Pig breeding support device, pig breeding support method, and pig breeding support program

Abstract

This swine rearing assistance apparatus is provided with: an acquisition unit that acquires image data of an image captured by a camera unit disposed so as to be directed towards a pen in which female swine are being raised in a group; a detection unit that detects a mounting behavior of the female swine on the basis of the image of the image data; and a count unit that counts the number of times the mounting behavior has been detected during a preset observation time. With such a swine rearing assistance apparatus, it is possible to notify, in a timely fashion without requiring excessive cost or labor, rearing staff of whether or not any of the female swine being reared in the pen are showing signs of heat.

Description

[Technical field]

[0001] The present invention relates to a pig breeding support device, a pig breeding support method, and a pig breeding support program. [Background technology]

[0002] 2. Description of the Related Art In order to recognize estrus symptoms in livestock, a device that detects riding behavior of the livestock is known (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2016-220588 A Summary of the Invention [Problem to be solved by the invention]

[0004] In pig farming, it is also important to determine when sows are in heat, from the perspective of increasing the number of offspring born per year and maintaining the breeding cycle at an appropriate interval. There are two methods for raising sows: stall rearing and group rearing. In recent years, from the perspective of animal welfare, there has been a trend toward group rearing in a relatively spacious area generally called a pen. In group rearing, sensors can be attached to each pig to monitor their mounting behavior, but there is a risk that the attached sensors will fall off due to contact between the sows housed inside, causing injury or accidental ingestion. It is also not realistic for breeders to continue to monitor the mounting behavior of many sows.

[0005] The present invention has been made to solve such problems, and provides a pig breeding support device etc. that can notify breeders in a timely manner, without excessive cost or labor, whether or not any of the sows being raised in a pen are showing signs of estrus. [Means for solving the problem]

[0006] The pig breeding support device in a first aspect of the present invention comprises an acquisition unit that acquires image data of images taken by a camera unit installed facing a pen in which sows are bred in groups, a detection unit that detects the mounting behavior of the sows based on the image of the image data, and a counting unit that counts the number of mounting behaviors detected during a set observation time.

[0007] In addition, a pig breeding support method in a second aspect of the present invention includes an acquisition step of acquiring image data of an image captured by a camera unit installed facing a pen in which sows are bred in groups, a detection step of detecting mounting behavior of sows based on the image of the image data, and a counting step of counting the number of mounting behaviors detected during a set observation time.

[0008] In addition, a pig breeding support program in a third aspect of the present invention causes a computer to execute an acquisition step of acquiring image data of an image taken by a camera unit installed facing a pen in which sows are bred in groups, a detection step of detecting mounting behavior of sows based on the image of the image data, and a counting step of counting the number of mounting behaviors detected during a set observation time. Effect of the Invention

[0009] The present invention makes it possible to provide a pig breeding support device etc. that can inform breeders in a timely manner, without excessive cost or effort, whether or not any of the sows being raised in a pen are showing signs of estrus. [Brief description of the drawings]

[0010] [Figure 1] FIG. 1 is a diagram showing an overall picture of a pig farming environment in which the pig farming support device according to this embodiment is adopted. [Diagram 2] FIG. 1 is a diagram illustrating a riding action. [Diagram 3] 13 is a diagram illustrating the procedure of a process for detecting riding behavior using a learning model. FIG. [Figure 4]FIG. 2 is a diagram showing the hardware configuration of the pig breeding assistance device and peripheral devices. [Diagram 5] FIG. 13 is a diagram illustrating a reference observation time. [Figure 6] FIG. 13 is a diagram illustrating a count list. [Figure 7] FIG. 13 is a diagram showing an example of the display on a zookeeper terminal that has received an excess notification. [Figure 8] FIG. 4 is a flowchart illustrating a processing procedure of a calculation unit. [Figure 9] FIG. 13 is a diagram showing an overall picture of a pig farming environment in which a pig farming support device according to another embodiment is adopted. [Figure 10] FIG. 13 is a diagram illustrating a counting list of a pig raising support device according to yet another embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] The present invention will be described below through embodiments of the invention, but the invention according to the claims is not limited to the following embodiments. Furthermore, not all of the configurations described in the embodiments are necessarily essential as means for solving the problems.

