Bird egg classification apparatus using magnetic resonance imaging

Abstract

A bird egg classification apparatus using magnetic resonance imaging according to the present invention may comprise: a memory that stores a bird egg classification artificial intelligence model that has learned to determine, on the basis of a magnetic resonance image generated by photographing a bird egg, at least one of whether the bird egg is a fertilized egg according to a first classification criterion, whether the bird egg is a blood spot egg according to a second classification criterion, whether the bird egg is a double-yolk egg according to a third classification criterion, whether the bird egg is a broken egg according to a fourth classification criterion, and which size range the bird egg falls into among a plurality of size ranges according to a fifth classification criterion; a processor that receives bird egg classification result data from the bird egg classification artificial intelligence model; and a classification unit controlled by means of the processor to classify the bird egg in response to the bird egg classification result data.

Description

Magnetic resonance imaging-based egg classification device

[0001] The present invention relates to an egg classification device using magnetic resonance imaging, and more particularly, to an egg classification device using magnetic resonance imaging, which inputs a magnetic resonance image of an egg into an egg classification artificial intelligence model, outputs classification result data indicating whether the egg is fertilized, unfertilized, blood-filled, double, or blue, and which size range it falls into among a plurality of size ranges, and classifies the egg on a conveyor belt in response to the classification result data.

[0002]

[0003] Since ancient times, farms and factories that ship eggs in bulk have spent a considerable amount of time and labor on sorting out and excluding eggs that are bad for the product, such as blue eggs with broken or cracked surfaces or blood-coated eggs, and packing only eggs that are in good condition. However, this problem is gradually being solved with the development and spread of egg sorting machines that automatically sort out bad eggs.

[0004] This egg sorting device is a sorting device that sorts eggs transported from a chicken coop, etc., by a transport device such as a roller conveyor into a predetermined number of rows, aligns them in the same direction, and then performs at least one process of crack detection or weight measurement. Depending on the type, the device may be configured in various ways, such as by further including a process of washing and drying the transported eggs.

[0005] In this regard, conventional technology suggests a method of generating multiple high-resolution images of each of a plurality of eggs continuously transported along an egg processing device using a high-resolution camera and a non-linearly shaped diffuser for uniform illumination, and classifying eggs with defects detected in the images into acceptable and those to be rejected, thereby enabling selective shipment of only eggs that are of good quality as a commodity.

[0006] Another conventional technology is to provide a defective sorting unit for sorting out defective eggs and a weight sorting unit for sorting out weight, thereby effectively removing broken eggs and quickly sorting them by weight, and additionally, to provide a fertilized egg reading unit for determining implantation, thereby enabling fertilized and unfertilized eggs to be distinguished and individually packaged.

[0007] However, existing egg sorting machines have problems with the reliability of sorting, such as not being able to detect fine cracks or blood during the process of sorting the quality of eggs, and there is a problem in that sorting itself cannot be performed when sorting is required based on the internal condition of the eggs.

[0008]

[0009] The purpose of the present invention is to provide a device for classifying eggs using magnetic resonance imaging, which inputs a magnetic resonance image of an egg into an artificial intelligence model for classifying eggs, outputs classification result data indicating whether the egg is fertilized, unfertilized, blood-filled, double, or blue, and which size range it falls into among multiple size ranges, and classifies the egg on a conveyor belt in response to the classification result data.

[0010] The objectives of the present invention are not limited to those mentioned above. 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.

[0011]

[0012] The device for classifying eggs using magnetic resonance imaging according to the present invention comprises: a memory for storing an egg classification artificial intelligence model learned to determine at least one of whether the egg is a fertilized egg according to a first classification criterion, whether the egg is a blood egg according to a second classification criterion, whether the egg is a double egg according to a third classification criterion, whether the egg is a blue egg according to a fourth classification criterion, and whether the egg falls into a size range among a plurality of size ranges according to a fifth classification criterion, based on a magnetic resonance image generated by photographing the egg; The method may include: a processor for inputting the magnetic resonance image into the egg classification artificial intelligence model, and receiving egg classification result data from the egg classification artificial intelligence model, wherein at least one of whether the egg is a fertilized egg according to the first classification criterion, whether the egg is a blood egg according to the second classification criterion, whether the egg is a double egg according to the third classification criterion, whether the egg is a blue egg according to the fourth classification criterion, and which size range the egg falls into among a plurality of size ranges according to the fifth classification criterion is determined; and a classification unit controlled by the processor to classify the egg in response to the egg classification result data.

