System for calculating daily average weight of poultry by using poultry weight and image analysis

Abstract

The present invention relates to a system for calculating the daily average weight of poultry by using poultry weight and image analysis, and provides a system for calculating the daily average weight of poultry by using poultry weight and image analysis, the system comprising: a weight measurement device which is provided in a poultry house, and which has a cell plate on which poultry can be placed, thereby measuring the weight of the poultry placed on the cell plate; a camera for photographing a cell plate area so that the poultry placed on the cell plate can be photographed; and a weight calculation server for collecting and analyzing a weight value and a captured image from the weight measurement device and the camera, and calculating the daily average weight of the poultry on the basis of the analysis result, wherein the weight calculation server analyzes the captured image so as to determine the number of individual poultry placed on the cell plate, acquires the weight value measured at the photography time point of the captured image according to the determination result, and averages the acquired weight value in units of 24 hours so as to calculate the daily average weight of the poultry.

Description

System for calculating average weight of poultry by age through weight and image analysis

[0001] The present invention relates to a system for calculating the average weight of poultry by age through the analysis of the weight and images of poultry. More specifically, the invention relates to a system for calculating the average weight of poultry by age through the analysis of the weight and images of poultry, which can selectively obtain only the weight value when the number of poultry placed on the weight measuring device is 1 through the analysis of images captured by a camera, thereby accurately measuring the weight value of the poultry, and thereby accurately calculating and predicting the average weight of the poultry by age.

[0002] This research was conducted with funding from the Ministry of Agriculture, Food and Rural Affairs, the Ministry of Science and ICT, and the Rural Development Administration, and supported by the Smart Farm Multi-Ministerial Package Innovation Technology Development Project of the Korea Institute of Planning and Evaluation for Agricultural and Food Technology (IPET) and the Smart Farm Research & Development Foundation (Project ID: RS-2024-00402271).

[0003] Project ID: RS-2024-00402271

[0004] Project Number: RS-2024-00402271

[0005] Ministry Names: Ministry of Agriculture, Food and Rural Affairs, Ministry of Science and ICT, Rural Development Administration

[0006] Research Management Agencies: Korea Institute of Planning and Evaluation for Food, Agriculture and Forestry, Smart Farm Research & Development Foundation

[0007] Research Project Name: 2024 Smart Farm Multi-Ministry Package Innovation Technology Development Project

[0008] Research Project Title: Local Construction and Demonstration (Middle East, Southeast Asia) of Export-Type K-Broiler Smart Farms

[0009] Contribution rate: 1 / 1

[0010] Organizer: Emotion Co., Ltd.

[0011] Research Period: 2024.04.01 - 2025.12.31

[0012] Generally, poultry such as chickens and ducks are introduced into farm barns as chicks for egg laying or meat production, raised for a certain period, and then shipped out as egg-laying or meat poultry.

[0013] Broiler chickens, which are an example of edible poultry, may include chickens such as Samgye and native chickens. In this case, broiler chickens are raised in flat-ground chicken houses, and approximately 30,000 or more chickens are raised in the house.

[0014] In addition, it takes approximately 35 days for broiler chickens to reach adulthood from the time they are introduced, and management during the week prior to shipment is very important.

[0015] For example, in the case of broiler chickens, chickens are shipped when their individual weight is in the range of approximately 1.35 kg to 1.60 kg. At this time, the weight of a sample of approximately 3% of chickens per barn (about 300 chickens based on 10,000 chickens per barn) is measured, and the chickens are shipped only when they reach the above weight range.

[0016] Therefore, when raising broiler chickens, it is necessary to monitor for disease infection by checking weight changes by age, while also appropriately adjusting feed and water amounts according to weight changes to ensure the chickens grow healthily along the standard growth curve.

[0017] In other words, when raising chickens, measuring their weight once a day and taking appropriate measures based on weight changes is of the utmost importance for improving productivity.

[0018] Conventionally, a scale was used in which the platform on which the object to be weighed was placed had a flat surface. When weighing a chicken, a method was used in which one person held the chicken with both hands and placed it on the flat platform, while another person read and recorded the scale reading.

[0019] However, there was a problem in that it was difficult to measure the accurate weight value because the scale kept fluctuating depending on the movement of the chicken placed on the flat board.