[0012] 1 is a diagram showing an overall picture of a pig farming environment in which a pig breeding support device according to this embodiment is adopted. The pig farm has a plurality of pens 301 separated by walls and fences. Each pen 301 houses a plurality (for example, about 10) sow pigs 302, which are bred collectively. The number of sow pigs 302 bred in each pen 301 can be adjusted according to the breed of sow pigs 302, the breeding environment, etc.

[0013] A camera unit 210 for observing the housed sows 302 is installed in each pen 301. The camera unit 210 is installed, for example, suspended from the ceiling, facing the pen 301 to be observed so that the entire pen 301 can be imaged from above. The camera unit 210 converts the captured image into image data and transmits it to the server 100 via the network 200. Specifically, a wireless unit 230 installed in the facility is connected to the network 200, and the camera unit 210 can transmit the image data to the server 100 by establishing wireless communication with the wireless unit 230. The network 200 connecting the camera unit 210 and the server 100 may be the Internet or an intranet, or short-distance wireless communication may be adopted when the management facility in which the server 100 is installed is installed in a pig farm.

[0014] A keeper who takes care of the sow 302 may carry a keeper terminal 220. The keeper terminal 220 is, for example, a tablet terminal or a smartphone, and can send and receive various information to and from the server 100 via the wireless unit 230 and the network 200. The keeper can, for example, input breeding records into the keeper terminal 220 and transfer them to the server 100, or can call up breeding records stored in the server 100. Furthermore, when the keeper terminal 220 receives an excess notification, which will be described later, it displays that fact.

[0015] A server 100 is installed in the management facility as a pig breeding support device. The server 100 is connected to a network 200. The server 100 sequentially acquires image data from a camera unit 210 installed in each pen 301, detects mounting behavior of sows 302 based on the image data, and counts the number of times the behavior occurs. The server 100 can display the results on a display monitor 150 in response to a request from a manager or a zookeeper. The display monitor 150 is, for example, a monitor equipped with a liquid crystal panel. The server 100 is also connected to an input device 160 that accepts operations from the manager or the zookeeper. The input device 160 is composed of a keyboard, a mouse, a touch panel superimposed on the display surface of the display monitor 150, and the like.

[0016] In pig farming, it is important to determine whether the sows 302 are in heat in order to increase the number of pigs born per year and to maintain the breeding cycle at an appropriate interval. However, in a pen where pigs are raised in groups, it is a task that requires a great deal of effort and skill for the keeper to constantly observe a large number of sows and determine whether they are in heat. The pig breeding support device in this embodiment detects the mounting behavior exhibited by the sows 302 in the pen 301 and notifies the keeper that at least one of the sows 302 in the pen 301 shows signs of heat, thereby supporting the breeding work.

[0017] In the case where there are multiple pens 301 in a pig farm as shown in Fig. 1, the keeper can reduce a lot of work just by being informed by the pig breeding support device of which pen 301 contains sows 302 showing signs of estrus. In other words, the keeper does not need to extract sows 302 showing signs of estrus until notified by the pig breeding support device, and even if notified, he or she can only extract the sows 302 in the specified pen 301. If the presence of sows 302 showing signs of estrus is indicated for a limited number of pigs in one pen 301, it is relatively easy to extract them even for an inexperienced keeper.

[0018] The pig breeding support device in this embodiment detects the mounting behavior of the sow 302 by using images captured by the camera unit 210. Mounting behavior is known as a behavior of the sow 302 that indicates a sign of estrus. FIG. 2 is a diagram explaining mounting behavior, showing a view of the sow 302 in the pen as viewed from above. FIG. 2(A) shows two examples in which a sow 302b is mounting on the back of a sow 302a. In this way, the mounting behavior of the sow 302b mounting on another pig is known as a behavior that indicates a sign of estrus.

