CN117915861A - Detecting oestrus in animals for insemination - Google Patents

Abstract

A method for detecting oestrus of sow comprises inseminating after determining. In a first method, images of the posterior contours of each sow are analyzed to determine if the sow is in estrus based on characteristics related to shape, size and/or color of the vulva. The analysis may use a program in which a number of images showing the vulvar region of a multi-headed sow known to be in estrus and known not to be in estrus have been previously analyzed to generate data on the previous images to create a program which uses the data to generate a signal to make the determination. In the second method, boars are housed and in intimate contact with sows, and the image and/or sound of each sow is analyzed to determine if the sow is in estrus.

Description

Detecting oestrus in animals for insemination

Technical Field

The present invention relates to a method of detecting oestrus in animals for insemination.

Background

For efficient animal husbandry, many attempts have been made to find effective methods to determine when the relevant animal is in estrus in order to inseminate with a better probability of success.

It is well known that changes in exercise and behavior can be used to detect estrus, including changes in the ratio of stance to lie. In addition, the temperature of the body, particularly the temperature of the vaginal region, may also be used to make the determination.

But further improvements in techniques are needed to make the determination.

The invention is particularly applicable to sows where workers typically contact less with animals and therefore a more automated method is needed to determine oestrus. But the method can be used for other animals.

Disclosure of Invention

The invention aims to provide a method for raising livestock, which comprises the following steps:

defining a containment area in which a group of sows are housed, which may be in estrus;

Providing a male pig which is housed or housed;

The containment area and the container are arranged so that sows in the containment area can be in close contact with the boars through movement of the boars or sows;

Providing at least one collecting device arranged to collect images and/or sounds of each sow upon said intimate contact; and

The images and/or sounds are analyzed using the control system to determine if the sow is in estrus for insemination.

Thus, the methods herein may use images from a camera, may use sound produced by a sow, or may combine both sets of data for analysis when the sow is in intimate contact with a boar.

In a preferred method, images are taken from the front of the sow to show the body and head, as these images indicate changes in the movement and the behavior that determine oestrus. In particular, when a sow is in close proximity to the front of a boar, an image of the front view of the sow may be taken from the front where the predicted changes will be most pronounced.

In a preferred method, the images are analyzed for movement of sow oestrus characteristics as determined by previous analysis of sows known to be in oestrus and comparison with one or more images currently obtained. Multiple comparisons may be made using statistical analysis to obtain more accurate results to discard errors.

In particular, the image may be analyzed for known estrus characteristics, including:

Movement speed characteristics of oestrus.

Body movement characteristics of oestrus.

Head movement characteristics of oestrus.

Ear movement characteristics of oestrus.

Angular velocity characteristic of the ear in oestrus.

Posterior region movement characteristics of oestrus.

Body posture change characteristics of oestrus.

In particular, sound may be analyzed for known estrus characteristics, including:

Sow oestrus characteristics produce a specific type of sound, such as screaming sounds, gurgling sounds.

Spectral features of oestrus.

A series of sound features of oestrus.

Preferably, the collection means comprises a camera mounted on and directed at the sow on a means for housing or housing the boar.

In one embodiment, the means for containment or containment includes a movable structure that brings the boar aside the sow while in a separate containment pen. This is typically a boar cart in which the boar is confined and then moved along a small aisle over a supporting floor to a sow housed in a crate along the small aisle. Boar carts provide the best way to house and control large boars as they are transferred from one sow to another. However, other means of transporting the boar may be used, such as a simple tethering or guiding robot means, on which the boar is tethered when walking from one sow to another.

In another arrangement, the means for housing or housing the boar is secured as a fixed cage within an open containment area where the sow is free to move from place to place, including close to the boar, and the boar is housed separately within its own cage within the containment area.

In both cases, the camera is preferably located on a cart or on a cage facing the sow to be imaged, so as to obtain a series of sow images as the sow approaches or is approached, to use these images to analyze the movement of the sow, as described above.

Preferably, the microphone is also located on the boar cage and is of the type that is directional to receive the primary sound from the sow at the boar container.

Although the above statement uses the term "boar", it is understood that it may be a truly living animal. However, a boar may be a virtual structure that provides one or more stimuli to a sow that mimics a living animal. In extreme cases, a "boar" may simply use a pheromone spray to adequately simulate a boar. Living animals have been found to be more effective in stimulating sows to detect oestrus and receive insemination. However, the improvement in simulated stimulation may avoid the necessity of transporting or housing actual living organisms and very large animals.

