previously captured a first preset number of times from the plurality of first area images among the periodic area images as the shipment time of the livestock population.

[0011] A device for calculating shipment information of a livestock population raised on a farm according to one embodiment of the present invention includes a communication unit that receives a plurality of first zone images of the farm captured by a surveillance camera, and a processor that calculates a count value of shapes of two or more livestock individuals included in each of the plurality of first zone images, and when the count values ​​of each of the plurality of first zone images are all less than a preset threshold count value, calculates a shooting time of a second zone image captured by the surveillance camera a preset first number of times before from the plurality of first zone images as a shipment time of the livestock population.

[0012] A method for calculating shipment information of a livestock population raised on a farm according to one embodiment of the present invention includes the steps of receiving a plurality of first zone images of the farm photographed by a surveillance camera, calculating a count value of shapes of two or more livestock individuals included in each of the plurality of first zone images, and calculating a shipment time of the livestock population based on the shooting time of a second zone image photographed by the surveillance camera a first number of times previously from the plurality of first zone images when the count value of each of the plurality of first zone images is less than a preset threshold count value.

[0013] According to the present invention, the shipping time of a livestock population can be accurately determined without intervention of a farm manager based on area images periodically captured by surveillance cameras installed in the farm.

[0014] Additionally, according to the present invention, the shipping weight of a livestock population can be accurately estimated at the determined shipping time.

[0015] In addition, the effects of the present invention are not limited to the above-described effects, and should be understood to include all effects that can be inferred from the composition of the invention described in the detailed description or claims of the present invention.

[0016] FIG. 1 is a diagram schematically illustrating a configuration of a system for producing shipment information of a livestock population according to one embodiment of the present invention.

[0017] FIG. 2 illustrates an example of an area image captured by a surveillance camera according to one embodiment of the present invention.

[0018] FIG. 3 is a flowchart illustrating a method for calculating shipment information of a livestock population according to one embodiment of the present invention.

[0019] FIG. 4 is a drawing for explaining the concept of setting a threshold count value according to one embodiment of the present invention.

[0020] FIG. 5 is a diagram for explaining a concept of calculating the shipment time of a chicken population according to one embodiment of the present invention.

[0021] FIG. 6 is a drawing for explaining the concept of L weight calculation zone images used to calculate the shipment weight according to one embodiment of the present invention.

[0022] Figures 7 and 8 are diagrams showing detailed flowcharts of some of the steps in Figure 3.

[0023] The present invention is susceptible to various modifications and embodiments. Specific embodiments are illustrated and described in detail in the drawings. However, this is not intended to limit the present invention to specific embodiments, but rather to encompass all modifications, equivalents, and alternatives falling within the spirit and technical scope of the present invention. Throughout the description of each drawing, similar reference numerals have been used to designate similar components.

[0024] While terms such as "first," "second," etc. may be used to describe various components, these components should not be limited by these terms. These terms are used solely to distinguish one component from another. The term "and / or" includes any combination of multiple related listed items or any one of multiple related listed items.

[0025] The terminology used herein is merely used to describe specific embodiments and is not intended to limit the present invention. The singular expression includes the plural expression unless the context clearly indicates otherwise. In this application, it should be understood that the terms "comprise" or "have" indicate the presence of a feature, number, step, operation, component, part, or combination thereof described in the specification, but do not preclude the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof.

[0026] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. Terms defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and will not be interpreted in an idealized or overly formal sense unless explicitly defined herein.

[0027] Hereinafter, embodiments of the present invention will be described in detail with reference to the attached drawings. In this embodiment, the "livestock" is assumed to be chicken (broiler), a type of poultry. However, this embodiment can be equally applied to other poultry, such as ducks and geese, as well as other livestock.

[0028] FIG. 1 is a diagram schematically illustrating the configuration of a livestock population shipment information production system (1) according to one embodiment of the present invention.

[0029] Referring to FIG. 1, the shipment information generation system (1) may include a surveillance camera (10), a gateway server (20), and a cloud server (30).

[0030] A surveillance camera (10) may be installed in a farm where a population of chickens is raised in groups. For example, the surveillance camera (10) may be a CCTV camera.

[0031] According to an embodiment, a surveillance camera (10) may be installed on the ceiling of a farm, and the lens of the surveillance camera (10) may be rotated or moved 360°. In this case, when the lens is positioned vertically, the surveillance camera (10) may capture an image for a zone corresponding to the vertical direction, i.e., a zone image. In other words, the zone image may be a top view image for the zone.

[0032] Fig. 2 illustrates an example of an area image captured by a surveillance camera (10). Referring to Fig. 2, the area image in the form of a top view may include the shapes of some chickens among a population of chickens raised on a farm.

[0033] The surveillance camera (10) can repeatedly capture area images based on a preset cycle. For example, the capture cycle may be 5 minutes.