[0019] It is known that signs of estrus can also be predicted from the mounting behavior of cows, but because cows are kept in small numbers and each one is highly valuable, a method is adopted in which an infrared sensor or acceleration sensor is attached to each cow to detect the mounting behavior. However, in group breeding where a relatively large number of pigs are housed in a pen with limited space, it is not realistic to attach a sensor to each sow.

[0020] In view of this background, the pig breeding support device according to the present embodiment detects the riding behavior of sows 302 bred in a group based on an image captured by a camera unit 210 installed facing the pen 301. From an overhead image of the pen 301, it is possible to detect whether the sows 302 are in a riding state based on the degree of overlap between them. For example, the contour line of each sow is extracted, and if the areas surrounded by the contour lines overlap, it is determined that the sows are in a riding state. In this case, the proportion or part of the overlapping area, the shape of the contour line for each sow 302, etc. may be taken into consideration. For example, if the proportion of the overlapping area is 30% or more of the size of the upper sow pig, it is determined that the sows are in a riding state. In addition, if the heads and torsos overlap, it is determined that the sows are in a riding state, but if the heads overlap, it is not determined that the sows are in a riding state. In addition, if the contour line shape indicates that the upper sow pig is lying down, it is not determined that the sows are in a riding state. By analyzing the overlapping areas of pigs that appear in the images in this way, it is possible to detect whether or not pigs are mounting each other.

[0021] 2(B) shows two examples of sow pigs 302a and 302b standing next to each other, which are not recognized as a riding state. As shown in the figure, it is common to see a part of the head of sow pig 302b overlapping with the torso of sow pig 302a, or sow pigs 302a and 302b standing close to each other facing the same direction, but these are not recognized as a riding state.

[0022] In this way, the method of detecting the riding behavior from an overhead image of the sows 302 kept in groups eliminates the risk of false detection, such as when two pigs equipped with sensors for detecting riding are adjacent to each other. Furthermore, unlike livestock such as cows that are kept in groups in large outdoor pastures, it is relatively easy to install an overhead camera unit 210 for sows 302 kept in groups in indoor pens, so a detection method using images is suitable.

[0023] The riding state can be analytically determined by extracting contour lines from the acquired image as described above, but in this embodiment, the riding behavior is detected by inputting the acquired image into a learning model that has been trained using a teacher image including an area where pigs overlap. FIG. 3 is a diagram explaining the procedure of the detection process of the riding behavior using the detection neural network 121, which is a learning model. The detection neural network 121 is created in advance by supervised learning, in which a large amount of teacher data is given to an overhead image of a pen in which sows are kept in a group, and the number of detections, which is the number of sows that should be determined to be in a riding state in the image, is linked as a correct answer. The number of detections in the teacher data is determined, for example, by an operator extracting an area where pigs overlap in the overhead image to determine whether or not the pigs are in a riding state, and counting how many sows are determined to be in a riding state in the overhead image. The operator or the like determines whether or not the pigs are in a riding state based on the proportion and part of the area where the pigs overlap, and the shape and direction of the extracted sow's contour.

[0024] The detection neural network 121, which is a learning model generated in this way, is incorporated into the server 100, which is a pig breeding support device, and is used. Specifically, for example, if eight pens 201 to be observed are provided in a pig farm, image data is sequentially sent to the server 100 from eight camera units 210 that overlook each of the pens 201. Images img1 to img8 of the image data are sequentially input to the detection neural network 121. Each time an image is input, the detection neural network 121 outputs the number of sows that are determined to be in a riding state in the image with the highest probability as the detection number. In this way, the detection number for each pen 201 is obtained, for example, the detection number for the image img1 of the first pen is "1", the detection number for the image img2 of the second pen is "2", the detection number for the image img3 of the third pen is "0", ..., the detection number for the image img8 of the eighth pen is "1". The server 100 repeats this process for a preset observation time, and counts and integrates the number of detections for each pen 201.