In an important alternative function, which may be used with any of the above statements or any of the functions defined below, the control system is operative to distinguish between oestrus determination requiring a simple analysis in which an output signal indicative of a determination of oestrus of a sow is provided and complex analysis in which an output signal requiring additional intervention is provided prior to oestrus determination. That is, the system may generate and indicate a confidence level in the results, allowing the system to directly guide to determining or indicating that the confidence is low, requiring further data to be provided. For example, the further data may be for other interventions from the methods described below or from workers. As another example, during collection of sound and/or images, a worker may be employed to apply pressure to the back area of the sow.

As a result of the determination, insemination is performed after receiving a signal indicating that a final decision has been made, thereby increasing the likelihood that the insemination procedure will be successful.

In another method, which may be used as a stand alone system or as part of the above analysis, there is provided a method of raising livestock comprising:

defining a containment area in which a group of sows are housed, which may be in estrus;

periodically collecting images of the posterior contours of each sow;

analyzing each image to determine an area on the image that shows the sow's vulva;

Analyzing the image of the area using a control system to determine if the sow is in estrus based on characteristics related to shape, size and/or color of the vulva;

Providing an output signal in response to a determination that the sow is in estrus;

wherein the insemination may be performed upon receipt of a signal indicating that the determination has been made.

Determining whether a sow is in estrus may be used for timing insemination or simply to know that the animal is entering estrus. In some cases, breeders of some sows wish to skip estrus cycles to optimize reproductive outcome.

In a preferred method, the analysis is performed by the control system using a program in which a number of images showing the vulvar region of a multi-headed sow known to be in estrus and known not to be in estrus have been previously analyzed to generate data on the previous images to create a program which uses the data to generate signals to make the determination. Thus, the program operating the control system is a neural network that learns characters in advance to determine the existence of estrus by analyzing a large number of images of the vulva region of an animal known to be in estrus and known not to be in estrus. Such pre-learning programming may be stored in a control unit, which may be local or cloud, for application to the new image to be analyzed.

While the machine learning described above is performed by the system itself, the inventors have found through extensive analysis of the image that a determination may be made based on, for example, some visual features that may be extracted from the image. These features may include one or more of the following:

-a-texture or wrinkles of the vulva; this is associated with a degree of swelling of the area, as no wrinkles are created by the increased volume caused by blood congestion, which stretches the skin to reduce the depth and appearance of wrinkles.

-B-height or width of the vulva; again, this is associated with a degree of swelling of the region, as the height and width of the region increases due to swelling.

-C-vulva colour; the presence of engorged blood can darken red, which can be observed.

-D-outward protrusion of the vulva. Also, swelling this area by increasing the blood supply will act to spread the vulva towards the outside of the body to produce an enhanced 3D effect which can even be observed from a 2D image when taken or selected at a suitable angle.

In some cases, particularly for elderly animals that have born many young animals, the vulva region may have less pronounced variation, and therefore analysis of a single image may not be decisive. In this case, the time-dependent changes can be observed using a comparison with previous images from the same animal, wherein at least one image of the region is compared with previous images of the same sow region to make the determination using the time-dependent differences. That is, in many sows, the change is very noticeable, and the presence of oestrus can be detected from a single image or a series of images taken simultaneously, and thus can represent the current state at the time of testing. It will be appreciated that the image and its determination are repeated over a period of time (e.g., daily) to be determined immediately after the onset of estrus. Thus, there is a series of images available for comparison over a period of time. In other cases, a comparison with the old image is required to make the determination. However, of course, this requires that the image be associated with the identification of the relevant animal, which is easily done if the animal is housed, but requires additional action to be taken if the animal is free to move.