[0013] The processor maps first plurality of training magnetic resonance images and first plurality of training egg classification result data indicating whether the eggs shown in each of the first plurality of training magnetic resonance images are fertilized eggs according to the first classification criterion to form first training data, and maps second plurality of training magnetic resonance images and second plurality of training egg classification result data indicating whether the eggs shown in each of the second plurality of training magnetic resonance images are blood eggs according to the second classification criterion to form second training data, and maps third plurality of training magnetic resonance images and third plurality of training egg classification result data indicating whether the eggs shown in each of the third plurality of training magnetic resonance images are double eggs according to the third classification criterion to form third training data, and maps fourth plurality of training magnetic resonance images and fourth plurality of training egg classification result data indicating whether the eggs shown in each of the fourth plurality of training magnetic resonance images are blue eggs according to the fourth classification criterion to form fourth training data. The fifth training data can be configured by mapping the fifth plurality of training magnetic resonance images and the fifth plurality of training magnetic resonance images, each of which indicates which size range among the plurality of size ranges according to the fifth classification criteria the eggs are included in, into the fifth training data.

[0014] The processor can train the egg classification artificial intelligence model using one or more of the first training data, the second training data, the third training data, the fourth training data, and the fifth training data.

[0015] The egg classification device using magnetic resonance imaging according to the present invention may include a conveyor unit controlled by the processor to move the egg; and a photographing unit controlled by the processor to photograph the egg moved by the conveyor unit to generate the magnetic resonance image.

[0016] The processor may set a photographing area in which each of the eggs moved by the conveyor unit is photographed by the photographing unit based on at least one of the first classification criterion, the second classification criterion, the third classification criterion, the fourth classification criterion, and the fifth classification criterion, and control the conveyor unit and the photographing unit so that the photographing area of ​​the eggs is photographed.

[0017] The above processor can preprocess the magnetic resonance image by performing one or more of cropping, rotation, color correction, shade correction, and noise removal.

[0018]

[0019] According to the present invention, by inputting a magnetic resonance image of an egg into an egg classification artificial intelligence model, classification result data is output, such as whether the egg is fertilized, unfertilized, blood-filled, double, or blue, and which size range it falls into among multiple size ranges, and by classifying the egg on a conveyor belt in response to the classification result data, the egg can be quickly and accurately classified using various classification criteria.

[0020]

[0021] Figure 1 is a perspective view of a device for classifying eggs using magnetic resonance imaging according to an embodiment of the present invention.

[0022] Figure 2 is a top view of a classification device using magnetic resonance imaging according to an embodiment of the present invention.

[0023] Figure 3 is a block diagram of a device for classifying eggs using magnetic resonance imaging according to an embodiment of the present invention.

[0024] FIG. 4 is a diagram for explaining the process of learning an artificial intelligence model for egg classification using a magnetic resonance imaging-based egg classification device according to an embodiment of the present invention.

[0025] FIG. 5 is a drawing for explaining a process of photographing an area of ​​a hen's egg by a hen's egg classification device using magnetic resonance imaging according to an embodiment of the present invention.

[0026] FIG. 6 is a diagram for explaining a process in which an egg classification device using magnetic resonance imaging according to one embodiment of the present invention outputs egg classification result data from an egg classification artificial intelligence model.

[0027]

[0028] Hereinafter, various embodiments of the present invention will be described with reference to the accompanying drawings. However, this is not intended to limit the present invention to specific embodiments, but should be understood to include various modifications, equivalents, and / or alternatives of the embodiments of the present invention. In connection with the description of the drawings, similar reference numerals may be used for similar components.

[0029] In this document, the expressions “has”, “may have”, “includes”, or “may include” indicate the presence of a feature (e.g., a number, function, operation, or component such as a part), but do not exclude the presence of additional features.

[0030] In this document, the expressions "A or B", "at least one of A and / or B", or "one or more of A and / or B" can include all possible combinations of the items listed together. For example, "A or B", "at least one of A and B", or "at least one of A or B" can all refer to cases where (1) at least one A is included, (2) at least one B is included, or (3) at least one A and at least one B are included.

[0031] The terms "first," "second," "first," or "second" used in this document can describe various components, regardless of order and / or importance, and are only used to distinguish one component from another, but do not limit the components. For example, a first user device and a second user device can represent different user devices, regardless of order or importance. For example, a first component can be referred to as a second component, and similarly, a second component can be referred to as a first component, without departing from the scope of the rights set forth in this document.

[0032] When it is said that a component (e.g., a first component) is "(operatively or communicatively) coupled with / to" or "connected to" another component (e.g., a second component), it should be understood that said component is directly coupled to / to the other component, or may be connected via another component (e.g., a third component). Conversely, when it is said that a component (e.g., a first component) is "directly coupled to" or "directly connected to" another component (e.g., a second component), it should be understood that no other component (e.g., a third component) exists between said component and said other component.