[0034] Meanwhile, as chickens age, their area increases, so the number of chickens included in a single area image may decrease. To accurately calculate the shipping weight of a chicken population, sufficient area data must be secured. Therefore, the imaging cycle may become shorter as the chickens age.

[0035] Depending on the size of the farm, one or more surveillance cameras (10) may be installed on the farm. For convenience of explanation, it is assumed that one surveillance camera (10) is installed on the farm.

[0036] Meanwhile, lighting can be installed on the farm, and the surveillance camera (10) can capture area images regardless of whether the lighting is turned on or off.

[0037] The surveillance camera (10) may include a communication unit for transmitting periodically captured area images to a gateway server (20).

[0038] Again, referring to FIG. 1, the gateway server (20) may be an edge terminal provided on the farm side.

[0039] The gateway server (20) can calculate the shipping time of the chicken population based on area images periodically captured by the surveillance camera (10), and estimate or calculate the shipping weight of the chicken population at the calculated shipping time. In addition, the gateway server (20) can transmit the shipping time and shipping weight to the cloud server (30) via the network (40). That is, the gateway server (20) can correspond to a shipping information calculation device that calculates shipping information (shipping time and shipping weight) of the livestock population.

[0040] According to an embodiment, the gateway server (20) can calculate the count value of a plurality of chicken individuals from each of the area images periodically captured by the surveillance camera (10), and calculate the shipping time of the chicken population based on the calculated count value of the area image. Here, the count value may be information indicating how many chicken individuals exist within the area image.

[0041] According to an embodiment, the gateway server (20) can estimate the weight of a chicken population by inputting information extracted from each of a plurality of sequential area images captured by a surveillance camera (10) into a pre-learned integrated algorithm model.

[0042] Here, the integrated algorithm model is an algorithm model that inputs area information of a chicken individual and outputs the weight of the chicken individual, and can be received from a cloud server (30) via a network (40). A detailed description of the integrated algorithm model will be described later.

[0043] Although not shown in FIG. 1, the gateway server (20) may include a communication unit, memory, and a processor.

[0044] The communication unit of the gateway server (20) can be connected to a surveillance camera (10) and a cloud server (30). For example, the communication unit of the gateway server (20) can include at least some of a wired communication module, a short-range wireless communication module, and a long-range wireless communication module.

[0045] The memory of the gateway server (20) may be volatile and / or non-volatile memory and may store commands or data related to at least one other component of the gateway server (20). In particular, the memory of the gateway server (20) may store commands or data related to a computer program, application, recording medium, etc. executed in the gateway server (20).

[0046] The processor of the gateway server (20) may include one or more of a central processing unit, an application processor, or a communication processor. The processor of the gateway server (20) may perform calculations or data processing related to the control and / or communication of at least one other component of the gateway server (20). In particular, the processor of the gateway server (20) may execute instructions of a computer program / application.

[0047] Referring again to FIG. 1, the cloud server (30) may be a central terminal provided on the external side of the farm.

[0048] Similar to the gateway server (20), the cloud server (30) may include a communication unit, memory, and a processor. A detailed description thereof is omitted.

[0049] The cloud server (30) can be connected to the gateway server (20) to receive shipment information of the chicken population and transmit it to related companies such as slaughterhouses.

[0050] Additionally, the cloud server (30) may be a learning device for learning an integrated algorithm model used to estimate the weight of a chicken population. Learning may be performed periodically, thereby updating the integrated algorithm model. The learned integrated algorithm model may be transmitted to the gateway server (20).

[0051] Although not illustrated in FIG. 1, the cloud server (30) can communicate with multiple gateway servers (20) located in each of the multiple farms, thereby monitoring the breeding status of chicken populations raised in each of the multiple farms. Furthermore, the cloud server (30) can receive shipment information on chicken populations from multiple farms and transmit it to related businesses, such as slaughterhouses.

[0052] Meanwhile, according to another embodiment, the production of shipment information for a chicken population may be performed on a cloud server (30) rather than a gateway server (20), and the learning of the integrated algorithm model may be performed on a gateway server (20) rather than a cloud server (30).

[0053] Below, the operation of the shipment information generation device to generate shipment information will be described in more detail. For convenience of explanation, it is assumed that the shipment information generation device is a gateway server (20).

[0054] FIG. 3 is a flowchart illustrating a method for calculating shipment information of a livestock population according to one embodiment of the present invention.

[0055] The steps illustrated in FIG. 3 may be performed within a predefined first period. The first period may have a length of 24 hours. For example, the first period may start at 14:00 on a specific day, and the first period may end at 13:59 on the day following the specific day. In another example, the first period may start at 00:00 on a specific day, and the first period may end at 23:59 on the day following the specific day.

[0056] Below, the process performed at each step is described with reference to Fig. 3.