In the first step of analysis, a plurality of images of the rear outline are preferably obtained and analyzed to select one or more suitable images that best display the region. The cameras are mounted and arranged so that the angle and position relative to the animal's back may be varied and many images of the back contour may not be suitable for displaying the vulvar region so that the control system uses the program to select the appropriate images from the images and uses those images for determination. As part of this analysis, the system must select the relevant vulvar region from the entire image of the posterior contour so that a portion of the algorithm requires that the region be first determined and selected onto the image component for further analysis. That is, after the vulvar region is selected, it is necessary to select those image components from the subset of image components that are most likely to display the region and apply to the determination. The system then evaluates each image to determine if it indicates a positive or negative result. After the determination, a comparison of all positive and negative results will result in a final determination and a confidence level, where the confidence level is represented by how much of the results are consistent. This information can be used to assess whether insemination should be performed, i.e. whether the vulva is showing signs of oestrus, or whether further analysis and further data are required.

After the first analysis, the numbers in the subset that best or suitably show the vulvar region are selected from the posterior contour image, and the multiple images of that region are selected and compared to obtain a determination using statistical analysis.

In some embodiments, the sow is housed in a separate pig pen and the image is obtained by a camera mounted on the pig pen. In this case, the camera may be mounted in a fixed position on the pig pen and timed and triggered to repeatedly take images of the rear profile at a systematically determined frequency (e.g., daily). The camera may then take multiple images of the rear profile for the analysis described above. This allows for the animal to be out of position or at a suitable angle relative to the camera to obtain the best posterior profile image. In other cases, the images are taken by a moving camera (on track, robot, etc.), and therefore these images may not be all suitable.

It is preferable to take an image of the posterior contour while the animal is standing, as this ensures that the vulvar region is presented correctly posteriorly and positions it at a specific location in the image, making selection easier. Furthermore, such standing positions may ensure that portions of the area are presented in a consistent manner in the image for more efficient analysis.

Thus, in acquiring images, it is desirable to ensure that the animal is standing, even by some form of physical stimulus, moving the animal to a standing position to obtain the image.

In some embodiments, the sows are not housed in separate pigs so that they can move freely. In this case, the image may be acquired by a camera mounted in a suitable location, such as on a feeder or water rail that is frequently attended by a sow. This may also be at a boar exposure station where there is an opening for the sow to see the boar for sexual stimulation. In this way, each sow will move near the camera at random times and will be imaged each time it moves to the relevant location. In one arrangement, in this case, it is necessary to apply an Identity (ID) to the imaged animal so that the animal can be identified for insemination. Thus, in this case, the animal may be tagged with a readable tag, such as an RFID tag. This Identity (ID) may also be used for the history comparison described above.

Alternatively, in the absence of Identity (ID) information, after the determination, a marker may be applied to the sow that determines the subject for subsequent insemination. Such marking systems by painting or tapping are available and can be activated to mark the relevant sow. This is particularly effective when the imaging camera is located on a feeder boom whose walls restrict one animal at a time so that individual animals can enter, can be restricted to taking pictures in the proper direction, imaging and determining before placing the animals back into the population. Alternatively, a lane may be provided between the two areas, such as a rest area and a feeding area, where the animal is temporarily kept in place for subsequent action before proceeding to the next area.

In some cases, the system may be provided as a program that is applicable to existing imaging systems such as mobile phones. In this case, the image may be manually photographed by a worker of the hand-held imaging device, who provides information about the animal for data storage, and selects an image to be photographed. The control system takes over and selects the area image to be used from among these images, and makes a decision from among the plurality of images and communicates it to the worker. In other cases, the control system may be part of a stand-alone system including a camera, a processor for making the determination, and an interface for communicating with the worker. In both cases the control system may be arranged to provide data related to the determination of the multi-headed sow to the sow administration system.

As described above, the method of using the image may be used in combination with providing a sow that is housed or housed, wherein the containment area of the sow and the confinement system of the sow are arranged such that the sow within the containment area can be brought into intimate contact with the boar by movement of the boar or sow.

Drawings

An embodiment of the invention will now be described with reference to the accompanying drawings, in which:

fig. 1 is a schematic diagram of a livestock house showing an arrangement used in a first embodiment of the method according to the invention.

Fig. 2 is a schematic diagram of a stall showing an arrangement used in a second embodiment of the method according to the invention.

Fig. 3 is a flow chart illustrating the steps of a first embodiment of the method according to the invention.

Fig. 4 is a schematic diagram of a livestock house showing an arrangement used in a third embodiment of the method according to the invention.

Fig. 5 is a schematic diagram of one example of a rear view of an animal for use in a third embodiment of the method of the invention.