[0033] The expression "configured to" as used herein can be used interchangeably with, for example, "suitable for", "having the capacity to", "designed to", "adapted to", "made to", or "capable of". The term "configured to" does not necessarily mean something is "specifically designed to" in terms of hardware. Instead, in some contexts, the expression "a device configured to" can mean that the device, together with other devices or components, is "capable of". For example, the phrase "a control unit configured (or set) to perform A, B, and C" may mean a dedicated processor (e.g., an embedded processor) for performing those operations, or a generic-purpose processor (e.g., a CPU or application processor) that can perform those operations by executing one or more software programs stored in memory.

[0034] In particular, in this specification, “~device” may include one or more of a Central Processing Unit (CPU), an Application Processor (AP), and a Communication Processor (CP).

[0035] In this specification, the term "~device" refers to any type of hardware device including at least one processor, and may also be understood to encompass software components operating on the hardware device, depending on the embodiment. For example, the term "~device" may be understood to encompass, but is not limited to, smartphones, tablet PCs, desktops, laptops, and all user clients and applications running on each device.

[0036] The terms used in this document are used only to describe specific embodiments and may not be intended to limit the scope of other embodiments. The singular expression may include plural expressions unless the context clearly indicates otherwise. Terms used herein, including technical or scientific terms, may have the same meaning as commonly understood by a person of ordinary skill in the art described in this document. Terms defined in general dictionaries among the terms used in this document may be interpreted as having the same or similar meaning in the context of the relevant technology, and shall not be interpreted in an idealized or overly formal sense unless explicitly defined in this document. In some cases, even if a term is defined in this document, it cannot be interpreted to exclude embodiments of this document.

[0037] FIG. 1 is a perspective view of an egg classification device using magnetic resonance imaging according to an embodiment of the present invention, FIG. 2 is a top view of an egg classification device using magnetic resonance imaging according to an embodiment of the present invention, and FIG. 3 is a block diagram of a configuration of an egg classification device using magnetic resonance imaging according to an embodiment of the present invention.

[0038] Referring to FIGS. 1 to 3, a magnetic resonance imaging-based egg classification device (100) according to one embodiment of the present invention may include a memory (110), a processor (120), a conveyor unit (130), a photographing unit (140), and a classification unit (150).

[0039] The memory (110) can store an egg classification artificial intelligence model (AI) learned to determine at least one of whether the egg is a fertilized egg according to a first classification criterion, whether the egg is a blood egg according to a second classification criterion, whether the egg is a double egg according to a third classification criterion, whether the egg is a blue egg according to a fourth classification criterion, and whether the egg is included in a size range among a plurality of size ranges according to a fifth classification criterion, based on a magnetic resonance image generated by photographing the egg.

[0040] The processor (120) controls the overall operation of the magnetic resonance imaging-based egg classification device (100) according to one embodiment of the present invention.

[0041] The processor (120) is connected to the memory (110) and controls the overall operation of the magnetic resonance imaging-based egg classification device (100) according to one embodiment of the present invention.

[0042] In one embodiment, the magnetic resonance imaging-based egg classification device (100) may include at least one of a server, a smartphone, a tablet personal computer, a mobile phone, a video phone, a laptop PC, a netbook computer, a notebook computer, a personal digital assistant (PDA), a portable multimedia player (PMP), and a wearable device.

[0043] Meanwhile, the memory (110) can store various programs and data required for the operation of the magnetic resonance imaging-based egg classification device (100). The memory (110) can be implemented as a non-volatile memory, a volatile memory, a flash memory, a hard disk drive (HDD), a solid state drive (SSD), or the like.

[0044] The processor (120) can control the overall operation of the magnetic resonance imaging-based egg classification device (100) using various programs stored in the memory (110). The processor (120) can be composed of a RAM, a ROM, a graphic processing unit, a main CPU, first to n interfaces, and a bus. In this case, the RAM, ROM, graphic processing unit, main CPU, first to n interfaces, etc. can be connected to each other via a bus.

[0045] RAM stores the O / S and application programs. Specifically, when the magnetic resonance imaging-based egg classification device (100) is booted, the O / S is stored in RAM, and various application data selected by the user can be stored in RAM.

[0046] ROM stores a set of commands for system booting, etc. When a turn-on command is input and power is supplied, the main CPU copies the O / S stored in the memory (110) to RAM according to the commands stored in the ROM, executes the O / S, and boots the system. When booting is complete, the main CPU copies various application programs stored in the memory (110) to RAM, and executes the application programs copied to RAM to perform various operations.

