WO2019182497A2

WO2019182497A2 - Method, control unit and system to identify animals

Info

Publication number: WO2019182497A2
Application number: PCT/SE2019/050240
Authority: WO
Inventors: Karol FERENC
Original assignee: Delaval Holding Ab
Priority date: 2018-03-20
Filing date: 2019-03-18
Publication date: 2019-09-26
Also published as: WO2019182497A3

Abstract

Method (400), control unit (120) and system (500) for animal identification, based on an individual milk flow characteristic. The method (400) comprises obtaining (402) information related to a milk flow when milking an animal (100a, 100b, 100c, 100d); determining (403) a first characteristic of the milk flow based on the obtained (402) information; and identifying (405) the animal (100a, 100b, 100c, 100d), based on said determined (403) first characteristic of the milk flow.

Description

METHOD, CONTROL UNIT AND SYSTEM TO IDENTIFY ANIMALS

TECHNICAL FIELD

This document discloses a method, a control unit and a system. More particularly, a method a control unit and a system are described, for animal identification, based on an individual milk flow characteristic of the respective animals.

BACKGROUND

On a dairy farm, it is in general a good idea to keep track on the milk produced by each respective animal, such as e.g. cows, on the farm. Thereby several advantages are achieved; advantages/ disadvantages with a modified diet of the animal may for example be observed (e.g. with regard to produced amount of milk). It may for example also be deter mined whether an animal is ill, or if she is ready for retirement. This requires identification of each individual animal.

However, in many milking systems, both conventional and automatic milking systems, proper animal identification is problematic, let be for different reasons. For example, farms where the animals are tied up (Stanchion) typically do not have any automated systems for animal identification. Therefore, in such environments, manual input of an animal identity reference is typically needed.

A known solution for animal recognition is to have a transponder attached to the animal and using a transponder reader to determine the identity of each transponder/ animal at the mo ment of milking. However, identification by using transponders is not always reliable. Tran sponders may be lost or malfunction, two closely situated animals may be confused, or the transponder reader may be dysfunctional due to the rough environment at the farm.

Another possible animal identification is by using cameras in combination with an image recognition program. The animal may for example have a tag with an identification number attached to the ear, which number may be recognised by the image recognition program. However, the animals may lose the tag, or the tag may become too dirty to be readable.

Alternatively, the look/ skin pattern of the animal may be used for animal identification. How ever, also this identification method has problems, as similar looking animals may be con fused, the camera lens may be sensitive to dirt, the camera may by mistake be misdirected during work at the farm, etc. It would be desired to find a reliable manner to identify animals; and/or to confirm identifica tion of the animals at a farm, preferably without the need to invest in very expensive equip ment.

SUMMARY

It is therefore an object of this invention to solve at least some of the above problems and identify animals in a herd, in a reliable manner.

According to a first aspect of the invention, this objective is achieved by a method for animal identification, based on an individual milk flow characteristic. The method comprises obtain ing information related to a milk flow when milking an animal. Further, the method also com prises determining a first characteristic of the milk flow based on the obtained information. In addition, the method also comprises identifying the animal, based on said determined first characteristic of the milk flow.

In a first implementation of the method according to the first aspect, the identification of the animal may comprise comparing the determined first characteristic of the milk flow with pre viously determined individual milk flow characteristics being related to individual animals at one or more previous milkings, each previously determined individual milk flow characteristic being associated with an animal identity. The method may also comprise extracting the ani mal identity, associated with the previously determined milk flow characteristic having the highest correlation with the determined first characteristic.

In a second implementation of the method according to the first aspect, or the first imple mentation thereof, the previously determined individual milk flow characteristics may be based on a mean value over a plurality of milkings.

In a third implementation of the method according to the first aspect, or any implementation thereof, the previously determined individual milk flow characteristics are determined based on previous milking events within seven days preceding the milking from which information related to the milk flow has been obtained.