[0020] Another method involves using a scale with a funnel-shaped container to measure the weight by placing the chicken upside down in the container, but this had the problem of increasing stress on the chicken.

[0021] In addition, when weighing chickens using these conventional scales, there was an inconvenience in that at least two workers had to work together, as the person placing the live chicken on the scale and the person reading and recording the scale reading had to work together.

[0022] In addition, in conventional chicken weighing, the operator had to manually record the readings on the scale one by one, which took too much time to measure the weight of hundreds or thousands of chickens every day. Consequently, most farms faced the problem that measuring the weight of chickens required a lot of time and labor.

[0023] Recently, various methods to improve the convenience of weighing poultry have been researched and developed; however, problems such as reduced accuracy of weight measurement still persist. Consequently, the calculated average weight by age is inaccurate, and there are issues such as the inability to accurately predict future average weights by age based on this data.

[0024] The present invention was invented to solve the problems of the prior art, and the objective of the present invention is to provide a system for calculating the average weight of poultry by age through the analysis of the weight and images of poultry, which can accurately determine the number of poultry placed on a weight measuring device through the analysis of images captured by a camera, can selectively obtain only the weight value when the number of poultry placed on the weight measuring device is 1, thereby accurately measuring the weight value of the poultry, and can improve the accuracy of the average weight of poultry by age calculated by averaging these values.

[0025] Another objective of the present invention is to provide a system for calculating the average weight of poultry by age through weight and image analysis, which can accurately predict the average weight by age in the future by analyzing the average weight by age to date using a regression analysis method, and which can expand various additional functions, such as adjusting the timing of shipment or controlling the rearing environment, through such prediction.

[0026] Another objective of the present invention is to provide a system for calculating the average weight of poultry by age through weight and image analysis, which utilizes a machine learning algorithm during the image analysis process to more accurately determine the number of poultry individuals placed on a weight measuring device, thereby further improving the accuracy of calculating and predicting the average weight by age.

[0027] The present invention provides a system for calculating the average weight of poultry by age through the analysis of the weight and images of poultry, comprising: a weight measuring device installed in a poultry house and equipped with a cell plate so that poultry can climb onto it, for measuring the weight of poultry that climbs onto the cell plate; a camera that photographs the cell plate area so as to photograph the poultry that climbs onto the cell plate; and a weight calculation server that collects and analyzes weight values ​​and captured images from the weight measuring device and the camera, and calculates the average weight of poultry by age based on the analysis results, wherein the weight calculation server analyzes the captured images to determine the number of poultry that climb onto the cell plate, obtains weight values ​​measured at the time of shooting the corresponding images according to the determination result, and calculates the average weight of poultry by age by averaging the obtained weight values ​​in 24-hour increments.

[0028] At this time, the weight calculation server may include: an information collection unit that collects weight values ​​and captured images from the weight measuring device and the camera; an image analysis unit that analyzes the captured images collected through the information collection unit using a machine learning algorithm to determine whether the number of poultry individuals placed on the cell plate is 1; a weight value acquisition unit that, if it is determined by the image analysis unit that the number of poultry individuals placed on the cell plate is 1, acquires the weight value measured at the time of shooting among the weight values ​​collected through the information collection unit; a weight value storage unit that stores the weight values ​​acquired by the weight value acquisition unit in a time series; and an average weight calculation unit that calculates the average weight of the poultry by age by averaging the weight values ​​stored in the weight value storage unit in 24-hour increments.

[0029] In addition, the camera is coupled to the top of the weight measuring device and can photograph the cell plate area in a vertical downward direction.

[0030] In addition, the camera can transmit images captured repeatedly at preset reference intervals to the information collection unit, and the weight measuring device can transmit weight values ​​measured at the same time as the camera's shooting time to the information collection unit.

[0031] In addition, the weight value acquisition unit can select and acquire the heaviest weight value among the weight values ​​measured at each shooting point when there are multiple consecutive shooting points in which the number of poultry individuals is determined to be 1.

[0032] In addition, the image analysis unit can determine the number of poultry individuals on the cell plate by learning the image of the cell plate and the image of the poultry through the machine learning algorithm and analyzing at least one of the mutual separation distance between the center of the cell plate image and the center of the poultry image and the size of the mutually overlapping area between the cell plate image and the poultry image.