[0047] The graphic processing unit generates a screen including various objects such as items, images, and texts using a calculation unit (not shown) and a rendering unit (not shown). Here, the calculation unit may be configured to calculate attribute values ​​such as coordinate values, shape, size, and color for each object to be displayed according to the layout of the screen using a control command received from the input unit. In addition, the rendering unit may be configured to generate screens of various layouts including objects based on the attribute values ​​calculated by the calculation unit. The screen generated by the rendering unit may be displayed within the display area of ​​the display.

[0048] The main CPU accesses the memory (110) and performs booting using the OS stored in the memory (110). In addition, the main CPU performs various operations using various programs, contents, data, etc. stored in the memory (110).

[0049] The first to nth interfaces are connected to the various components described above. One of the first to nth interfaces may be a network interface that connects to an external device via a network.

[0050] Meanwhile, furthermore, the processor (120) can control an artificial intelligence model. In this case, it goes without saying that the processor may include a graphics-only processor (e.g., GPU) for controlling the artificial intelligence model.

[0051] Meanwhile, the artificial intelligence model according to the present invention may be a supervised learning or unsupervised learning-based model. Furthermore, the artificial intelligence model according to the present invention may include a support vector machine (SVM), a decision tree, a neural network, and other methods using these methods.

[0052] In one embodiment, the AI ​​model according to the present invention may be an AI model based on a convolutional deep neural network (AI) trained by inputting training data. However, the present invention is not limited thereto, and it goes without saying that various AI models can be applied to the present invention. For example, models such as a deep neural network (DNN), a recurrent neural network (RNN), and a bidirectional recurrent deep neural network (BRDNN) can be used as the AI ​​model, but are not limited thereto.

[0053] Here, convolutional deep neural networks (CNNs) are a type of multilayer perceptron designed to utilize minimal preprocessing. A CNN consists of one or more convolutional layers and regular artificial neural network layers layered on top of them, with additional weight and pooling layers. This structure allows a CNN to fully utilize two-dimensional input data. Furthermore, a CNN can be trained using standard backpropagation. Compared to other feedforward artificial neural network techniques, a CNN is easier to train and uses fewer parameters.

[0054] Additionally, deep neural networks (DNN) are artificial neural networks (ANN) composed of multiple hidden layers between the input layer and the output layer.

[0055] At this time, the structure of a deep neural network can be composed of perceptrons. A perceptron consists of multiple inputs, a processor, and a single output. The processor multiplies each input by a weight and then sums all the weighted inputs. The processor then inputs the summed value into an activation function to produce a single output. If a specific value is desired as the output of the activation function, the weights multiplied by each input can be modified and the output can be recalculated using the modified weights. Each perceptron can use a different activation function. Furthermore, each perceptron receives the outputs from the previous layer as input and uses the activation function to derive the output. The derived output is then passed on to the input of the next layer. Through the above process, several output values ​​can be ultimately obtained.

[0056] A recurrent neural network (RNN) is a neural network in which the connections between the units forming the artificial neural network form directed cycles. Unlike feed-forward neural networks, RNNs can utilize internal memory to process arbitrary inputs.

[0057] Deep Belief Networks (DBNs) are a generative graphical model used in machine learning. In deep learning, they refer to deep neural networks comprised of multiple layers of latent variables. They are characterized by connections between layers, but no connections between units within a layer.

[0058] Deep belief neural networks, due to their generative nature, can be used for pre-training. After learning initial weights through pre-training, they can be fine-tuned using backpropagation or other discriminative algorithms. This characteristic is extremely useful when training data is limited, as the influence of initial weight values ​​on the resulting model becomes stronger with less training data. Pre-trained initial weight values ​​are closer to the optimal weights than randomly set initial weight values, which enables improved performance and speed in the fine-tuning stage.

[0059] The artificial intelligence model according to the present invention may be a language model trained to output statistically most appropriate output values ​​based on input values ​​(e.g., text, sentences).

[0060] The above-described artificial intelligence and its learning methods are provided for illustrative purposes only, and the artificial intelligence and its learning methods utilized in the above-described embodiments are not limited. For example, any type of artificial intelligence technology and its learning methods applicable to the same task by a person skilled in the art can be utilized to implement the system according to the disclosed embodiments.

[0061] Meanwhile, the processor (120) may include one or more cores (not shown) and a graphics processing unit (not shown) and / or a connection path (e.g., a bus) for transmitting and receiving signals with other components.

[0062] A processor (120) according to one embodiment performs the method described in connection with the present invention by executing one or more instructions stored in a memory (110).