[0057] In step (S102), the gateway server (20) can receive a current area image from the surveillance camera (10).

[0058] In step (S104), the gateway server (20) can extract two or more chicken object shapes from the current area image.

[0059] According to an embodiment, the gateway server (20) may compare a plurality of chicken object shapes included in the current area image with a predefined reference chicken object shape to select two or more chicken object shapes from among the plurality of chicken object shapes. For example, the two or more chicken object shapes may be chicken object shapes whose similarity with the reference chicken object shape exceeds a threshold ratio (for example, 90%).

[0060] In step (S106), the gateway server (20) can calculate and store area information from each of two or more chicken object shapes.

[0061] In an embodiment, the area information may correspond to the number of pixels in a portion of the area image corresponding to the chicken shape.

[0062] In step (S108), the gateway server (20) can count the number of chicken object shapes of two or more to calculate the count value of the current area image.

[0063] In step (S110), the gateway server (20) can determine whether the count value of the current zone image is less than a preset threshold count value.

[0064] The threshold count value may be a reference count value for determining whether a chicken population has been released. The threshold count value may change over time.

[0065] According to an embodiment, the threshold count value may be set based on the distribution of count values ​​of a plurality of zone images (i.e., third zone images) captured by the surveillance camera (10) in a second period immediately prior to the first period. Here, no shipment of the chicken population is performed in the second period.

[0066] FIG. 4 is a drawing for explaining the concept of setting a threshold count value according to one embodiment of the present invention.

[0067] Referring to FIG. 4, if the current time is "7 / 02 06:00", the start time of the first period may be "7 / 01 14:00", the end time of the first period may be "7 / 02 13:59", and the second period may be set from "6 / 30 14:00" to "7 / 01 13:59". In this case, the threshold count value may be set based on the distribution of multiple area images captured from "6 / 30 14:00" to "7 / 01 13:59".

[0068] According to an embodiment, the threshold count value may be set based on the average and variance of the count values ​​of the plurality of area images captured in the second period. For example, the threshold count value may correspond to the difference (m-3σ) of the average and three standard deviations (three standard deviations) of the count values ​​of the plurality of area images captured in the second period. That is, it is based on the rule that for a specific distribution, almost all values ​​exist within a range of three standard deviations (3σ) on either side of the average (m). However, the present invention is not limited thereto.

[0069] Meanwhile, each of the multiple area images captured in the second period can be captured by the surveillance camera (10) when the lighting is on and the surveillance camera (10) is not contaminated. This is described below.

[0070] Referring back to FIG. 3, if the count value of the current zone image is greater than or equal to a threshold count value, then in step S112, the gateway server (20) may determine that the chicken population is not being shipped. For example, if the count value of the current zone image is "24" and the threshold count value is "9," then since there are a large number of chickens in the farm, the gateway server (20) may determine that the chicken population is not being shipped. In this case, the process may be repeated from step S102 for the next zone image.

[0071] Conversely, if the count value of the current zone image is less than the threshold count value, in step (S114), the gateway server (20) can determine whether the count value of the zone image immediately preceding the current zone image is less than the threshold count value.

[0072] At this time, there may be one or more immediate area images of the current area image. For example, referring to FIG. 4, two immediate area images captured immediately before the current area image may be used. In this case, step (S114) may determine whether all count values ​​of one or more immediate area images are below a threshold count value.

[0073] At this time, if the count value of the previous zone image is greater than or equal to the threshold count value, in step (S112), the gateway server (20) may determine that the chicken population is not shipped. For example, referring to FIG. 4, if the threshold count value is "9", the count value of the current zone image ("7 / 02 06:00") is "8", and the count values ​​of the two previous zone images ("7 / 02 05:50" and "7 / 02 05:55") are "24" and "7", respectively, the gateway server (20) may determine that the chicken population is not shipped. In this case, the process may be repeated from step (S11) for the next zone image.

[0074] Alternatively, if the count value of the previous zone image is less than the threshold count value, in step (S116), the gateway server (20) can determine that the chicken population has been shipped.

[0075] For example, referring to FIG. 4, if the threshold count value is "9", the count value of the current zone image ("7 / 02 06:00") is "8", and the count values ​​of the two previous zone images ("7 / 02 05:50", "7 / 02 05:55") are "6" and "7", respectively, the gateway server (20) can determine that the chicken population has been shipped.

[0076] In summary, among the periodic zone images captured in the first period, the current zone image and one or more immediately preceding zone images can constitute a plurality of first zone images. In this case, if all count values ​​of the plurality of first zone images are less than a threshold count value, the gateway server (20) can determine that the chicken population has been shipped. Alternatively, if the count value of any one of the plurality of first zone images is greater than or equal to the threshold count value, the gateway server (20) can determine that the chicken population has not been shipped.