In a fourth implementation of the method according to the first aspect, or any implementation thereof, further comprising a preceding step of, for a number of milkings: associating the identity of an animal with information related to a milk flow derived during milking of the ani mal. In a fifth implementation of the method according to the first aspect, or any implementation thereof, the method also may comprise determining whether the milking is performed during a first milking event of the day, e.g. a morning milking event, or a subsequent milking event of the day, e.g. an evening milking event. Furthermore, the comparison of the determined first characteristic of the milk flow is made with previously determined milk flow characteris tics determined during the corresponding milking sessions at either the first milking event of the day e.g. only morning, or the subsequent milking event of the day, e.g. only evening.

In a sixth implementation of the method according to the first aspect, or any implementation thereof, the first characteristic of the milk flow may comprise at least one of: flow increase rate, flow increase time, peak flow level, peak flow duration, flow decrease rate, flow de crease time and milk yield.

In a seventh implementation of the method according to the first aspect, or any implementa tion thereof, the first characteristic of the milk flow may comprise the flow increase rate and/or flow increase time.

In an eighth implementation of the method according to the first aspect, or any implementa tion thereof, the method furthermore may comprise: obtaining information of a characteristic of the milk derived during the milking. Also, the identification of the animal may be further based on said determined characteristic of the milk.

In a ninth implementation of the method according to the first aspect, or any implementation thereof, the characteristic of the milk may further comprise: conductivity, amount of fat, milk temperature, levels of progesterone, LDH (Lactate Dehydrogenase), BHB (Beta-Hydroxy- butyrat), urea, and/ or somatic cell count.

In a tenth implementation of the method according to the first aspect, or any implementation thereof, the method may additionally comprise storing the obtained information related to a milk flow in association with an identity reference of the identified animal.

According to a second aspect of the invention, this objective is achieved by a control unit for animal identification, based on an individual milk flow characteristic. The control unit is con figured to obtain information related to a milk flow during milking of an animal. Further, the control unit is configured to determine a first characteristic of the milk flow based on the obtained information. Additionally, the control unit is further configured to identify the animal, based on said determined first characteristic of the milk flow. In a first implementation of the control unit according to the second aspect, the control unit may be additionally configured to identify the animal by comparing the determined first char acteristic of the milk flow with a previously determined milk flow characteristics being related to individual animals at one or more previous milking events, each previously determined individual milk flow characteristic being associated with an animal identity reference. Further, the control unit may be configured to extract the animal identity, associated with the previ ously determined milk flow characteristic having the highest correlation with the determined first characteristic.

In a second implementation of the control unit according to the second aspect, or the first implementation thereof, the previously determined individual milk flow characteristics may be based on a mean value over a plurality of milkings.

In a third implementation of the control unit according to the second aspect, or any imple mentation thereof, the control unit may be further configured to determine the previously determined individual milk flow characteristics, based on previous milking events performed within seven days preceding the milking from which information related to the milk flow has been obtained.

In a fourth implementation of the control unit according to the second aspect, or any imple mentation thereof, the control unit may be further configured to associate the identity of an animal with information related to a milk flow derived during milking of the animal.

In a fifth implementation of the control unit according to the second aspect, or any implemen tation thereof, the control unit may be further configured to determine whether the milking is performed during a first milking event of the day or during a subsequent milking event of the day, e.g. morning or evening. Also, the control unit may be configured to compare the deter mined first characteristic of the milk flow with previously determined milk flow characteristics determined during the corresponding milkings at either a first milking event of the day or a subsequent milking event of the day, e.g. only morning, or only evening.

In a sixth implementation of the control unit according to the second aspect, or any imple mentation thereof, the first characteristic of the milk flow may comprise at least one of: flow increase rate, flow increase time, peak flow level, peak flow duration, flow decrease rate, flow decrease time and milk yield,. In a seventh implementation of the control unit according to the second aspect, or any im plementation thereof, the first characteristic of the milk flow may comprise the flow increase rate and/ or flow increase time.

In an eighth implementation of the control unit according to the second aspect, or any imple mentation thereof, the control unit may be further configured to: obtain information of a char acteristic of the milk derived during the milking. Further, the identification of the animal may be further based on said determined characteristic of the milk.

In a ninth implementation of the control unit according to the second aspect, or any imple mentation thereof, the characteristic of the milk may comprise any of: conductivity, amount of fat, milk temperature, levels of progesterone, LDH (Lactate Dehydrogenase), BHB (Beta- Hydroxybutyrat), urea, and/ or somatic cell count.