[0063] For example, the processor (120) may acquire new training data by executing one or more instructions stored in the memory (110), perform a test on the acquired new training data using a learned model, extract first training data for which labeled information is acquired with an accuracy higher than a predetermined first reference value as a result of the test, delete the extracted first training data from the new training data, and retrain the learned model using the new training data from which the extracted training data has been deleted.

[0064] Meanwhile, the processor (120) may further include a RAM (Random Access Memory, not shown) and a ROM (Read-Only Memory, not shown) that temporarily and / or permanently store signals (or data) processed within the processor (120). In addition, the processor (120) may be implemented in the form of a system on chip (SoC) that includes at least one of a graphics processing unit, RAM, and ROM.

[0065] The memory (110) can store programs (one or more instructions) for processing and controlling the processor (120). The programs stored in the memory (110) can be divided into multiple modules according to function.

[0066] The conveyor section (130) has a plurality of eggs positioned at the top, and can move the plurality of eggs positioned at the top in a predetermined direction.

[0067] To this end, the conveyor unit (130) may be equipped with a conveyor motor that rotates the conveyor belt to move a plurality of eggs located on the upper portion of the conveyor belt in a predetermined direction.

[0068] The operation of the conveyor unit (130) can be controlled by the processor (120).

[0069] Meanwhile, the photographing unit (140) is formed in a tunnel shape so that the conveyor unit (130) passes through the inside, and can photograph the eggs passing through the inside to create a magnetic resonance image of the eggs.

[0070] This photographing unit (140) may be equipped with an MRI photographing module that photographs an egg to create a magnetic resonance image and a shielding housing that shields the magnetic field generated by the photographing.

[0071] Meanwhile, the photographing unit (140) can create multiple detailed magnetic resonance images for one egg by taking a cross-sectional image of one egg and creating a magnetic resonance image.

[0072] That is, the photographing unit (140) can generate a magnetic resonance image including multiple detailed magnetic resonance images for each egg.

[0073] Afterwards, the shooting unit (140) can output the generated magnetic resonance image to the memory (110) or to the processor (120).

[0074] The operation of the shooting unit (140) can be controlled by the processor (120).

[0075] The classification unit (150) can classify the corresponding eggs in response to the egg classification result data output from the egg classification artificial intelligence model (AI) by the processor (120).

[0076] Here, classifying an egg may mean that the classification unit (150) moves the physical location of the egg to a location corresponding to the egg classification result data.

[0077] This classification unit (150) may be equipped with a support unit, a robot arm, and a grip unit to move the position of the eggs by gripping the eggs.

[0078] The support member is fixed to the ground adjacent to the end of the conveyor section of the conveyor section at the bottom, and one end of the robot arm is connected to the top to support the moving robot arm.

[0079] Additionally, the gripping portion may be formed in a forceps shape to pressurize both sides of the egg to grip the egg.

[0080] This classification unit (150) may be configured to implement mechanical movements. Specifically, the classification unit (150) may include one or more joints (or joints), links, gears, etc. capable of performing various tasks. This classification unit (150) may include a plurality of joints, a plurality of links interconnected by the plurality of joints, and a drive motor that rotates the plurality of joints. In this case, the processor (120) may control the classification unit (150) by controlling the operation of the drive motor.

[0081]

[0082] Below, we will explain in detail the process of training the artificial intelligence (AI) model for classification of orchids.

[0083] FIG. 4 is a diagram for explaining the process of learning an artificial intelligence model for egg classification using a magnetic resonance imaging-based egg classification device according to an embodiment of the present invention.

[0084] Referring further to FIG. 4, the processor (120) may map the first plurality of training magnetic resonance images (TI) and the first plurality of training egg classification result data (TC) indicating whether the eggs shown in each of the first plurality of training magnetic resonance images (TI) are fertilized eggs according to the first classification criterion, thereby configuring the first training data.

[0085] Here, a fertilized egg is an egg that has been fertilized by the meeting of an egg and sperm in the female of an oviparous animal, and can hatch into a young animal, while an unfertilized egg is an egg that has not been fertilized.

[0086] In addition, the processor (120) may map a second plurality of training magnetic resonance images (TI) and a second plurality of training egg classification result data (TC) indicating whether the eggs indicated by each of the second plurality of training magnetic resonance images (TI) are blood eggs according to the second classification criteria, to form second training data.

[0087] Here, a blood egg is an egg that contains blood inside.

[0088] In addition, the processor (120) can map the third plurality of training magnetic resonance images (TI) and the third plurality of training egg classification result data (TC) indicating whether the eggs shown in each of the third plurality of training magnetic resonance images (TI) are twin eggs according to the third classification criterion, thereby forming third training data.