[0033] In addition, the weight calculation server may further include an average weight prediction unit that predicts future average weights by age by analyzing the average weights by age up to the present, calculated through the average weight calculation unit, using a regression analysis method.

[0034] According to the present invention, a weight measuring device capable of measuring the weight of poultry is provided along with a camera capable of photographing poultry placed on the weight measuring device, and by analyzing the captured image of the camera to accurately determine the number of poultry placed on the weight measuring device and selectively obtaining only the weight value when the number of poultry placed on the weight measuring device is 1, the weight value for one poultry can be accurately measured, and the accuracy of the average weight of poultry by age calculated by averaging this is also significantly improved.

[0035] In addition, by accumulating the average weight by age to date and analyzing it using regression analysis, future average weights by age can be predicted more accurately. This prediction allows for the expansion of various additional functions, such as adjusting the timing of shipment or controlling the rearing environment.

[0036] In addition, by utilizing machine learning algorithms in the image analysis process, the number of poultry placed on the weighing device can be determined more accurately, thereby further improving the accuracy of calculating and predicting average weight by age.

[0037] FIG. 1 is a functional block diagram schematically illustrating the overall configuration of a system for calculating the average weight of poultry by age according to one embodiment of the present invention.

[0038] FIG. 2 is a functional block diagram schematically illustrating the configuration of a weight calculation server of a system for calculating the average weight of poultry by age according to one embodiment of the present invention.

[0039] FIG. 3 is a schematic diagram illustrating the configuration of a weight measuring device and a camera of a system for calculating the average weight of poultry by age according to one embodiment of the present invention.

[0040] FIG. 4 is a flowchart illustrating the method of calculating and predicting the average weight by age of poultry according to an embodiment of the present invention, according to the operation flow.

[0041] FIG. 5 is a diagram illustrating, exemplarily, the state of a captured image of a system for calculating the average weight of poultry by age according to one embodiment of the present invention.

[0042] Figure 6 is a diagram illustrating the method of analyzing captured images of a system for calculating the average weight of poultry by age according to one embodiment of the present invention.

[0043] Figure 7 is a graph showing the weight value over one day measured through the average weight calculation system for poultry by age according to one embodiment of the present invention.

[0044] FIG. 8 is a graph illustrating, exemplarily, a regression analysis process for predicting future average weight by age through a system for calculating the average weight by age of poultry according to one embodiment of the present invention.

[0045] The present invention relates to a system for calculating the average weight of poultry by age through the analysis of the weight and images of poultry, comprising: a weight measuring device installed in a poultry house and equipped with a cell plate to allow poultry to climb onto, for measuring the weight of poultry that climbs onto the cell plate; a camera that photographs the cell plate area to photograph the poultry that climbs onto the cell plate; and a weight calculation server that collects and analyzes weight values ​​and captured images from the weight measuring device and the camera, and calculates the average weight of poultry by age based on the analysis results. The weight calculation server may be configured to determine the number of poultry that climb onto the cell plate by analyzing the captured images, obtain weight values ​​measured at the time of shooting the corresponding images according to the determination result, and calculate the average weight of poultry by age by averaging the obtained weight values ​​in 24-hour increments.

[0046] Hereinafter, specific embodiments for implementing the present invention will be described in detail with reference to the drawings.

[0047] First, it should be noted that when assigning reference numerals to the components of each drawing, the same components are assigned the same reference numeral whenever possible, even if they are shown in different drawings. Furthermore, in describing the present invention, if it is determined that a detailed description of related known components or functions could obscure the essence of the invention, such detailed description is omitted.

[0048] Furthermore, when it is stated that one component is 'connected,' 'supported,' 'connected,' 'supplied,' 'transmitted,' or 'contacted' with another component, it should be understood that while the connection, support, connection, supply, transmission, or contact may be direct to that other component, there may also be other components present in between.

[0049] The terms used in this specification are used merely to describe specific embodiments and are not intended to limit the invention. Singular expressions include plural expressions unless the context clearly indicates otherwise.

[0050] Furthermore, it should be noted in advance that expressions such as "upper side," "lower side," and "side" in this specification are described based on the drawings, and may be expressed differently if the orientation of the object changes. For the same reason, some components in the attached drawings may be exaggerated, omitted, or schematically depicted, and the size of each component does not entirely reflect its actual size.