[0077] That is, the gateway server (20) can determine whether or not to ship a chicken population using multiple first-zone images, rather than using a single first-zone image. This can reduce errors that may occur when determining whether or not to ship a chicken population. In addition, by determining whether or not to ship a chicken population using first-zone images periodically captured by a surveillance camera (10), it is possible to accurately determine whether or not to ship a chicken population without the intervention of a farm manager.

[0078] Hereinafter, for convenience of explanation, a group of zone images including a current zone image and one or more immediately preceding zone images is referred to as the “multiple first zone images” described above.

[0079] Continuing, at step (S118), the gateway server (20) can calculate the shipping time of the chicken population.

[0080] Here, the shipping time of the chicken population can correspond to the reference time for calculating the shipping weight of the chicken population.

[0081] Specifically, the gateway server (20) can calculate the shooting time of the second zone image, which was previously shot a preset first number of times from a plurality of first zone images, as the shipping time of the chicken population.

[0082] In an embodiment, the second zone image may be a zone image captured a first number of times before the first zone image, which was captured the fastest among the plurality of first zone images. For example, if the capture cycle is 5 minutes, the first number of times may be 12.

[0083] Meanwhile, the surveillance camera (10) can capture an area image regardless of whether the lights in the farm are turned on or off, and can also capture an area image regardless of whether the surveillance camera (10) is contaminated. The area image captured in this way is transmitted to the gateway server (20), and the gateway server (20) can determine whether the lights are turned on or off and whether the surveillance camera (10) is contaminated based on the area image. At this time, a predefined lighting on / off determination algorithm and a contamination determination algorithm can be used. For example, the gateway server (20) can determine whether the lights are turned on or off by analyzing the RGB of the pixels constituting the area image, and can input the area image into a pre-trained artificial neural network-based contamination determination algorithm model to determine whether the surveillance camera (10) is contaminated.

[0084] Referring to the above, when selecting a second zone image corresponding to the shipment time of the chicken population, the gateway server (20) may exclude zone images captured when the lights are turned off and zone images captured when the surveillance camera (10) is contaminated when counting the first number of times. That is, the gateway server (20) may exclude, from among the previous zone images of the plurality of first zone images, the previous light-off zone image captured when the lights are turned off and the previous contaminated zone image captured when the surveillance camera (10) is contaminated from the counting of the first number of times.

[0085] Figure 5 is a diagram illustrating a concept for calculating the shipment time of a chicken population according to one embodiment of the present invention. In this case, it is assumed that the first number of times is 12.

[0086] Fig. 5 (a) illustrates an example of a zone image captured in a first period when the light is on and the surveillance camera (10) is not contaminated. Referring to Fig. 5 (a), three zone images captured from "7 / 01 20:40" to "7 / 01 20:50" may be a plurality of first zone images, and the gateway server (20) may select the zone image of "7 / 01 19:40" captured 12 times earlier than the zone image of "7 / 01 20:40", which is the first zone image captured earliest among the plurality of first zone images, as a second zone image, and may calculate "7 / 01 19:40" as the shipping time of the chicken population.

[0087] In addition, Fig. 5 (b) illustrates an example of a zone image captured in the first period when the lights are turned off only from "7 / 02 01:50" to "7 / 02 05:45" for the sleeping of the chicken population and the surveillance camera (10) is not contaminated. Referring to Fig. 5 (b), three zone images captured from "7 / 02 05:50" to "7 / 02 06:00" may be a plurality of first zone images, and the gateway server (20) may find a second zone image captured 12 times earlier from the zone image of "7 / 02 05:50", which is the first zone image captured earliest among the plurality of first zone images. At this time, since the area image in which the light is turned off is excluded when calculating the count of 12, the gateway server (20) can select the area image of "7 / 02 00:50" as the second area image, and calculate "7 / 02 00:50" as the shipment time of the chicken population.

[0088] Referring again to FIG. 3, at step (S120), the gateway server (20) can calculate the shipping weight of the chicken population at the shipping time.

[0089] Specifically, the gateway server (20) can calculate the shipping weight of the chicken population at the shipping time based on L (an integer greater than or equal to 2) weight calculation zone images composed of a second zone image and one or more fourth zone images sequentially captured immediately before the second zone image, a pre-learned integrated algorithm model, and L time weights corresponding to the shooting times of each of the L weight calculation zone images. Here, the integrated algorithm model can include N (an integer greater than or equal to 2) pre-learned algorithm models and N combination weights for combining the N algorithm models.

[0090] FIG. 6 is a drawing for explaining the concept of L weight calculation zone images used to calculate the shipment weight according to one embodiment of the present invention.

[0091] Figure 6 illustrates images of weight calculation zones when the lights are on, the surveillance camera (10) is not contaminated, and L is "72". In this case, the gateway server (20) can calculate the shipping weight of the chicken population at the shipping time based on the 72 weight calculation zone images, the integrated algorithm model, and the 72 time weights.