[0089] Here, a double egg is an egg with multiple yolks inside.

[0090] In addition, the processor (120) may map the fourth plurality of training magnetic resonance images (TI) and the fourth plurality of training magnetic resonance images (TI) and the fourth plurality of training magnetic resonance images (TI) indicating whether the eggs are blue according to the fourth classification criterion, respectively, to form the fourth training data.

[0091] Here, blue is an egg with a broken or cracked shell.

[0092] In addition, the processor (120) may map the fifth plurality of training magnetic resonance images (TI) and the fifth plurality of training magnetic resonance images (TI) and the fifth plurality of training magnetic resonance images (TI) to determine which size range among the plurality of size ranges according to the fifth classification criteria the eggs indicated by each of the fifth plurality of training magnetic resonance images (TI) are included in, thereby forming the fifth training data.

[0093] Thereafter, the processor (120) can train an artificial intelligence model (AI) for classification of eggs using one or more of the first training data, the second training data, the third training data, the fourth training data, and the fifth training data.

[0094] In particular, when the processor (120) trains an egg classification artificial intelligence model (AI) using the first training data, the processor (120) can train the egg classification artificial intelligence model (AI) by linking the connection shape between the embryo and the latebra of the egg on the first plurality of training magnetic resonance images (TI) with the first plurality of training egg classification result data (TC) indicating whether the egg is fertilized.

[0095] Meanwhile, the processor (120) may preprocess one or more of the first plurality of training magnetic resonance images (TI), the second plurality of training magnetic resonance images (TI), the third plurality of training magnetic resonance images (TI), the fourth plurality of training magnetic resonance images (TI), and the fifth plurality of training magnetic resonance images (TI) to train an egg classification artificial intelligence model (AI) using one or more of the first training data, the second training data, the third training data, the fourth training data, and the fifth training data.

[0096] Specifically, the processor (120) may perform preprocessing by performing one or more of cropping, rotation, color correction, shade correction, and noise removal on one or more of the first plurality of training magnetic resonance images (TI), the second plurality of training magnetic resonance images (TI), the third plurality of training magnetic resonance images (TI), the fourth plurality of training magnetic resonance images (TI), and the fifth plurality of training magnetic resonance images (TI).

[0097] Meanwhile, the above-described classification artificial intelligence model (AI) may not be limited in type as long as it is an AI model that can input an image and classify what kind of object the image represents according to various classification criteria.

[0098] For example, the egg classification artificial intelligence (AI) model according to the present invention may be an AI model based on convolutional deep neural networks (AI). However, this is not a limitation, and various AI models can be applied to the present invention. For example, models such as a deep neural network (DNN), a recurrent neural network (RNN), and a bidirectional recurrent deep neural network (BRDNN) can be used as AI models, but are not limited thereto.

[0099]

[0100] Below, the process of filming the eggs by the filming unit (140) will be explained.

[0101] FIG. 5 is a drawing for explaining a process of photographing an area of ​​a hen's egg by a hen's egg classification device using magnetic resonance imaging according to an embodiment of the present invention.

[0102] Referring further to FIG. 5, when an egg that has not been classified by the egg classification device (100) using magnetic resonance imaging according to one embodiment of the present invention is positioned at the upper end of the conveyor unit (130), it can be moved in a predetermined direction by the conveyor unit (130) and then enter the interior of the photographing unit (140).

[0103] Afterwards, the camera unit (140) can capture the eggs that have entered the interior and create a magnetic resonance image of the eggs.

[0104] At this time, the processor (120) can set a shooting area in which the eggs are shot according to the set classification criteria when classifying the eggs.

[0105] For example, the processor (120) can set the central region (R1) of the egg as the shooting region or set the entire region (R2) of the egg as the shooting region.

[0106] More specifically, when classifying an egg, the processor (120) can set the central region (R1) of the egg as the shooting region if at least one of the first classification criterion, whether it is a fertilized egg, and the third classification criterion, whether it is a double egg, is set as the classification criterion.

[0107] In addition, when the processor (120) classifies the eggs, if at least one of the second classification criterion of whether the eggs are blood eggs, the fourth classification criterion of whether the eggs are blue eggs, and the fifth classification criterion of whether the eggs fall into a size range among multiple size ranges is set as a classification criterion, the entire area (R2) of the eggs can be set as a shooting area.

[0108] Thereafter, the processor (120) can control one or more of the shooting time of the shooting unit (140), the shooting interval of the shooting unit (140), and the moving speed of the egg by the conveyor unit (130) so that only the set shooting area is captured by the shooting unit (140).