[0051] Additionally, terms including ordinal numbers, such as first, second, etc., may be used to describe various components, but such components are not limited by such terms. These terms are used solely for the purpose of distinguishing one component from another.

[0052] The meaning of "comprising" as used in the specification is to specify certain characteristics, regions, integers, steps, actions, elements, and / or components, and does not exclude the existence or addition of other specific characteristics, regions, integers, steps, actions, elements, components, and / or groups.

[0053]

[0054] FIG. 1 is a functional block diagram schematically illustrating the overall configuration of a system for calculating the average weight of poultry by age according to an embodiment of the present invention, FIG. 2 is a functional block diagram schematically illustrating the configuration of a weight calculation server of a system for calculating the average weight of poultry by age according to an embodiment of the present invention, FIG. 3 is a diagram schematically illustrating the configuration of a weight measuring device and a camera of a system for calculating the average weight of poultry by age according to an embodiment of the present invention, and FIG. 4 is a flowchart illustrating the method for calculating and predicting the average weight by age according to an embodiment of the present invention according to an operation flow.

[0055] A system for calculating the average weight of poultry by age according to one embodiment of the present invention is a system capable of calculating and predicting the average weight of poultry by age by analyzing captured images and weight values ​​obtained through camera shooting and weight measurement of poultry raised in a poultry house, and may be configured to include a weight measuring device (100), a camera (200), and a weight calculation server (300).

[0056] A weight measuring device (100) is installed inside a poultry house and is configured to measure the weight of poultry placed on the cell plate (160) by providing a cell plate (160) on which poultry can be placed. A camera (200) photographs the cell plate area to photograph the poultry placed on the cell plate (160), and a weight calculation server (300) receives weight values ​​and captured images from the weight measuring device (100) and the camera (200), analyzes them, and calculates the average weight of the poultry by age based on the analysis results.

[0057] A weight measuring device (100) may be installed in a poultry house in multiple units, and the weight of multiple poultry may be measured simultaneously through multiple weight measuring devices (100). These weight measuring devices (100) may be formed to allow poultry to climb on their own, reflecting the behavioral habits of poultry such as broiler chickens, and may be configured to measure weight through a load cell.

[0058] To look more specifically, the weight measuring device (100) may include a cell plate (160), a cell moving pipe (110), a guide holder (120), a cell control unit (130), and a cell support (140).

[0059] The cell plate (160) is a space on which poultry can be placed, and when a broiler chicken is placed on the cell plate (160) of the weight measuring device (100), its weight can be measured. The cell plate (160) may be formed in the shape of a circular plate, but this is merely an embodiment of the present invention and is not limited thereto. The cell moving pipe (110) supports the cell plate (160) and can move up and down. To this end, the lower end of the cell moving pipe (110) is connected to the cell plate (160) to fix the cell plate (160), and the upper end of the cell moving pipe (110) can be connected to the cell control unit (130) to allow for up and down movement.

[0060] Additionally, the upper part of the cell plate (160) can be inserted into the guide holder (120) and connected to the cell control unit (130), whereby the guide holder (120) acts as a guide for vertical movement.

[0061] That is, when poultry is placed on the cell plate (160), the cell moving pipe (110) moves downward. At this time, the amount by which the cell moving pipe (110) moves downward varies depending on the weight of the poultry. Also, when the poultry is on the cell plate (160) and then moves down, the cell moving pipe (110) moves upward and returns to its original position.

[0062] The guide holder (120) can guide the movement of the cell moving pipe (110) and fix the cell moving pipe (110) so that it does not shake. Additionally, the guide holder (120) has a hole formed in the center so that the cell moving pipe (110) can move up and down along the hole.

[0063] The cell control unit (130) has a guide holder (120) and a cell moving pipe (110) connected to its lower side, and is equipped with a load cell so that it can measure weight according to the downward movement of the cell moving pipe (110).

[0064] That is, the weight of the poultry can be measured by a method in which the cell moving pipe (110) moves downward according to the weight of the poultry, the load cell expands and deforms as much as it moves downward, and the deformation amount is detected as an electrical signal by a deformation measuring device and converted into a digital signal to derive the weight as a number.