[0092] Meanwhile, similar to the above, the L weight calculation area images may be area images captured by a surveillance camera (10) in which the farm's lights are turned on and there is no contamination.

[0093] Hereinafter, the process performed in step (S120) will be described in more detail with reference to FIGS. 7 and 8.

[0094] Figure 7 is a drawing showing a detailed flowchart of step (S120) among the steps of Figure 3.

[0095] Below, the process performed at each step is described with reference to Fig. 7.

[0096] In step (S121), the gateway server (20) can calculate L first weights corresponding to each of the L weight calculation zone images based on the L weight calculation zone images and the integration algorithm.

[0097] Here, the first weight may be related to the weight of two or more chicken individuals included in the area image for weight calculation.

[0098] Hereinafter, step (S121) will be described in more detail with reference to FIG. 8.

[0099] Figure 8 is a drawing showing a flowchart of step (S121) among the steps of Figure 7.

[0100] The steps of step (S121) described in FIG. 8 can be performed for each of L weight calculation zone images. For convenience of the following description, among the L weight calculation zone images, the weight calculation zone image for which step (S121) is performed is referred to as a “first weight calculation zone image.”

[0101] In step (S1211), the gateway server (20) can input area information of two or more chicken individuals into the integrated algorithm model to produce a first weight output of the two or more chicken individuals. Meanwhile, the area information of the two or more chicken individuals is information stored in the above-described step (S106).

[0102] That is, the first weight output of a chicken individual may be the output value of an integrated algorithm model that is produced when the area information of the chicken individual is input into the integrated algorithm model. The specific details of producing the first weight output of a chicken individual using the integrated algorithm model are described below.

[0103] In step (S1212), the gateway server (20) can calculate a representative weight output of the first weight output of two or more chicken individuals as a first weight corresponding to the first weight calculation area image.

[0104] According to an embodiment, a representative weight output of the first weight output of two or more chicken individuals may correspond to an average output of the first weight output of the two or more chicken individuals.

[0105] In another embodiment, the representative weight output of the first weight output of the two or more chicken individuals may be a weighted sum of the first weight output of the two or more chicken individuals and at least one preset weight. Here, the weight may be set to be proportional to the similarity between the shape of the reference chicken individual and the shape of the chicken individual.

[0106] Hereinafter, the step (S1212) of calculating the first weight output of a chicken individual using an integrated algorithm model is described in more detail.

[0107] The integrated algorithm model is an algorithm model that inputs area information of a chicken individual and outputs the weight of the chicken individual, and can be transmitted to the gateway server (20) after being trained in advance in a cloud server (30).

[0108] According to an embodiment, the integrated algorithm model may be composed of N algorithm models and N combination weights. In this case, each of the N algorithm models may correspond to each of the N combination weights.

[0109] Each of the N algorithm models may be an algorithm model that inputs the area of ​​a chicken population and outputs the weight of the chicken population, and the N algorithm models may be generated using different modeling methods. In addition, each of the N algorithm models may be trained in a cloud server (30) based on area-weight information pairs acquired prior to the time of estimating the weight of the chicken population. As an example, each of the N algorithm models may be a regression analysis model, an artificial neural network model, etc.

[0110] That is, the gateway server (20) can input area information of a chicken individual into each of N algorithm models to produce N weight outputs, and produce the weighted sum of the N combined weights and the N weight outputs as the weight of the chicken individual. This can be expressed by the following mathematical expression 1.

[0111]

[0112]

[0113]

[0114] Here, W C is the weight of the produced chicken individual, α i is the i-th combination weight corresponding to the i-th algorithm model among N algorithm models, MO i Each represents the i-th weight output that is output when the area information of the chicken individual is input to the i-th algorithm model. And, α i can have a value near 1.

[0115] Accordingly, in the above-described step (S1212), the gateway server (20) can input area information of each of two or more chicken individuals into N algorithm models to produce N second weight outputs, which are weight outputs of each of the N algorithm models, and can output the first weight of the chicken individual as a weighted sum of the N combined weights and the N second weight outputs.

[0116] Meanwhile, according to an embodiment, the N algorithm models may include two or more algorithm models among the first, second, and third algorithm models.

[0117] The first algorithm model may be an algorithm model learned by a cloud server (30) based on a plurality of first area-weight information pairs corresponding to pairs of area and weight of a single chicken. For example, the first algorithm model may be an artificial neural network model.

[0118] Specifically, the area and weight of a single chicken can be calculated based on at least one area image and a plurality of weight data captured and measured by a biometric data acquisition device provided in the farm during a period A prior to the time of calculating the shipment weight.

[0119] The second algorithm model may be an algorithm model learned by the cloud server (30) based on a plurality of second area-weight information pairs corresponding to pairs of representative areas of X (an integer greater than or equal to 2) chicken individuals and representative weights of Y (an integer greater than or equal to 2) chicken individuals. For example, the second algorithm model may be a regression analysis model.