[0025] 4 is a diagram showing the hardware configuration of the server 100 and peripheral devices as a pig breeding support device. The server 100 can be connected to the display monitor 150, the input device 160, the camera unit 210, and the keeper terminal 220, as described above.

[0026] The server 100 is mainly composed of a calculation unit 110, a storage unit 120, and a communication unit 130. The calculation unit 110 is a processor (CPU: Central Processing Unit) that controls the server 100 and executes programs. The processor may be configured to cooperate with a calculation processing chip such as an ASIC (Application Specific Integrated Circuit) or a GPU (Graphics Processing Unit). The calculation unit 110 reads out a pig breeding support program stored in the storage unit 120 and executes various processes related to pig breeding support.

[0027] The storage unit 120 is a non-volatile storage medium, and is configured, for example, by an HDD (Hard Disk Drive). The storage unit 120 may store various parameter values, functions, display element data, lookup tables, and the like used for control and calculation, in addition to programs for executing control and processing of the server 100. The storage unit 120 particularly stores a detection neural network 121 and a counting list 122. As described above, when an image captured by the camera unit 210 is input, the detection neural network 121 outputs a detection number indicating the number of horses in a riding state present in the image. The counting list 122 is a record related to the counting of riding behavior, and will be described in detail later. The storage unit 120 may be configured by multiple hardware components, and for example, a storage medium for storing a program and a storage medium for storing the detection neural network 121 may be configured by separate hardware components.

[0028] The communication unit 130 includes, for example, a LAN unit, and transmits an imaging control signal generated by the calculation unit 110 to the camera unit 210, and passes image data sent from the camera unit 210 to the calculation unit 110, via the network 200. It also relays data transmission between the keeper terminal 220 and the calculation unit 110. The communication unit 130 can also relay data and control signals to and from other external devices. For example, it can be used to import update data for the pig breeding support program and the detection neural network 121 from an external server.

[0029] The calculation unit 110 also serves as a functional calculation unit that executes various calculations in response to processing instructed by the pig breeding support program. The calculation unit 110 can function as an acquisition unit 111, a detection unit 112, and a counting unit 113. The acquisition unit 111 mainly acquires image data of images captured by the camera unit 210 and passes it on to the detection unit 112. The detection unit 112 mainly detects the mounting behavior of the sow 302 based on the image of the image data received from the acquisition unit 111, and passes the result on to the counting unit 113. The counting unit 113 mainly counts the number of mounting behaviors detected by the detection unit 112 during a set observation time.

[0030] Next, the process of the counting unit 113 will be described. Fig. 5 is a diagram for explaining a reference observation time as a target period for counting the number of times of riding behavior by the counting unit 113. The counting unit 113 resets the counted number for each preset reference observation time.

[0031] In this embodiment, the reference observation time is set to 24 hours. The counting unit 113 counts how many times the climbing behavior is detected in each pen 201 during this reference observation time. Specifically, as described with reference to FIG. 3, the acquisition unit 111 periodically acquires the images img1 to img8 of the image data of each pen 201, and the detection unit 112 detects the number of pigs (detection number) performing the climbing behavior in each image using the detection neural network 121. For example, if the detection number for the image img2 of the second pen is "2", it means that at the time of this detection, two sows 302 are each climbing over another sow 302, and two climbing behaviors have been observed. This is repeated during the reference observation time, and the number of times the climbing behavior is detected in each pen 201 is counted.

[0032] However, even if a period is included in the standard observation time, periods that correspond to certain circumstances are excluded from the counting. Here, the zookeeper approach period and the boar introduction period are set as such excluded periods. The zookeeper approach period is set as a fixed period including the scheduled time from when the zookeeper approaches the pen 301 to when he leaves. Also, the boar introduction period is set as a fixed period including the scheduled time from when the boar is introduced into the pen 301 to when it is removed.