[0065] Additionally, the cell control unit (130) can measure the weight of the poultry in the manner described above and transmit the measured weight of the poultry to the weight calculation server (300). Accordingly, the weight calculation server (300) can collect weight values ​​by receiving the weight of the poultry measured from the weight measuring device (100).

[0066] The cell support (140) can be connected to the cell control unit (130) and fixedly installed inside the poultry house. At this time, by supporting the cell control unit (130), the cell support (140) can also support the cell plate (160), cell moving pipe (110), and guide holder (120) that are linked together.

[0067] Additionally, the weight measuring device (100) may further include a haptic motor (150). The haptic motor (150) may be installed on the lower surface of the cell plate (160), but since this is merely an embodiment of the present invention, it may be installed at various locations to provide vibration to the cell plate (160).

[0068] That is, the haptic motor (150) can provide vibration to the cell plate (160), thereby preventing specific poultry from remaining on the cell plate (160) due to the vibration and inducing a large number of poultry to rise evenly on the cell plate (160).

[0069] This haptic motor (150) can be controlled by the cell control unit (130) to vibrate. For example, it can be controlled to vibrate after the weight of the poultry is measured. Additionally, the operation control method can be set in various ways, such as the haptic motor (150) being controlled to vibrate when the number of poultry placed on the cell plate (160) is determined to be 1 through the weight calculation server (300) as described later.

[0070] A camera (200) can be coupled to the upper part of a weight measuring device (100) to photograph the cell plate area so as to photograph poultry placed on the cell plate (160). More specifically, the camera (200) can be coupled to the upper part of the weight measuring device (100), for example, to one side of the cell control unit (130), and positioned to photograph the cell plate area in a vertical downward direction. Such a camera (200) can be coupled to each of a plurality of weight measuring devices (100) to photograph each cell plate area.

[0071] The weight calculation server (300) collects and analyzes weight values ​​and captured images from a plurality of weight measuring devices (100) and a camera (200), and calculates the average weight of poultry by age based on the analysis results. More specifically, the weight calculation server (300) analyzes the captured images transmitted from the camera (200) to determine how many individuals of poultry are placed on the cell plate (160), obtains the weight value measured at the time of taking the corresponding captured image according to the determination result, calculates the average weight of poultry by age by averaging the obtained weight values ​​in 24-hour increments, and predicts the average weight of poultry by age in the future.

[0072] Looking more closely at the configuration of the weight calculation server (300), the weight calculation server (300) includes an information collection unit (310), an image analysis unit (320), a weight value acquisition unit (330), a weight value storage unit (340), an average weight calculation unit (350), and an average weight prediction unit (360), as shown in FIG. 2.

[0073] The information collection unit (310) is connected to a plurality of weight measuring devices (100) and cameras (200) via a communication unit (not shown), and collects weight values ​​and captured images in real time from each weight measuring device (100) and camera (200) and stores them temporarily.

[0074] At this time, the camera (200) is configured to repeatedly photograph the cell plate (160) area at time intervals pre-set by the user, and the captured images taken at each interval are continuously transmitted to the information collection unit (310). The weight measuring device (100) continuously measures weight in real time, and the measured weight values ​​are continuously transmitted to the information collection unit (310) at time intervals that are the same as the time of shooting by the camera (200). The shooting interval of the camera (200) can be set to various values ​​according to the user's needs, such as, for example, 1 second or 3 seconds.

[0075] The image analysis unit (320) analyzes the captured image collected through the information collection unit (310) using a machine learning algorithm to determine how many poultry are placed on the cell plate (160). More specifically, the image analysis unit (320) determines whether the number of poultry placed on the cell plate (160) is 1.

[0076] If we examine the image analysis process of the image analysis unit (320) in more detail, first, as shown in FIG. 5, the captured image of the cell plate (160) is taken in a top-view manner according to the shooting angle of the camera (200). At the center of the captured image, there is a cell plate image (P1), which can be achieved through the initial setting process of the camera (200). When poultry is placed on the cell plate (160) or poultry is present around the cell plate (160), the captured image contains the cell plate image (P1) and the poultry image (P2).

[0077] When poultry is placed on the cell plate (160), the poultry image (P2) is superimposed on the cell plate image (P1), and the image analysis unit (320) analyzes this image superimposition state to determine how many poultry are placed on the cell plate (160).