Fig. 6 is an illustration of one example of an image of an animal area taken from fig. 5 and used in a third embodiment of the method of the present invention.

Fig. 7 is a flow chart illustrating the steps of a third embodiment of the method of the present invention.

Detailed Description

Fig. 1 shows an assembly for use in a method of raising livestock. The containment zone 10 includes a plurality of breeding pens 11 side by side for containment of sows arranged in a row along a lane 14.

The pen consists of a pen wall of welded strips with a trough 12 on the front wall to keep the animals in containment during the breeding period in which a group of sows may be in estrus.

A boar is accommodated in a movable boar cart 15 on wheels 16 and is thus movable along the lane 14 to be brought to the side of each sow in turn and brought into close contact with the boar by the movement of the boar. Suitable carts are shown in prior U.S. patent 6196975 (Labrecque) issued 3/6/2001 and U.S. patent 6508202 (KLEINSASSER) issued 1/21/2003, both of which are incorporated herein by reference.

The cart carries a camera 17 and microphone 18 in a fixed position on the cart so that the camera takes images and the microphone collects the sound of each sow as it comes into said intimate contact. Thus, when the cart is aligned with the sow pen, the camera and microphone are in place on the cart frame to guide to the desired location.

The images and sounds are transmitted by wire or wirelessly to the control system 19 which analyzes the data as described below to determine if the sow is in estrus for insemination.

As shown in the flow chart of fig. 3, the method uses images from a camera or sound generated by the sow, or may combine analysis of both sets of data, where the images and sound are acquired while the sow is in close contact with the boar. The steps of the method are shown in fig. 3.

In a preferred method, images are taken from the front of the sow as front views to show the body and head, as these images show the changes in the movements and the behavior that determine oestrus. In particular, when a sow is near the front of a boar, an image of the front view of the sow may be taken from the front where the expected changes will be most pronounced.

As shown in fig. 3, the images are analyzed for movement of sow oestrus characteristics determined by previous analysis of sows known to be in oestrus and comparing the currently acquired image or images with the materials in the database. Multiple comparisons may be made using statistical analysis to obtain more accurate results to discard errors.

In particular, the image may be analyzed for known estrus characteristics, including:

Movement speed characteristics of oestrus.

Body movement characteristics of oestrus.

Head movement characteristics of oestrus.

Ear movement characteristics of oestrus.

Angular velocity characteristic of the ear in oestrus.

Posterior region movement characteristics of oestrus.

Body posture change characteristics of oestrus.

In particular, sound may be analyzed for known estrus characteristics, including:

Sow oestrus characteristics produce a specific type of sound, such as screaming sounds, gurgling sounds.

Spectral features of oestrus.

A series of sound features of oestrus.

As shown in fig. 2, the sow is free to move within the containment pen 20 and to move to the feeder 21, the water supply 22 or the rest area as desired. The boar enclosure 23 is secured as a fixed cage within an open containment area where the sow is free to move from place to place, including close to the boar, which is individually restrained within its own cage within the containment area. In this case, the camera 24 and microphone 25 are located at a position on the cage to face the sow when the sow approaches the boar cage from the front of the boar. In this way, a series of images of the sow may be obtained as the sow approaches to use these images to analyze the movement of the sow, as shown in fig. 3. In fig. 2, short dividing walls 25A and 25B extending forward from the boar cages on each side of the sow are provided when standing in front of the boar, in order to obtain and analyze signals from only one animal and separate from other animals in the same general area.

Microphone 25 is an elongated directional microphone to receive primary sound from the sow directed onto the boar container and to discard sound from other areas of the restricted area, such as other sows.

As shown in fig. 3, the control system operates to distinguish between oestrus determinations requiring a simple analysis in which an output signal indicative of a determination of oestrus in a sow is provided and complex analysis in which an output signal requiring additional intervention is provided prior to oestrus determination. That is, the system may generate and indicate a confidence level in the results, allowing the system to directly guide to determining or indicating that the confidence is low, requiring further data to be provided. For example, the further data may be a response to a request from the method described below or from other intervention by a worker. As another example, during collection of sound and/or images, a worker may be employed to apply pressure to the back area of the sow.

As a result of the determination, insemination is performed after receiving a signal indicating that a final decision has been made, thereby increasing the likelihood that the insemination procedure will be successful.