In a tenth implementation of the control unit according to the second aspect, or any imple mentation thereof, the control unit may be further configured to store the obtained information related to a milk flow in association with an identity of the identified animal.

According to a third aspect of the invention, this objective is achieved by a system for animal identification, based on an individual milk flow characteristic. The system comprises a control unit according to the second aspect, or any implementation thereof. The system also com prises a database configured to store information related to a respective milk flow of animals, which information is associated with an identity reference of the respective animal.

Thanks to the described aspects, by determining a characteristic of milk extracted from an individual animal, in particular a characteristic related to milk flow of the animal, a certain pattern and/ or level which is typical/ unique for the individual animal may be detected and defined (i.e. typical/ unique at least at a farm level; not necessarily globally unique). The detected unique characteristic may then be stored, associated with an identity reference of the animal having delivered the milk sample. During later made milking events, a character istic of the flow of milk extracted from an animal may be compared with previously stored milk flow characteristics in the database and when a match is made, the animal identity ref erence associated with the stored milk flow characteristic with which the match is made, is extracted. Thereby, the animal identity of the animal of the later milking event is determined and/ or confirmed. A“match” should be understood to be made between the extracted char acteristic and the one previously stored characteristic (out of a set of previously stored char acteristics) being most similar, e.g. having the shortest distance or smallest difference, in some respect, to the extracted characteristic (out of the previously stored characteristics in the set).

By assuring that milk flow data and/ or other characteristic is correctly associated with the correct animal, quality of herd management system data will increase, which in turn will im prove herd management and animal feeding at the farm.

Other advantages and additional novel features will become apparent from the subsequent detailed description.

FIGURES

Embodiments of the invention will now be described in further detail with reference to the accompanying figures, in which:

Figure 1 illustrates an example of an animal and a control unit for identification of ani mals, according to an embodiment of the invention.

Figure 2 illustrates an example of a system for identification of animals, according to an embodiment of the invention.

Figure 3 schematically illustrates milk flow characteristics of an animal, according to an example.

Figure 4 is a flow chart illustrating an embodiment of a method.

Figure 5 is an illustration depicting a system according to an embodiment.

DETAILED DESCRIPTION

Embodiments of the invention described herein are defined as a control unit, a method, and a system, which may be put into practice in the embodiments described below. These em bodiments may, however, be exemplified and realised in many different forms and are not to be limited to the examples set forth herein; rather, these illustrative examples of embodi ments are provided so that this disclosure will be thorough and complete.

Still other objects and features may become apparent from the following detailed description, considered in conjunction with the accompanying drawings. It is to be understood, however, that the drawings are designed solely for purposes of illustration and not as a definition of the limits of the herein disclosed embodiments, for which reference is to be made to the appended claims. Further, the drawings are not necessarily drawn to scale and, unless oth- erwise indicated, they are merely intended to conceptually illustrate the structures and pro cedures described herein.

Figure 1 illustrates a scenario with an animal 100a which may be comprised in a herd of animals at a dairy farm.

“Animal” may be any arbitrary type of domesticated milk producing (non-human) animal, such as e.g. cow, goat, sheep, camel, dairy buffalo, yak, etc.

It is per se well known to measure milk yield, and even milk flow, for animals at a farm. The purpose has, however, never been to use this information in order to identify an individual animal. Instead, this information has been used for other purposes, such as for determining when to remove (take-off) the milking cluster, or for keeping track of the milk production of each animal.

It is also well known that the milk yield of an individual animal 100a is not constant, but varies over a lactation period of the animal 100a, as well as with the age of the animal 100a. Other factors that may influence the milk yield, at least when suddenly changed, are, for example, fodder and water supply of the animal 100a, and also the preparations before a milking event. Illness may also influence the milk yield negatively.

Rather unexpectedly, it has been discovered that certain milk flow characteristics of the an imal 100a could be recognised also at different moments in time, in particular when for ex ample preparations, environment, fodder, etc., are kept relatively constant, or at least is not drastically changed. Thereby, the milk flow characteristics may also be used for identification of animals 100a at a farm; and/ or for correction of animal identity in case other system for animal identification have malfunctioned. The milk flow characteristics could also be used for verification of an animal identity determined by another system for animal identification. These slightly different aspects of identification are intended to be covered by the scope of the claims.