[0078] For example, as shown in FIG. 5(a), if one poultry image (P2) is placed overlappingly within the area of ​​the cell plate image (P1), the image analysis unit (320) determines that the number of poultry individuals on the cell plate (160) is 1. As shown in FIG. 5(b), if two poultry images (P2) are placed overlappingly within the area of ​​the cell plate image (P1), the image analysis unit (320) determines that the number of poultry individuals on the cell plate (160) is 2.

[0079] Meanwhile, if the cell plate image (P1) and the poultry image (P2) among the captured images do not completely overlap and are arranged so that only some areas overlap each other, it may be ambiguous whether the poultry is placed on the cell plate (160).

[0080] In one embodiment of the present invention, the number of poultry individuals on the cell plate (160) can be determined more accurately by analyzing the mutual separation distance (d1, d2, d3) between the center (C1) of the cell plate image (P1) and the center (C2) of the poultry image (P2), or by analyzing the size of the area where the cell plate image (P1) and the poultry image (P2) overlap.

[0081] For example, as shown in FIG. 6(a), if the poultry image (P2) completely overlaps the cell plate image (P1), it is determined that the poultry has come onto the cell plate (160). As shown in FIG. 6(b) and (c), if the poultry image (P2) partially overlaps the cell plate image (P1), if the mutual overlap area is greater than or equal to a threshold value as in FIG. 6(b), it is determined that the poultry has come onto the cell plate (160) and is included in the population count; if the mutual overlap area is less than the threshold value as in FIG. 6(c), it is determined that the poultry has not come onto the cell plate (160) and is not included in the population count. At this time, the threshold value may be specified as a value pre-set by the user, for example, it may be set to half the size of the poultry image area.

[0082] Additionally, it can be determined whether poultry has come onto the cell plate (160) based on the mutual separation distance between the center (C1) of the cell plate image (P1) and the center (C2) of the poultry image (P2). As illustrated in FIG. 6 (a) and (b), if the mutual separation distance (d1, d2) between the center (C1) of the cell plate image (P1) and the center (C2) of the poultry image (P2) is less than a threshold value, it is determined that poultry has come onto the cell plate (160) and can be included in the population count. As illustrated in FIG. 6 (c), if the mutual separation distance (d3) between the center (C1) of the cell plate image (P1) and the center (C2) of the poultry image (P2) is greater than the threshold value, it is determined that poultry has not come onto the cell plate (160) and can be excluded from the population count. At this time, the threshold value can be specified as a value pre-set by the user and can be freely set according to the size of the poultry, etc.

[0083] In addition, the process of determining whether poultry has been placed on the cell plate (160) may be performed by comprehensively determining the distance between the center (C1) of the cell plate image (P1) and the center (C2) of the poultry image (P2) described above, and the size of the area where the cell plate image (P1) and the poultry image (P2) overlap each other.

[0084] This image analysis unit (320) learns the cell plate image (P1) and the poultry image (P2) using a machine learning algorithm during the image analysis process, and determines the number of poultry on the cell plate (160) based on the learning results.

[0085] The machine learning algorithm used in the image analysis unit (320) may be a CNN (Convolutional Neural Network).

[0086] The CNN algorithm is a deep learning model that is very useful for image processing, and it performs exceptionally well in tasks such as image classification, object detection, and image comparison, and can accurately determine the number of poultry on the cell plate (160).

[0087] In the CNN algorithm, first, boundary lines and planar structures are learned for a reference plane image of the cell plate and a reference plane image of the poultry, and based on the learning results, features are extracted from the cell plate image and the poultry image through various filters in the actual captured image, and the two images are compared to accurately determine the number of poultry on the cell plate (160).

[0088] When the weight value acquisition unit (330) determines by the image analysis unit (320) that the number of poultry placed on the cell plate (160) is 1, it acquires the weight value measured at the time of shooting among the weight values ​​collected through the information collection unit (310).

[0089] That is, if the number of poultry placed on the cell plate (160) in the captured image is determined to be 1 as a result of analyzing the captured image through the image analysis unit (320), the weight value measured at the time of capturing the image is obtained. The weight value obtained in this way represents the weight value of one poultry at the time of capturing.