[0109] In addition, the processor (120) can control the conveyor unit (130) by considering the time required for each process in which the eggs are moved by the conveyor unit (130), photographed by the photographing unit (140), and physically moved and classified by the classification unit (150), thereby controlling the speed of movement of the eggs by the conveyor unit (130).

[0110] Here, the central region (R1) may be the region from the first point, which is 20% of the egg height from the lowest point of the egg, to the second point, which is 80% of the egg height from the lowest point of the egg, when the egg is erected.

[0111] For example, the central area of ​​an egg with a height of 10 cm may be the area from a point 2 cm high from the lowest point of the egg to a point 8 cm high from the lowest point of the egg when the egg is standing.

[0112]

[0113] Below, we will explain in detail the process of outputting the egg classification result data.

[0114] FIG. 6 is a diagram for explaining a process in which an egg classification device using magnetic resonance imaging according to one embodiment of the present invention outputs egg classification result data from an egg classification artificial intelligence model.

[0115] Referring further to FIG. 6, the processor (120) inputs the magnetic resonance image (I) into the egg classification artificial intelligence model (AI) and can output egg classification result data (C) from the egg classification artificial intelligence model (AI) in which at least one of whether the egg is a fertilized egg according to the first classification criterion, whether the egg is a blood egg according to the second classification criterion, whether the egg is a double egg according to the third classification criterion, whether the egg is a blue egg according to the fourth classification criterion, and which size range the egg falls into among a plurality of size ranges according to the fifth classification criterion is determined.

[0116] At this time, the processor (120) can output only the egg classification result data (C) including a specific classification result from the egg classification artificial intelligence model (AI) according to the set classification criteria when classifying the eggs.

[0117] Specifically, the processor (120) can set one or more of the first to fifth classification criteria described above in an egg classification artificial intelligence model (AI) for classifying eggs, input a magnetic resonance image (I) into the egg classification artificial intelligence model (AI) for which one or more classification criteria are set, and receive an output including a classification result according to one or more of the set classification criteria.

[0118] Meanwhile, when there are multiple magnetic resonance images (I) generated by photographing one egg, the processor (120) sequentially inputs the multiple magnetic resonance images (I) into the egg classification artificial intelligence model (AI), and outputs multiple egg classification result data (C) corresponding to each of the multiple magnetic resonance images (I) from the egg classification artificial intelligence model (AI).

[0119] At this time, the processor (120) can select the egg as a fertilized egg if at least one of the plurality of egg classification result data (C) indicates that the egg is a fertilized egg according to the first classification criterion.

[0120] At this time, the processor (120) can select the egg as a blood egg if at least one of the egg classification result data (C) among the plurality of egg classification result data (C) indicates that the egg is a blood egg according to the second classification criterion.

[0121] At this time, the processor (120) can select the egg as a pair of eggs if at least one of the egg classification result data (C) among the plurality of egg classification result data (C) indicates that the egg is a pair of eggs according to the third classification criterion.

[0122] At this time, the processor (120) can select the egg as blue if at least one of the plurality of egg classification result data (C) indicates that the egg is blue according to the fourth classification criterion.

[0123]

[0124] Meanwhile, the processor (120) according to another embodiment may control the conveyor unit (130) by considering the time required for each process in which eggs are moved by the conveyor unit (130), photographed by the photographing unit (140), and physically moved and classified by the classification unit (150), as described above, thereby controlling the speed of egg movement by the conveyor unit (130).

[0125] At this time, the processor (120) according to another embodiment can calculate the egg classification speed of the egg classification device (100) using magnetic resonance imaging according to the present invention based on the egg movement speed and the egg spacing between each of the plurality of eggs moved in one row by the conveyor unit (130).

[0126] Specifically, a processor (120) according to another embodiment can divide the egg movement speed by the egg spacing to calculate the egg classification speed.

[0127] For example, if the egg moving speed is 20 cm / sec and the egg spacing is 10 cm, a processor according to another embodiment can calculate the egg classification speed as 2 / sec, which is the egg moving speed of 20 cm / sec divided by the egg spacing of 10 cm.

[0128] Through this, the processor (120) can calculate the egg classification speed, which is one of the performances of the egg classification device (100) using magnetic resonance imaging according to the present invention.

[0129] Here, the spacing between the eggs can be determined in advance according to the structure and shape of the conveyor section (130).