The above mentioned milk production variation of an animal 100a may be compensated for, by establishing a comparison value comprising a mean value based on e.g. the x latest made milking events of the animal 100a; or, for morning/ evening milking events, the x latest made milking events of the animal 100a made during morning/ evening, respectively; wherein 1 £ x < about 40; or perhaps preferably 1 £ x < about 10. By applying a retro-perspective sliding time frame, a mean average value may be calculated which could be used as a reference level for identifying a particular animal 100a, also when the milk yield, duration of milking, milk flow, or other parameter is slowly varying over time. The identification may be made by comparing a later determined milk flow characteristic of the animal 100a with the stored ref erence levels. When a match is found, i.e. when the difference between the determined milk flow characteristic and the reference level is smaller than a threshold level; or alternatively smallest of all made comparisons, the animal identity associated with the matching reference level is extracted. This extracted animal identity may then be associated with the determined milk flow characteristic and the milking event of the animal 100a.

The animal 100a in the illustrated example is connected to a milk line 110 having a sensor 115, configured to measure milk flow of the animal 100a. The measurement values of the sensor 115 are provided to a control unit 120 for animal identification. The control unit 120 may thereby determine commencement and termination of the milking, and thereby also duration of the milking; flow increase up to a peak phase, a milk yield peak value, a flow decrease from the peak phase, total milk yield, etc. Further, the sensor 115, or one or more additional sensors, may be arranged and configured to also measure other characteristics of the extracted milk such as conductivity, amount of fat, milk temperature, levels of proges terone, LDH, BHB, urea, and/ or somatic cell count. These characteristics are not unique per individual, but some animals have milk with a higher conductivity, temperature and/or so matic cell count than other animals over a period of time or on a regular basis, and thereby this information could be used in combination with the milk flow characteristics in order to identify an animal. Other characteristics of the milk that could possibly be used for this pur pose are progesterone, LDH, BHB, and/ or urea.

When a reference value or level of at least one characteristic of the milk flow or milk is to be established during an initialisation period, the animal 100a must be identified and the identity of the animal 100a is to be associated with the reference value or level of the characteristic. The initialisation period may start at an arbitrary moment in time during a lactation period of the animal, but may preferably be made in the beginning of the lactation period, e.g. during the first days of lactation after parturition. The initialisation period may then proceed over a number of days, e.g. 2, 3, 4, 5, 6 or 7 days.

The milk flow characteristics of the animal 100a may be stored e.g. in a database 130 in association with a reference to an established identity of the animal 100a. The identity refer ence of the animal 100a may be determined either manually by a farmer; or automatically, e.g. by a conventional identification system using RFID tags, scanning of a visible code on an ear tag, image recognition of animals, by using GPS (Global Positioning System) posi tioning, or in some other convenient manner. The database 130 may comprise individual milk flow characteristics associated with a respective animal identity reference of various animals of the herd at the farm.

After having established, in an initialisation phase, the database 130 comprising the milk flow characteristics associated with the respective animal identity reference of the animals at the farm, a comparison may be made whenever a new milking event is made. Alternatively, or in addition, a comparison can be made when a milking event is detected, wherein the animal identity could not be established by other means, such as by reading an RFID or other tag of the animal 100a. By comparing the milk flow characteristics of the new milking event with the stored milk flow characteristics, an identity reference may be extracted when a match is made between a determined and a previously stored milk flow characteristic.

The milk flow characteristics may comprise one or more measurements out of for example: flow increase rate, flow increase time, peak flow level, peak flow duration, flow decrease rate, flow decrease time and milk yield, milk temperature, levels of progesterone, LDH, BHB, and/ or somatic cell count.