[0090] Since the camera (200) repeatedly takes pictures and transmits the captured images at each shooting cycle (e.g., 1 second), if one poultry is on the cell plate (160) for 5 seconds, the number of poultry can be determined to be 1 in 5 consecutive captured images taken during that 5 seconds. At this time, the weight values ​​measured at the 5 shooting points may differ from each other due to reasons such as the poultry flapping its wings or moving while on the cell plate (160). In this case, if the poultry flaps its wings, the weight measured by the weight measuring device (100) will decrease relatively, so the heaviest weight value among the different weight values ​​can be considered the accurate weight value for the poultry.

[0091] Accordingly, the weight value acquisition unit (330) can select and acquire the heaviest weight value among the weight values ​​measured at each shooting point in order to prevent different weight values ​​from being acquired and to acquire a more accurate weight value when multiple shooting points are consecutive in which the number of poultry individuals is determined to be 1.

[0092] In this way, by obtaining the weight value at the time of shooting only when the number of poultry placed on the cell plate (160) through the image analysis unit (320) is 1, the accuracy of the weight measurement for one poultry is improved, and the accuracy of the daily average weight calculated by averaging this can also be improved.

[0093] If all weight values ​​measured by a weight measuring device are acquired without analyzing the captured images, even if the number of poultry placed on the device is 2 or 3, they are all recognized as the weight of a single animal, and consequently, the deviation of the measured weight values ​​becomes very large. When calculating the daily average weight by averaging weight values ​​with such large deviations, the accuracy of the calculated daily average weight is also significantly reduced. Although various correction methods can be applied to correct such weight value deviations, in one embodiment of the present invention, the deviation of weight values ​​can be minimized through the analysis of the captured images of the camera, so that an accurate daily average weight can be calculated without separate correction.

[0094] The weight value storage unit (340) stores the weight values ​​obtained by the weight value acquisition unit (330) in a time-series manner according to the time order. That is, whenever the number of poultry placed on the cell plate (160) is determined to be 1 through the image analysis unit (320), the weight value measured at the time of shooting is obtained by the weight value acquisition unit (330), and the obtained weight value is sequentially stored in the weight value storage unit (340).

[0095] The average weight calculation unit (350) calculates the daily average weight of poultry by averaging the weight values ​​stored in the weight value storage unit (340) in 24-hour increments. By repeating this process every 24 hours, the average weight for each age group can be continuously calculated.

[0096] FIG. 7 illustrates a graph of weight values ​​stored in a weight value storage unit (340) for 24 hours. As can be seen from the graph in FIG. 7, most of the weight values ​​measured over 24 hours are between 1000 and 1500. The daily average weight can be calculated by averaging these total weight values. In order to improve the accuracy of the daily average weight, the average weight can be calculated by excluding weight values ​​that fall outside the concentrated distribution range of 1000 to 1500 (or the range of 500 to 2000).

[0097] If the daily average weight calculated in this way is repeated in 24-hour intervals, the average weight by age can be continuously calculated.

[0098] The average weight prediction unit (360) can predict the future average weight by age by analyzing the average weight by age calculated so far through the average weight calculation unit (350) using a non-linear second-order regression analysis method. That is, the average weight prediction unit (360) accumulates the average weight by age calculated through the average weight calculation unit (350) and, as shown in FIG. 8, can predict the future average weight by age by completing the trend line of the second-order equation through regression analysis.

[0099] By predicting the average weight by age, it is possible to estimate the average weight of poultry at a specific age. This offers high potential for application, such as adjusting the timing of shipment or controlling the rearing environment within the poultry house.

[0100] To summarize the process of calculating and predicting the average weight of poultry by age through the configurations described above, as shown in FIG. 4, first, the cell plate area is photographed at each reference period and the weight value is measured simultaneously through the camera (200) and the weight measuring device (100) (S10). The photographed image and weight value captured at each time of shooting are transmitted to and collected by the information collection unit (310) (S20). The image analysis unit (320) analyzes the photographed image (S30) and determines whether the number of poultry placed on the cell plate (160) is 1 (S40). The image analysis unit (320) continuously analyzes the photographed image collected at each time of shooting in real time.

[0101] When it is determined that the number of poultry placed on the cell plate (160) is 1, the weight value acquisition unit (330) acquires the weight value measured at the time of taking the corresponding image (S50), and the acquired weight value is sequentially stored in the weight value storage unit (340) (S60). It is determined whether this process has been carried out for 24 hours (S70), and if 24 hours have elapsed, the average weight by age is calculated by averaging the weight values ​​stored for 24 hours through the average weight calculation unit (350) (S80). Subsequently, the average weight prediction unit (360) predicts the future average weight by age by analyzing the average weight by age up to the present using a regression analysis method (S90).