[0130] The steps of a method or algorithm described in connection with an embodiment of the present invention may be implemented directly in hardware, implemented as a software module executed by hardware, or implemented by a combination thereof. The software module may reside in a random access memory (RAM), a read only memory (ROM), an erasable programmable ROM (EPROM), an electrically erasable programmable ROM (EEPROM), a flash memory, a hard disk, a removable disk, a CD-ROM, or any other form of computer-readable recording medium well known in the art to which the present invention pertains.

[0131] Additionally, different embodiments of the present invention may be complementary or combined with each other.

[0132] The components of the present invention may be implemented as a program (or application) and stored on a medium to be executed in conjunction with a computer, which is hardware. The components of the present invention may be implemented as software programs or software elements. Similarly, the embodiments may be implemented in a programming or scripting language such as C, C++, Java, assembler, Python, etc., including various algorithms implemented as a combination of data structures, processes, routines, or other programming components. Functional aspects may be implemented as algorithms executed on one or more processors.

[0133] While the embodiments of the present invention have been described above with reference to the attached drawings, those skilled in the art will appreciate that the present invention can be implemented in other specific forms without altering the technical concept or essential features thereof. Therefore, the embodiments described above should be understood to be illustrative in all respects and not restrictive.

Claims

1. A memory storing an egg classification artificial intelligence model learned to determine at least one of the following based on a magnetic resonance image generated by photographing an egg: whether the egg is a fertilized egg according to a first classification criterion; whether the egg is a blood egg according to a second classification criterion; whether the egg is a double egg according to a third classification criterion; whether the egg is a blue egg according to a fourth classification criterion; and whether the egg falls into a size range among multiple size ranges according to a fifth classification criterion; A processor that inputs the magnetic resonance image into the egg classification artificial intelligence model and outputs egg classification result data from the egg classification artificial intelligence model, wherein at least one of whether the egg is a fertilized egg according to the first classification criterion, whether the egg is a blood egg according to the second classification criterion, whether the egg is a double egg according to the third classification criterion, whether the egg is a blue egg according to the fourth classification criterion, and which size range the egg falls into among a plurality of size ranges according to the fifth classification criterion is determined; and A classification unit that is controlled by the processor and classifies the eggs in response to the egg classification result data; characterized in that it includes A device for classifying eggs using magnetic resonance imaging.

2. In paragraph 1, The above processor The first plurality of training magnetic resonance images and the first plurality of training egg classification result data, which are data indicating whether the eggs shown in each of the first plurality of training magnetic resonance images are fertilized eggs according to the first classification criterion, are mapped to form the first training data, The second plurality of training magnetic resonance images and the second plurality of training egg classification result data, which are data indicating whether the eggs shown in each of the second plurality of training magnetic resonance images are blood eggs according to the second classification criteria, are mapped to form second training data, The third plurality of training magnetic resonance images and the third plurality of training egg classification result data, which are data indicating whether the eggs shown in each of the third plurality of training magnetic resonance images are double eggs according to the third classification criteria, are mapped to form the third training data. The fourth plurality of training magnetic resonance images and the fourth plurality of training magnetic resonance images are mapped to form the fourth training data by mapping the fourth plurality of training magnetic resonance images to determine whether the eggs are blue according to the fourth classification criteria. The fifth plurality of training magnetic resonance images and the fifth plurality of training magnetic resonance images are each mapped to form fifth training data, which is characterized by mapping the fifth plurality of training magnetic resonance images and the fifth plurality of training magnetic resonance images to determine which size range among the plurality of size ranges according to the fifth classification criteria the eggs are included in. A device for classifying eggs using magnetic resonance imaging.

3. In paragraph 2, The above processor Characterized in that the egg classification artificial intelligence model is trained using at least one of the first training data, the second training data, the third training data, the fourth training data, and the fifth training data. A device for classifying eggs using magnetic resonance imaging.

4. In paragraph 1, A conveyor unit controlled by the processor to move the eggs; and A photographing unit that is controlled by the processor and moves by the conveyor unit to photograph the egg to generate the magnetic resonance image; characterized in that it includes; A device for classifying eggs using magnetic resonance imaging.

5. In paragraph 4, The above processor A method characterized in that, based on at least one of the first classification criterion, the second classification criterion, the third classification criterion, the fourth classification criterion and the fifth classification criterion, a photographing area is set in which each of the eggs moved by the conveyor unit is photographed by the photographing unit, and the conveyor unit and the photographing unit are controlled so that the photographing area of ​​the eggs is photographed. A device for classifying eggs using magnetic resonance imaging.

6. In paragraph 1, The above processor Preprocessing is characterized by performing one or more of cropping, rotation, color correction, shade correction, and noise removal on the magnetic resonance image. A device for classifying eggs using magnetic resonance imaging.

Images