[0120] Specifically, the representative area of ​​X chicken individuals can be calculated based on a plurality of learning area images captured by a surveillance camera (10) of the farm during a preset period B prior to the aforementioned shipping weight calculation time point. And, the representative weight of Y chicken individuals can be calculated based on a plurality of weight data measured by a weight measurement module of a biometric data acquisition device installed in the farm where the surveillance camera (10) is installed during the period B. At this time, the surveillance camera (10) and the biometric data acquisition device may be installed at different points in the farm, and thus, X chicken individuals and Y chicken individuals may not be identical to each other.

[0121] More specifically, the surveillance camera (10) can transmit a plurality of learning area images captured during period B to the gateway server (20), and the biometric data acquisition device can transmit a plurality of weight data measured during period B to the gateway server (20).

[0122] After this, the gateway server (20) can extract X chicken individual shapes from each of the plurality of learning area images, calculate the area of ​​X chicken individuals from the X chicken individual shapes, and calculate the average area or weighted average area of ​​the areas of X chicken individuals as the representative area of ​​X chicken individuals. Meanwhile, similarly to the above, each of the X chicken individual shapes can have a similarity rate exceeding a critical ratio with the reference chicken individual shape.

[0123] In addition, the gateway server (20) can accumulate multiple weight data to generate a histogram based on weight and frequency, and calculate a representative weight (for example, an average weight or a weighted average weight) of Y chicken individuals based on the generated histogram.

[0124] The gateway server (20) can transmit a second area-weight information pair corresponding to the pair of representative areas of X chicken individuals and representative weights of Y chicken individuals to the cloud server (30). The cloud server (30) can learn a second algorithm model using a plurality of second area-weight information pairs received periodically.

[0125] The third algorithm model may be an algorithm model learned by the cloud server (30) based on a plurality of third area-weight information pairs corresponding to pairs of representative areas of multiple chicken individuals and preset reference weights of the chickens. For example, the third algorithm model may be a regression analysis model or an artificial neural network model.

[0126] Specifically, the representative area of ​​multiple chicken individuals can be calculated based on multiple learning area images captured by a surveillance camera (10) installed in the farm on a specific day set in advance before the weight estimation time. Since the calculation of the representative area of ​​multiple chicken individuals is similar to the calculation of the representative area of ​​the X chicken individuals described above, a redundant description is omitted.

[0127] In addition, the chicken reference weight may be the standard weight of livestock of the breeding date corresponding to the specific day among the predefined standard weights of chickens by breeding date. In Korea, the standard weight of chickens by breeding date is defined as shown in Table 1 below. For example, if a chicken population was bred from May 4th and multiple study area images were taken on May 26th, the chicken reference weight may be 971g, the standard weight of livestock for 23 days of age.

[0128]

[0129]

[0130]

[0131] That is, the cloud server (30) can receive representative areas of multiple chicken individuals based on multiple learning area images taken during a specific day from the gateway server (20), and can learn a third algorithm model using multiple pairs of third area-weight information, which are pairs of representative areas of multiple chicken individuals and standard weights of livestock of a breeding day corresponding to a specific day.

[0132] In summary, the present invention calculates the weight of a chicken by combining N different algorithm models using N combination weights. This allows for more accurate calculation of the weight of a chicken.

[0133] Specifically, conventional techniques use a single algorithm model to calculate the weight of a chicken, which can result in underestimated or overestimated weights. However, the present invention utilizes N different algorithm models trained on N different area-weight information pairs, enabling more accurate estimation of the weight of a chicken.

[0134] In particular, the weight of a chicken can be calculated using a first algorithm model trained with the area and weight of a single chicken (the first area-weight information pair), a second algorithm model trained with the representative area and representative weight of a group of chickens raised on a farm (the second area-weight information pair), and a third algorithm model trained with the representative area and chicken reference weight of a group of chickens raised on a farm (the third area-weight information pair). In this case, the first and third weight outputs by the first and third algorithm models have a tendency to overestimate the calculated weight of the chicken, and the second weight output by the second algorithm model has a tendency to underestimate the calculated weight of the chicken. Therefore, the method for estimating the weight of a chicken according to the present invention can accurately calculate the weight of a chicken by combining and using three algorithm models.

[0135] Again, referring to FIG. 4, in step (S122), the gateway server (20) can calculate or estimate the weighted sum of the preset L time weights and the L first weights as the shipping weight of the chicken population.

[0136] Specifically, in order to more accurately calculate the weight of a chicken population, the gateway server (20) may use a combination of multiple, i.e., L, weight calculation area images rather than using only one weight calculation area image at a single point in time.