[0102] A system for calculating the average weight of poultry by age according to one embodiment of the present invention can calculate and predict a more accurate average weight by age through such a process. In particular, by selectively obtaining a weight value when the number of poultry individuals placed on the cell plate (160) is 1 through the analysis of the captured image of the camera (200), the accuracy of calculating the average weight of poultry by age can be further improved.

[0103]

[0104] The foregoing description is merely an illustrative explanation of the technical concept of the present invention, and those skilled in the art to which the present invention pertains will be able to make various modifications and variations within the scope of the essential characteristics of the present invention. Accordingly, the embodiments disclosed in the present invention are intended to explain, not limit, the technical concept of the present invention, and the scope of the technical concept of the present invention is not limited by these embodiments. The scope of protection of the present invention shall be interpreted by the claims below, and all technical concepts within an equivalent scope shall be interpreted as being included within the scope of rights of the present invention.

Claims

1. A weight measuring device installed inside a poultry house, equipped with a cell plate to allow poultry to climb onto, and measuring the weight of poultry placed on the cell plate; A camera that photographs the cell plate area to photograph poultry placed on the cell plate; and A weight calculation server that collects and analyzes weight values ​​and captured images from the weight measuring device and the camera, and calculates the average weight of poultry by age based on the analysis results. ...including, and the weight calculation server A system for calculating the average weight of poultry by age through the analysis of poultry weight and images, which analyzes the above-mentioned captured image to determine the number of poultry individuals placed on the cell plate, obtains a weight value measured at the time of capturing the image according to the determination result, and calculates the average weight of the poultry by age by averaging the obtained weight value in 24-hour increments.

2. In Paragraph 1, The above weight calculation server is An information collection unit that collects weight values ​​and captured images from the above-mentioned weight measuring device and the above-mentioned camera; An image analysis unit that analyzes the captured image collected through the information collection unit using a machine learning algorithm to determine whether the number of poultry individuals placed on the cell plate is 1; If the number of poultry placed on the cell plate is determined to be 1 by the image analysis unit, a weight value acquisition unit acquires the weight value measured at the time of shooting among the weight values ​​collected through the information collection unit; A weight value storage unit that stores weight values ​​obtained by the above weight value acquisition unit in a time-series manner; and An average weight calculation unit that calculates the average weight of poultry by age by averaging the weight values ​​stored in the above-mentioned weight value storage unit on a 24-hour basis. A system for calculating the average weight of poultry by age through weight and image analysis of poultry, including 3. In Paragraph 2, A system for calculating the average weight of poultry by age through weight and image analysis of poultry, wherein the camera is coupled to the top of the weight measuring device and photographs the cell plate area in a vertical downward direction.

4. In Paragraph 2, The above camera transmits captured images repeatedly at preset reference intervals to the information collection unit, and The above-described weight measuring device transmits the measured weight value to the information collection unit at the same time as the shooting time of the camera, and is a system for calculating the average weight of poultry by age through weight and image analysis of poultry.

5. In Paragraph 4, The above weight value acquisition unit A system for calculating the average weight of poultry by age through weight and image analysis of poultry, wherein when multiple shooting points are consecutive in which the number of poultry individuals is determined to be 1, the heaviest weight value among the weight values ​​measured at each shooting point is selected and obtained.

6. In Paragraph 2, The above image analysis unit The image of the cell plate and the image of the poultry are learned through the machine learning algorithm, and A system for calculating the average weight of poultry by age through the analysis of poultry weight and images, which determines the number of poultry individuals placed on the cell plate by analyzing at least one of the mutual separation distance between the center of the cell plate image and the center of the poultry image and the size of the mutually overlapping area between the cell plate image and the poultry image.

7. In Paragraph 2, The above weight calculation server is A system for calculating the average weight of poultry by age through weight and image analysis of poultry, further comprising an average weight prediction unit that predicts the average weight of poultry by age in the future by analyzing the average weight of poultry by age calculated to date through the above average weight calculation unit using a regression analysis method.

Images