In some preferred embodiments, the milk flow characteristics that are stored in the database 130 may be continuously updated, or at least updated at regular time intervals. In some embodiments, an average over a number of measurements of the milk flow characteristics of each animal may be calculated, such as e.g. the latest five days, the latest seven days, etc. The milk flow information may be normalised, in case it is obtained from different milking equipment, having slightly different, known, calibration. The milk flow information, e.g. an averaged number of milk flow sequences from an animal, may be filtered, e.g. using a low pass filter, such that fast variations in the milk flow are smoothed. Thereby, a data base entry, or other type of reference information may be created, which is unique for each indi vidual animal at the farm; although it may not necessarily be globally unique.

Further, a distinction may be made between a first milking event of the day and a subsequent milking event of the day. The reason for this is that the milk flow characteristics of an animal may look slightly different for the different milking sessions during the day, and a more pre cise match can be made if the different milking events of the day are treated separately. For example, in many farms, animals may be milked twice a day, once in the morning, and once in the evening. In such cases, a distinction may be made between milking events made in the morning and milking events made in the evening. The milk flow characteristics of milking events made in the morning and milking events made in the evening of the respective ani mals in the herd may be handled separately. For example, milk flow characteristics deter mined in a morning milking may then be compared with previously stored milk flow charac teristics from morning milking events. Correspondingly, milk flow characteristics determined in an evening milking may be compared with previously stored milk flow characteristics from evening milking events.

In farms where animals (at least to some extent) themselves decide when to visit a milking station, some animals are milked more than twice a day, which is beneficial for their milk production. Animals may be given a new milking permission e.g. every eight hours, every ten hours, etc. In such cases, a distinction can be made between the first milking event of the day and any subsequent milking event of the day. For example, the first, the second and the third milking event of the day may be stored and treated separately for identification purposes.

In an embodiment, each average value of a number (such as e.g. 5, 7, 10, etc.) of measure ments of the milk flow characteristics of each animal may be built on some days’ average values calculated separately on morning and evening sessions.

In some milking installations, the milking duration may be set to be a constant value. During such circumstances the milking duration may not be used for identification purposes, since it then is not related to the characteristics of the animal 100a; i.e. the milking duration will be the same for all animals at the farm, as determined by the farmer.

Thanks to the described solution, individual animals 100a at the farm may be recognised merely on their milk flow characteristics, as determined during milking. Also, or alternatively, the described solution may be used as a support system for another, primarily used identifi cation system (e.g. based on RFID tags, photo identification, etc.). In such cases, the de scribed solution could be used for correcting a wrongly identified animal, or providing identi fication for a small set of animals which have not been identified by the primary identification system. The described solution can also provide suggestions to a herd management system concerning a probable or possible identity of the animal 100a being milked. The described solution may be applied as a support to e.g. rotary, parlour, and/ or stanchion barn custom ers.

Figure 2 schematically illustrates animals 100a, 100b, 100c, 100d of a herd 101 at a farm. In this illustration, some examples of state of the art automatic animal identification are illus trated. Each animal 100a, 100b, 100c, 100d may have a respective identification device 210 attached to the body in some embodiments, e.g. in a necklace around the neck of the animal 100, under the hide of the animal 100a, 100b, 100c, 100d, as ear tag/-s, around the tail of the animal 100a, 100b, 100c, 10Od and/ or around any, some or all of the legs of the animal 100a, 100b, 100c, 100d, etc.

The identification device 210 may comprise a transponder in some embodiments. The tran sponder of the identification device 210 may be recognised by a reader 215. The communi cation between the identification device 210 and the reader 215 may be made e.g. over a short range wireless signal such as Radio-Frequency Identification (RFID), Bluetooth, ZigBee, infrared transmission, Wi-Fi, Light Fidelity (Li-Fi), ANT+, Near Field Communication (NFC), Z-Wave, Wireless Universal Serial Bus (Wireless USB), etc.

The reader 215 may then provide an identity of the animal 100a, 100b, 100c, 100d to the control unit 120, over a wired or wireless communication interface.

Also, or alternatively, a camera 230 may capture an image of the animal 100a, 100b, 100c, 100d and automatically derive an estimate of the animal identity based on an image of at least a part of the animal 100a, 100b, 100c, 100d. In some embodiments, the camera 230 may also determine a time period and associate a time stamp with the determined animal identity. The camera 230 may comprise e.g. a camera, a video camera, a stereo camera, a time-of flight camera or similar entity.