[0033] The staff approach period is set to, for example, 8:00 to 9:00 when the staff enters the pen 301 to feed the animals. In addition, the staff approach period may also be set to include the period when the staff approaches the pen 301 to clean the pen 301 or to inspect the sows 302. The boar introduction period is set to 14:00 to 14:30 when the boar is temporarily brought into the pen 301 to promote the estrus of the sows 302. In this way, when staff or boars approach or enter the pen 301, the sows 302 become excited to a certain extent, and even sows 302 that do not show signs of estrus may behave irregularly, including mounting. Therefore, for events that may cause irregular behavior, a certain period including the implementation period of the event may be set as an exclusion period that excludes the sows 302 from the counting. Specifically, the exclusion period may be set by adding a period before and after the implementation period of the event during which the sows 302 sense the event and become excited, and a period during which the sows regain their composure after the event.

[0034] The exclusion period can be set in advance by an administrator or the like by operating the input device 160. The exclusion period does not have to be included in the reference observation time for each time, and for example, if the introduction of boars is carried out on a specific day, the exclusion period may be set only for the reference observation time on that specific day. The counting unit 113 excludes the exclusion period set in this way from the counting targets. Specifically, processing such as stopping counting by the counting unit 113, causing the detection unit 112 to stop detecting mounting behavior, causing the acquisition unit 111 to stop acquiring image data, and causing the camera unit 210 to stop capturing images may be adopted.

[0035] Furthermore, the cycle at which the detection unit 112 detects the mounting behavior using image data is adjusted and set, taking into consideration the agility of the movements of the sow 302 and the interval between each mounting behavior, etc., so that one mounting behavior is detected as one occurrence. The cycle can also be set taking into consideration other factors such as the breed, rearing environment, and age of the sow 302. Alternatively, several frames of images per second may be continuously obtained for a certain period of time, and the continuous behavior of the sow 302 being observed may be analyzed to detect one mounting behavior as one occurrence.

[0036] The counting unit 113 manages the number of times of jumping over, which is counted for each pen 301, using a counting list 122. Fig. 6 is a diagram for explaining an example of the counting list 122. One counting list 122 is generated for each reference observation time, and is updated appropriately during that observation time.

[0037] The counting list 122 includes a notification threshold and an observation date. The notification threshold is a value that is set in advance by the manager, etc., and when the number of times of riding counted in each pen 301 exceeds this notification threshold, the counting unit 113 outputs an excess notification to the zookeeper terminal 220, etc. to inform the zookeeper of the fact. The manager, etc. sets the notification threshold in consideration of the breed and rearing environment of the sows 302, particularly the number of sows 302 housed in one pen 301, etc. More specifically, the threshold that can determine that a sow 302 showing signs of estrus is present in the target pen 301 is set based on statistics and experience up to that point. In the example of FIG. 6, "30 times" is set. The observation date is the date on which the observation is performed, and when the counting list 122 is referred to after the observation, it represents the date on which the observation was performed. If the reference observation time is less than 24 hours, for example, the observation time when the observation started may be added.

[0038] The counting list 122 includes a tally table showing the number of mounting behaviors counted for each pen 301. The tally table is made up of pen numbers (for example, eight pens from the first to the eighth pens), the counted number, and flag information. When the counting unit 113 confirms that a pen 301 has appeared whose counted number exceeds the notification threshold, it generates an excess notification and outputs it to the keeper terminal 220 or the like. At this time, the counting unit 113 adds pen information related to the pen 301 whose counted number exceeds the notification threshold to the excess notification and outputs it. In this embodiment, the pen number is added as the pen information. For the pen 301 that has output the excess notification, the information "notified" is recorded as flag information.