[0137] At this time, L area images for weight calculation are periodically captured by the surveillance camera (10) over time. In addition, the chicken population in the farm continuously grows, and accordingly, the average area and average weight of the chicken population continuously increase. Therefore, the area of ​​a chicken in the area image captured at the first point in time may be smaller than the area of ​​a chicken in the area image captured at the second point in time after the first point in time. Therefore, in the case of the present invention, when calculating the weight of the chicken population at the time of shipment, L area images for weight calculation can be combined by applying a weight according to time.

[0138] Referring to the above, when the representative weight output of the first weight output of two or more chicken individuals included in the area image for weight calculation is the average weight output, the gateway server (20) can calculate or estimate the shipping weight of the chicken population at the shipping time based on the following mathematical expressions 2, 3, and 4.

[0139]

[0140]

[0141]

[0142]

[0143]

[0144]

[0145]

[0146] Here, W CP is the estimated shipping weight of the chicken population, μ j is the time weight of the jth weight calculation area image among L weight calculation area images, W j is the first weight of the jth weight calculation area image, W h is the first weight output of the h-th chicken among the K (integer greater than or equal to 1) chicken individuals included in the j-th weight calculation area image, α iis the i-th combination weight corresponding to the i-th algorithm model among N algorithm models, MO i Each represents the i-th weight output output when the area information of the h-th chicken individual in the j-th region image is input to the i-th algorithm model.

[0147] At this time, the j-th time weight among the L time weights can be set to be greater than the j-1-th time weight among the L time weights. That is, the first time weight is the smallest, and the last time weight is the largest. In other words, the closer the time point of capturing the area image for weight calculation is to the shipping time, the higher the time weight can be assigned. According to an embodiment, the L time weights can have a linear relationship that sequentially increases according to the time point of capture. Accordingly, the shipping weight of the chicken population can be more accurately calculated or estimated.

[0148] In short, the chicken population shipment information generation system according to one embodiment of the present invention can determine whether a chicken population has been shipped based on area images periodically captured by a surveillance camera (10). Accordingly, whether a chicken population has been shipped is automatically reported to a cloud server (30) without intervention from the farm manager, enabling more efficient shipment management.

[0149] In particular, if all of the area images used to determine the shipping time (i.e., the area images between the second area image and the plurality of first area images) are taken in a state where the lights are on and the surveillance camera (10) is not contaminated, the cloud server (30) can inform the shipping manager of the start time of the movement of the shipping truck to the slaughterhouse based on the previously acquired travel time information between the farm and the slaughterhouse and the time information taken by the shipping manager to move the chicken population in the farm to the shipping truck. Generally, since chickens are shipped at similar times from two or more farms that are close to each other, a situation in which multiple shipping trucks wait at the entrance of the slaughterhouse can be prevented.

[0150] Additionally, because chickens generally do not eat when the lights are off, images taken at this time can be excluded from the calculation of the chicken population's shipping time. This allows for accurate calculation of the chicken population's shipping weight.

[0151] In addition, a system for calculating shipment information of a chicken population according to one embodiment of the present invention uses an integrated algorithm model including N algorithm models and further uses L weight calculation area images with time weights, thereby calculating a more accurate shipment weight of a chicken population.

[0152] In addition, embodiments of the present invention may be implemented in the form of program commands that can be executed through various computer means and recorded on a computer-readable medium. The computer-readable medium may include program commands, data files, data structures, etc., alone or in combination. The program commands recorded on the medium may be those specially designed and configured for the present invention, or may be those 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, magneto-optical media such as floptical disks, and hardware devices specially configured to store and execute program commands, such as ROMs, RAMs, and flash memories. Examples of program commands include not only machine language codes generated by a compiler, but also high-level language codes that can be executed by a computer using an interpreter, etc. The hardware devices described above may be configured to operate as one or more software modules to perform operations of embodiments of the present invention, and vice versa.

[0153] Although the present invention has been described with specific details such as specific components and limited embodiments and drawings, these are provided only to help the overall understanding of the present invention, and the present invention is not limited to the above embodiments, and those with ordinary skill in the art to which the present invention pertains can make various modifications and variations based on these descriptions. Therefore, the spirit of the present invention should not be limited to the described embodiments, and all things that are equivalent or equivalent to the claims as well as the claims described below are considered to fall within the scope of the spirit of the present invention.

Claims

1. A system that produces shipment information on livestock populations raised on farms. Surveillance cameras that periodically capture images of the area of ​​the above farm; and A shipment information calculation device that calculates the shipment information based on periodic area images captured by the surveillance camera; The above shipment information production device is, For each of the above periodic zone images, the count values ​​of two or more livestock object shapes included in the zone image are calculated, If the count value of each of the plurality of first zone images among the periodic zone images is less than a preset threshold count value, the shooting time of the second zone image, which was shot a preset first number of times before from the plurality of first zone images among the periodic zone images, is calculated as the shipping time of the livestock population. A system for producing shipment information on livestock populations.