In another approach, shown in figures 4 to 7, a containment area 30 is provided housing a group of sows, wherein the sows may be in estrus. Containment zone 30 includes individual pens 31 and 32 into which sows can enter to obtain feed from a trough 33 or water supply 34. The pen is typically arranged so that sows can be accessed individually, while other sows are excluded from accessing the rear of the pen. Each pen is provided with a fixed camera 35 which operates to periodically collect images of the rear profile of each sow as it enters the pen. Since all sows will eat regularly, all sows will be the subject of the imaging system during the day. In this way, each sow is moved to the camera 35 at random times and can be imaged each time it is moved to the relevant location. In one arrangement, in this case, it is necessary to apply an Identity (ID) to the imaged animal so that the animal can be identified for insemination. Thus, in this case, the animal may be tagged with a readable tag, such as an RFID tag, detected by tag reader 37 at rail 31.

As shown in the flowchart of fig. 7, where these steps are shown in detail, the control system receives signals from the camera and performs a series of steps as illustrated for analyzing the image for oestrus determination.

Thus, each image of the posterior outline shown in fig. 5 was analyzed using the control system shown in fig. 7 to determine the area on the image shown in fig. 6 where the sow vulva was displayed.

In some cases, the flow chart may be slightly different depending on the method used to analyze the image. A method may not require selection from a larger image of the vulvar region before selecting the appropriate image. According to other criteria, the appropriate image may be selected for analysis based on the angle of image capture, the orientation of the animal at the time the image was acquired, etc., to ensure that the best image is used for analysis.

Thus, the vulvar region image of fig. 5 is analyzed to determine whether a sow is in estrus for insemination based on characteristics related to shape, size and/or color of the vulva. As described above, this may be done after selecting an appropriate image to extract the relevant region from a larger image. Or the entire image may be analyzed without selecting any region. Depending on the method of obtaining the image, the appropriate image may be selected for analysis to discard some images that are unsuitable due to the angle or orientation or position of the animal.

An output signal is generated indicating that the determination has been made, wherein insemination is performed after receipt of the signal.

In a preferred method, the control system uses a program to analyze, wherein prior data showing known estrus status of the vulvar area of the multi-headed sow is compared by the program. Thus, the program operating the control system is a neural network or other machine learning algorithm that has learned features in advance to determine the presence of oestrus by analyzing a large number of images of the vulvar region of the animal known to be in oestrus and known not to be in oestrus. The pre-learning program is stored in the control unit to be applied to a new image to be analyzed. Or it may even be a classical algorithm, rather than machine learning, using classical computer vision techniques.

In the first step of analysis, a plurality of images of the rear outline are preferably obtained and analyzed to select one or more suitable images that best display the region. The control system uses a program to select the appropriate images from among the images and uses them in the determination. As part of this analysis, in some cases, the system needs to select the relevant vulvar region from the entire image of the posterior contour, so that a portion of the algorithm requires that the region be first determined and selected on the image component for further analysis. The system then evaluates each image to determine if it indicates a positive or negative result. After the determination, a comparison of all positive and negative results will result in a final determination and a confidence level, where the confidence level is indicated by how much of the results are consistent. This information can be used to evaluate whether insemination should be performed or whether further analysis and further data is required.

One analysis that may be used to identify images is based on the determination of vulvar cleanliness by the analysis system. In this case, if the vulva region is too dirty and may produce erroneous results, a warning may be issued to the farmer. In this case, the image may be discarded and a signal provided to the farmer indicating that action must be taken to clean the area before an appropriate accurate decision can be made.

In a further step of the analysis, the size may not be determined on the basis of the actual size, but on the basis of the size of the vulva compared with a determinable area of the sow's rear outline, for example the size of the buttocks or the size of the total area of the animal present in the acquired rear or cross-section view. Thus, the analysis can determine the ratio of the two regions selected by the imaging system and the detected change in ratio caused by the animal entering estrus.

After the first analysis, the number of the most preferred or suitable for displaying the vulvar region in a subset is selected from the posterior contour image, and a plurality of images of that region are selected and compared to obtain a determination using statistical analysis.