[0039] The counting list 122 includes information on the exclusion period. Specifically, the exclusion period described with reference to FIG. 5 is recorded as list information. Note that the longer the exclusion period, the shorter the period to be counted, so even if there is a female pig showing signs of estrus, the number of times counted may not exceed the notification threshold. Therefore, the counting unit 113 may automatically correct the notification threshold in consideration of the ratio of the total time of the set exclusion period to the reference observation time. For example, if the reference observation time is 24 hours and the total of the exclusion periods is 3 hours, the notification threshold is corrected to 30×(24−3) / 24=26.25 times. In this case, the counting unit 113 outputs an excess notification when it is confirmed that a pen 301 has appeared whose counted number of times of overtaking exceeds the corrected notification threshold.

[0040] Furthermore, if the number of sows 302 housed in each pen 301 is different, the notification threshold may be corrected for each pen, taking into consideration the number of pigs housed. For example, if the notification threshold is set to "30 times" assuming that one pen houses 10 sows 302, the notification threshold is corrected to 30 x (8 / 10) = 24 times for a pen 301 housing eight sows 302. In this case, the counting unit 113 outputs an excess notification when the number of times of boarding counted for the pen 301 housing eight pigs exceeds 24 times.

[0041] Fig. 7 is a diagram showing an example of the display of the keeper terminal 220 that has received an excess notification. As described above, when the counting unit 113 outputs an excess notification, the keeper terminal 220 receives the excess notification and displays the contents of the excess notification on the display panel. Specifically, as shown in the figure, the keeper terminal 220 refers to the pen information added to the excess notification and displays the pen number where the number of mounting actions has exceeded the notification threshold (prescribed value) (in the example shown in the figure, "pen 6"). Furthermore, if an indoor map of the pen layout of the pig farm is held, it is also displayed so that the position of the corresponding pen can be recognized. The keeper terminal 220 may issue a notification sound together with this display.

[0042] Next, a process procedure of the pig breeding support method using the server 100 will be described. Fig. 8 is a flow diagram explaining the process procedure of the calculation unit 110. The flow starts from the start point of the reference observation time. Note that the process for the excluded period will be omitted here.

[0043] In step S101, as an initial process at the start of observation, the counting unit 113 starts measuring time using a time-elapsed timer T and starts a counter C for counting the number of times the sow 302 housed in the nth pen gets on top of the sow 302. n In this example, the number of pens in the pig farm is m, and the values ​​C1 to C m A counter is provided up to

[0044] The counting unit 113 proceeds to step S102 and sets the variable n to 1. In response to this, the counter to be processed is switched to C1. The acquisition unit 111 proceeds to step S103 and acquires img from the camera unit 210 directed at the n-th pen. n The image data of img1 is acquired via the communication unit 130. When n=1, image data of img1 is acquired from the camera unit 210 directed to the first pen. The acquisition unit 111 passes the acquired image data to the detection unit 112.

[0045] In step S104, the detection unit 112 detects the image img n is read from the storage unit 120 and input to the detection neural network 121 to generate the image img n The detection unit 112 outputs the number of sows 302 that are mounting within the cage (detection number). The detection unit 112 passes the detection number to the counting unit 113. The counting unit 113 outputs the C n The number of detections received from the detection unit 112 is added to the counter value of C n Update.

[0046] In step S106, the counting unit 113 calculates the updated C n The value of the notification threshold C dDetermine whether or not the notification threshold C has been exceeded. d If it is determined that the n-th pen has been exceeded, the process proceeds to step S107, where an excess notification is generated and output to the keeper terminal 220. At this time, the n-th pen is added as pen information. After outputting the excess notification, the process proceeds to step S108. d If it is determined that the number of times ...

[0047] The counting unit 113 increments the variable n in step S108 and proceeds to step S109. When proceeding to step S109, it is determined whether or not the variable n has exceeded the number of pens m in the pig farm. If it is determined that the variable n has not exceeded the number m, the process returns to step S103 and executes the same process on the incremented variable n. If it is determined that the variable n has exceeded the number m, the process proceeds to step S110.