2. In paragraph 1, The second zone image is captured a first number of times before the first zone image that was captured the fastest among the plurality of first zone images. A system for producing shipment information on livestock populations.

3. In paragraph 1, The above shipment information production device determines whether the lights of the farm are turned on or off at each shooting time of the periodic area image. Among the previous zone images of the plurality of first zone images, the previous light-off zone image captured when the light was turned off is excluded from the count of the first number of times. A system for producing shipment information on livestock populations.

4. In paragraph 1, The above shipment information generation device determines whether the surveillance camera is contaminated at each shooting time of the periodic area image, Among the previous zone images of the plurality of first zone images, the previous contaminated zone image captured when the surveillance camera was contaminated is excluded from the count of the first number. A system for producing shipment information on livestock populations.

5. In paragraph 1, The above periodic area images are taken in the first period, The above threshold count value is set based on the distribution of count values ​​of a plurality of third zone images captured by the surveillance camera in a second period immediately before the first period, The shipment of the above livestock population is not carried out in the above second period, A system for producing shipment information on livestock populations.

6. In paragraph 5, The above threshold count value is set based on the average and variance of the count values ​​of the plurality of third zone images. A system for producing shipment information on livestock populations.

7. In paragraph 1, The above shipment information calculation device calculates the shipment weight of the livestock population at the shipment time based on L (an integer greater than or equal to 2) weight calculation zone images composed of the second zone image and one or more fourth zone images sequentially photographed immediately before the second zone image, a pre-learned integrated algorithm model, and L time weights corresponding to the shooting times of each of the L weight calculation zone images, The above integrated algorithm model includes N (an integer greater than or equal to 2) pre-learned algorithm models and N combination weights for combining the N algorithm models. A system for producing shipment information on livestock populations.

8. In paragraph 7, Each of the above L weight calculation area images is taken by the surveillance camera with the lights of the farm on and uncontaminated. A system for producing shipment information on livestock populations.

9. In paragraph 7, The above N algorithm models are different algorithm models that input livestock area and output livestock weight, but the area-weight information pairs used for learning each of the N algorithm models are different. The above area-weight information pair is obtained before the time of calculating the above shipment weight. A system for producing shipment information on livestock populations.

10. In paragraph 7, The above N algorithm models include at least two of the first, second and third algorithm models, The first algorithm model is learned based on a plurality of first area-weight information pairs corresponding to pairs of area and weight of a single livestock individual, The second algorithm model is learned based on a plurality of second area-weight information pairs corresponding to pairs of representative areas of X (an integer greater than or equal to 2) livestock individuals and representative weights of Y (an integer greater than or equal to 2) livestock individuals, The third algorithm model is learned based on a plurality of third area-weight information pairs corresponding to pairs of representative areas of a plurality of livestock individuals and preset livestock reference weights. A system for producing shipment information on livestock populations.

11. In paragraph 7, The above shipment information calculation device calculates L first weights corresponding to each of the L weight calculation area images based on the integrated algorithm model, and calculates the weighted sum of the L time weights and the L first weights as the shipment weight of the livestock population. For each of the L weight calculation zone images, the shipment information calculation device calculates area information of at least one livestock individual from the weight calculation zone image, inputs each area information of the at least one livestock individual into the integrated algorithm model to calculate a first weight output of the at least one livestock individual, and calculates a representative weight output of the first weight output of the at least one livestock individual as a first weight corresponding to the zone image. For each of the at least one livestock individual, the shipment information producing device inputs area information of the livestock individual into each of the N algorithm models to produce N second weight outputs which are weight outputs of each of the N algorithm models, and produces a weighted sum of the N combination weights and the N second weight outputs as the first weight output of the livestock individual. A system for producing shipment information on livestock populations.

12. A device for producing shipment information on livestock populations raised on a farm. A communication unit receiving a plurality of first zone images of the above farm captured by surveillance cameras; and A processor that calculates a count value of two or more livestock object shapes included in each of the plurality of first zone images, and, if the count value of each of the plurality of first zone images is less than a preset threshold count value, calculates the shooting time of the second zone image captured by the surveillance camera a preset first number of times from the plurality of first zone images as the shipping time of the livestock object. A device for producing shipment information on livestock populations.

13. A method for producing shipment information on livestock populations raised on a farm, A step of receiving a plurality of first zone images of the above farm captured by surveillance cameras; A step of calculating a count value of two or more livestock object shapes included in the first zone image for each of the plurality of first zone images; and A step of calculating the shooting time of the second zone image captured by the surveillance camera a preset first number of times from the plurality of first zone images as the shipping time of the livestock population, when the count value of each of the plurality of first zone images is less than a preset threshold count value; Method for calculating shipment information for livestock populations.

14. A computer-readable recording medium recording a program for performing the method of Article 13.

Images