Another arrangement is shown in fig. 1, in which a camera 50 is located at the rear of the pen 11 for capturing images of the restrained animal in the presence or absence of a boar in the cart. The image of the rear outline is preferably taken while the animal is standing, as this ensures that the vulvar region is presented correctly backwards and positions it at a specific location in the image, making it easier to select. Furthermore, such standing positions may ensure that portions of the area are presented in a consistent manner in the image for more efficient analysis.

Thus, the camera 50 and associated control system operate to acquire or select images while the animal is standing. That is, the animal may be monitored to determine that an image is taken while it is standing in the crate, as shown in fig. 1, or that an actuator 51 is provided on the wall of the crate to move the animal to a standing position by physical stimulus to obtain an image. Or if the system determines that the animal is not standing, the images may be taken periodically and simply discarded.

If there is no Identity (ID) information on the animal, the animal is free to move, as shown in FIG. 4, and after the determination, a marker can be applied to the sow by the marker 36, the sow being the subject of the determination in the pen 32 for subsequent insemination. Such systems for marking by painting or tapping are known and commonly available and can be activated to mark the relevant sow to indicate that it has been determined to be in estrus.

It should be noted that the time of insemination may be offset in time from the analysis that determines the state of estrus. In some cases, the actual time of estrus may be delayed by analyzing and detecting a significant change in the appearance of the vulva. In fact, the vulva generally changes appearance some time before standing heat reflex (commonly used to determine sow oestrus), sometimes for several days after standing heat reflex (oestrus). Thus, the appearance of the vulva is a good indicator of the existence of oestrus, but not necessarily a good way to find the actual time of insemination. However, it can be used to confirm that estrus is coming or starting. Another system, such as the image/sound method described above, may then be employed to better determine the time of insemination required. Or the vulvar imaging system may be used to notify of estrus events that are preferably not to be inseminated. For young or new sows, it is important to find these oestrus events, as farmers may wish to inseminate them in a second oestrus, skipping the first oestrus to ensure better maturation of the animal. It is therefore desirable to detect both the first and second oestrus without missing them, otherwise the animal would become too old at the time of the first insemination, which would reduce the lifetime fertility.

As various modifications could be made to my invention, as described above, and many apparently widely different embodiments made within the spirit and scope of the claims without departing from such spirit and scope, it is intended that all matter contained in the appended description shall be interpreted as illustrative only and not in a limiting sense.

Claims (45)

1. A method of raising livestock comprising:

defining a containment area in which a group of sows are housed, which may be in estrus;

periodically collecting images of the posterior contours of each sow;

Analyzing the image using the control system to determine if the sow is in estrus based on characteristics related to shape, size and/or color of the vulva;

An output signal is provided in response to a determination that the sow is in estrus.

2. The method of claim 1 wherein the analyzing is performed by the control system using a program wherein a plurality of images showing the vulvar region of the multi-headed sow known to be in estrus and known not to be in estrus have been previously analyzed to generate data on the previous images to create a program which uses the data to generate signals to make the determination.

3. A method according to any preceding claim, wherein the features comprise texture or wrinkles of the vulva.

4. A method according to any preceding claim, wherein the features comprise the height or width of the vulva.

5. A method according to any preceding claim wherein the characteristics include comparing the size of the vulva with a determinable area of the posterior profile of the sow.

6. A method according to any preceding claim, wherein the characteristic comprises a vulvar colour.

7. A method according to any preceding claim, wherein the features comprise outward protrusions of the vulva.

8. A method according to any preceding claim, wherein at least one image of the region is compared with a previous image of the same sow region to make the determination using a time-varying difference.

9. A method according to any preceding claim wherein the image is analysed to select the area of the image on which the sow's vulva is displayed for analysis.

10. A method according to any preceding claim, wherein a plurality of images of the rear outline are analysed to select one or more suitable images which best show the region.

11. A method according to any preceding claim, wherein a plurality of images of the region are selected and compared to obtain the determination.

12. A method according to any preceding claim, wherein a plurality of images of the region are selected and compared to obtain the determination using statistical analysis.

13. A method according to any of the preceding claims, characterized in that the sow is housed in a separate pig pen and the images are obtained by means of a camera mounted on the pig pen.

14. The method of any one of claims 1 to 13, wherein the sow is not housed in a separate pig pen and the images are acquired by a camera mounted on the pig pen, wherein the pig pen is frequently accessed by the sow.

15. A method according to any of the preceding claims, characterized in that a system is provided for providing an identity to the cow from which the image was obtained.