[0048] When the process proceeds to step S110, the counting unit 113 counts up to the reference observation time T c It is determined whether the time has been exceeded. If it is determined that the time has not been exceeded, the process returns to step S102 after an interval according to a preset cycle. If it is determined that the time has been exceeded, the process ends. If successive observations are to be performed, the process starts again from step S101.

[0049] Next, some other examples according to this embodiment will be described. Figure 9 is a diagram showing an overall view of a pig farming environment in which a pig breeding support device according to another embodiment is adopted. The same elements as those in Figure 1 are given the same reference numerals and their description will be omitted.

[0050] In the embodiment shown in FIG. 9, the keeper does not have a keeper terminal 220, and instead, a notification light 240 is installed adjacent to each pen 301. Each notification light 240 is connected to the server 100 via the wireless unit 230 and the network 200. For example, when the number of times of mounting behavior counted in the fifth pen exceeds the notification threshold, the server 100 transmits a notification signal equivalent to an excess notification to the notification light 240 installed adjacent to the fifth pen, and turns on the notification light 240. By using such notification light 240, the keeper can recognize the pen 301 to go to even if he does not have the keeper terminal 220. The keeper can search for the sow 302 housed in the pen 301 with the notification light 240 turned on, which shows signs of estrus.

[0051] FIG. 10 is a diagram for explaining the counting list 122' of a pig breeding support device according to yet another embodiment. In the embodiment described above, a plurality of sows 302 housed in one pen 301 were not recognized as being distinguished from one another. Therefore, if any of the sows 302 housed in a specific pen 301 to be observed performs a mount action, it is detected as one mount action. For example, when the notification threshold is set to 30 times, the counting unit 113 outputs an excess notification even if one sow 302 performs a mount action more than 30 times, or even if ten sows 302 each perform a mount action 3 to 4 times. That is, even if the keeper knows which pen 301 has detected a mount action exceeding the specified value, it is necessary for the keeper to determine whether a specific sow 302 is showing strong signs of estrus, whether there are multiple sows 302 that have begun to show signs of estrus, or whether the state is a combination of these.

[0052] Meanwhile, techniques for distinguishing and recognizing the multiple sow pigs 302 housed in each pen 301 are becoming known. For example, individual identification can be realized by attaching an identification marker to each sow pig 302 and analyzing an image captured by the camera unit 210. Alternatively, each sow pig 302 may be imaged at the time of being housed in the pen 301 and associated with an identification number in the image, and a learning model may be used to detect which image associated with which identification number a sow pig 302 detected to be climbing on top of another pig matches.

[0053] The counting list 122' shown in Fig. 10 is a counting list in the case where the sows 302 housed in each pen 301 can be individually identified. Each pen 301 houses, for example, ten sows 302, and each sow 302 is distinguished by an assigned identification number. The detection unit 112 detects a mounting behavior and identifies the identification number of the sow 302 that has performed the mounting behavior. The counting unit 113 receives this information from the detection unit 112 and updates the number of mounting times of the sow corresponding to the identified identification number.

[0054] In addition, when individual identification is performed, it is preferable to set the notification threshold to a smaller value than when individual identification is not performed. In the counting list 122', the notification threshold is set to 10 times. When a specific sow 302 exceeds this notification threshold, the counting unit 113 outputs an excess notification to the keeper terminal 220, which includes information on the identification number of the specific sow 302. If the notification light 240 described with reference to FIG. 9 has a display unit, the counting unit 113 may output an excess notification to the notification light 240 adjacent to the pen in which the specific sow 302 is housed, and display the identification number on the display unit of the notification light 240. If the keeper can also obtain information on the specific sow 302, he or she can easily find the specific sow 302 from among the multiple sows 302 housed in the pen 301.

[0055] In the present embodiment described above through several examples, one camera unit 210 is provided for each pen 301, but a camera unit may be provided that overlooks a plurality of pens 301 collectively. In that case, the acquisition unit 111 may divide the image acquired from the camera unit along the boundaries of the pens 301 and sequentially transfer each of the divided images to the detection unit 112.

[0056] In the present embodiment described above, the number of times the pigs climb on each of the multiple pens 301 is counted, but the pig rearing support device may count the number of times the pigs climb on each of the single pens 301. Even if the pig rearing support device is only informed that a female pig showing signs of estrus is present among the female pigs 302 reared collectively in the pen 301, the work load of the keeper can be reduced.

[0057] In the embodiment described above, the output destination of the excess notification is the keeper terminal 220 or the notification light 240, but this is not limited to the above. The counting unit 113 may directly display information related to the excess notification on the display monitor 150 connected to the server 100. In addition to outputting the excess notification when the counted number of climbs exceeds the notification threshold, or instead of outputting the excess notification, the counting unit 113 may constantly output the number of climbs being counted. For example, the current number of climbs counted in each pen 301 may be displayed in a list on the keeper terminal 220.

[0058] In the above embodiment, the server 100 functions as a pig raising support device, but the hardware configuration is not limited to this. If the mobile terminal described as the keeper terminal 220 performs the same processing as the server 100, the mobile terminal can function as a pig raising support device. For example, if the keeper terminal 220 is configured to take on part of the processing of the server 100, a system in which the server 100 and the keeper terminal 220 cooperate with each other can become a pig raising support device. [Explanation of symbols]

[0059] 100...server, 110...calculation unit, 111...acquisition unit, 112...detection unit, 113...counting unit, 120...storage unit, 121...detection neural network, 122, 122'...counting list, 130...communication unit, 150...display monitor, 160...input device, 200...network, 210...camera unit, 220...keeper terminal, 230...wireless unit, 240...notification light, 301...pen, 302, 302a, 302b...sow pig

Claims

1. an acquisition unit that acquires image data of an image captured by a camera unit that is installed facing a pen in which sows are bred in groups; A detection unit that detects the mounting behavior of the sow based on an image of the image data without distinguishing between individuals; a counting unit that counts the number of times the mounting behavior is detected during a set observation time for each pen in which the sow is housed; A pig breeding support device equipped with:

2. The pig raising support device according to claim 1 , wherein the counting unit outputs an excess notification when the number of times exceeds a set threshold value.

3. The pig breeding support device described in claim 2, wherein the counting unit counts the number of times the climbing behavior occurs for each of the multiple pens, and when outputting the excess notification, adds pen information regarding the pen that is the observed object that exceeded the threshold to the excess notification.

4. The pig raising support device according to claim 1 , wherein the detection unit detects the riding behavior using a learning model trained from a teacher image including an area where pigs overlap each other.

5. The pig breeding support device according to claim 1 , wherein the counting unit excludes from the counting a certain period of time including a time from when a keeper approaches the pen to when he or she leaves the pen.

6. 6. The pig raising support device according to claim 1, wherein the counting unit excludes boars from the counting for a certain period of time including the time from when they are introduced into the pen until when they are removed.

7. An acquisition step of acquiring image data of an image captured by a camera unit installed facing a pen in which sows are group-raised; a detection step of detecting a mounting behavior of the sow based on an image of the image data without distinguishing between the sows; a counting step of counting the number of times the mounting behavior is detected during a set observation time for each pen in which the sow is housed; A pig breeding support method having the above-mentioned features.

8. An acquisition step of acquiring image data of an image captured by a camera unit installed facing a pen in which sows are group-raised; a detection step of detecting a mounting behavior of the sow based on an image of the image data without distinguishing between the sows; a counting step of counting the number of times the mounting behavior is detected during a set observation time for each pen in which the sow is housed; A pig-raising support program that runs on a computer.

Images