16. A method according to any of the preceding claims, characterized in that after estrus determination, a marking can be applied to the sow of the subject for whom a subsequent insemination is determined.

17. A method according to any preceding claim, wherein the control system is part of a stand-alone system comprising a camera, a processor for making the determination, and an interface for communicating with a worker.

18. A method according to any of the preceding claims, characterized in that the control system is arranged to provide data related to the determination of a multi-headed sow to the sow management system.

19. A method according to any preceding claim, wherein the image is manually captured by a worker of the hand-held imaging device.

20. A method according to any of the preceding claims, wherein a housed or housed boar is provided, wherein the containment area and the container are arranged such that sows within the containment area can be brought into intimate contact with the boar by movement of the boar or sow.

21. A method according to claim 20, wherein at least one collecting means is provided, the collecting means being arranged to collect an image and/or sound of each sow at the time of said intimate contact and to analyze the image and/or sound using the control system to determine if the sow is in estrus.

22. A method of raising livestock comprising:

defining a containment area in which a group of sows are housed, which may be in estrus;

Providing a male pig which is housed or housed;

The containment area and the container are arranged so that sows in the containment area can be in close contact with the boars through movement of the boars or sows;

Providing at least one collecting device arranged to collect images and/or sounds of each sow upon said intimate contact; and

The images and/or sounds are analyzed using the control system to determine if the sow is in estrus.

23. The method of claim 21 or 22, wherein the image is taken from the front of the sow to display the body and head.

24. The method of any one of claims 21 to 23 wherein the images are analysed for movement of oestrus features of the sow.

25. The method of any one of claims 21 to 24 wherein the images are analysed for movement speed of oestrus characteristics of the sow.

26. The method of any one of claims 21 to 25 wherein the images are analysed for somatic movements characteristic of sow oestrus.

27. The method of any one of claims 21 to 26 wherein the images are analysed for head movements characteristic of sow oestrus.

28. A method according to any one of claims 21 to 27, wherein the images are analysed for ear movements characteristic of oestrus in sows.

29. The method of any one of claims 21 to 28 wherein the images are analysed for angular ear velocity characteristic of sow oestrus.

30. The method of any one of claims 21 to 29 wherein the images are analysed for movement of the back region of the oestrus characteristics of the sow.

31. The method of any one of claims 21 to 30 wherein the image is analysed for a somatic posture change characteristic of sow oestrus.

32. The method according to any one of claims 21 to 31, characterized in that sound is analyzed for specific types of sounds generated by oestrus characteristics of sows, such as screaming, gurgling.

33. A method according to any one of claims 21 to 32, wherein sound is analysed to find spectral features of oestrus.

34. A method according to any one of claims 21 to 33, wherein sound is analysed for a series of sound features of oestrus.

35. A method according to any one of claims 21 to 34, wherein the collecting means comprises a camera mounted on the means for housing or housing the boar and directed towards the sow on the container.

36. A method according to any one of claims 21 to 35 wherein the means for containment or containment comprises a movable structure, the boar being brought aside the sow when the sow is in a separate containment pen.

37. A method according to any one of claims 21 to 36, wherein the means for housing or containment comprises a container for the boar.

38. A method according to any one of claims 21 to 37, wherein the means for housing or containment comprises a removable tethering means for inducing movement of a live boar.

39. A method according to any one of claims 21 to 38, wherein the means for containment or containment is fixed within the containment area and the sow is free to move towards the boar.

40. The method of any one of claims 21 to 39 wherein the microphone is oriented to receive primary sound from a sow on the boar container.

41. The method of any one of claims 21 to 40, wherein the boar is a real living animal.

42. The method of any one of claims 21 to 41, wherein the boar is virtual, providing one or more stimuli to the sow, mimicking a living animal.

43. The method of any one of claims 21 to 42 wherein pressure is applied to the back area of the sow during collection of the sound and/or image.

44. A method according to any one of the preceding claims, wherein the control system is operative to distinguish oestrus determination requiring a simple analysis in which an output signal indicative of a determination of oestrus in a sow is provided from complex analysis in which an output signal requiring additional intervention is provided prior to oestrus determination.

45. A method according to any preceding claim, wherein insemination is performed after receipt of a signal indicating that a final decision has been made.