In order to capture the image of the animal 100a, 100b, 100c, 100d, the camera 230 may be mounted in a ceiling of the barn, or on an elevated structure in the barn, in order to capture images from behind/ above and/ or front of the animal 100a, 100b, 100c, 100d. The camera 230 and/ or the reader 215 may be situated at a location in the barn where the animal 100a, 100b, 100c, 100d may be identified, without being mistaken for another animal 100a, 100b, 100c, 100d, e.g. at a milking station or the like, when the animal identity is associated with the individual milk flow characteristics determined during milking.

In some embodiments, the identification device 210 may comprise an identification number, which may be recognised by the camera 230, or possibly a separate camera, in cooperation with an image recognition program. In some further embodiments, the identification device 210 may comprise the identification number encoded in a graphic encoding such as e.g. barcode, European Article Number (EAN) code, data matrix, Quick Response (QR) code, which may be printed on a tag associated with each respective animal 100a, 100b, 100c, 100d, painted and/ or tattooed into the hide of the animal 100a, 100b, 100c, 100d. The reader 215 may then comprise a scanner in some embodiments. Alternatively, the camera 230 may recognise the graphic encoding in cooperation with the image recognition program.

In some alternative embodiments, the animals 100a, 100b, 100c, 100d may be identified by iris scanning, which may be performed by the reader 215.

In yet some embodiments, the animal 100a, 100b, 100c, 100d may be identified based on the pattern of markings in the animal skin, as determined by the camera 230 and a compar ison made with a register over pre-stored patterns of animals 100a, 100b, 100c, 100d in the herd 101.

The control unit 120 may repeatedly obtain information from various sources and sensors in the barn wherein the animals 100a, 100b, 100c, 100d are kept, including the identification device 210, the reader 215 and/ or the camera 230, possibly together with a time stamp. It thereby becomes easy to match animal identification obtained from the reader 215, with im ages of the animals 100a, 100b, 100c, 100d, captured by the camera 230. This information may be stored in the database 130. Various measured data associated with the animal 100a, 100b, 100c, 100d, and possibly all animals 100a, 100b, 100c, 100d of the herd 101 may thus be continuously stored, e.g. associated with a time stamp in the database 130.

Further, the control unit 120 may be connected to a transceiver 225, configured to transmit and receive wireless signals to/ from the identification device 210, the reader 215, and/ or the camera 230. Thereby, information concerning the animal identity and/ or milk flow char acteristics of the animals 100a, 100b, 100c, 100d may be obtained by the control unit 120.

Such wireless communication interface may function according to any suitable state-of-the art wireless technology, such as Wi-Fi, Wireless Local Area Network (WLAN), Bluetooth (BT) or other similar wireless technology, e.g. as previously enumerated.

The communication between the reader 215, the camera 230 and/ or the sensor 115 and the control unit 120 may alternatively be made over a wired connection.

The control unit 120 may be comprised in a portable entity such as a mobile cellular tele phone, a portable computing device, a pair of intelligent glasses, an augmented reality de vice, a smart watch or similar device having a user interface and wireless communication ability.

The control unit 120 may be carried by a human such as e.g. a farmer or other person work ing at a farm. The“farm” as herein used may be a barn, a ranch, a stable or other similar agricultural structure or area (indoors or outdoors) for keeping animals 100a, 100b, 100c, 100d.

In some embodiments, the functionalities of the control unit 120 may be comprised in an application software (app) and downloaded to the portable entity of the farmer.

Figure 3 illustrates an example of a milk flow graph for a certain animal 100a, 100b, 100c, 100d.

On a horizontal axis, time since commencement of the milking is shown, while the vertical axis illustrates milk flow in Kg/ minute, during the milking. The total milk yield during the milking may be regarded as the size of the area beneath the milk flow curve, i.e. the integral of the milk flow curve.

The total milk yield of each respective animal 100a, 100b, 100c, 100d is typically very inter esting for the farmer to notice, for economic reasons and also for detecting sudden devia tions, e.g. due to illness, insufficient fodder (a low rank animal may not be allowed to access the fodder table/ feed station), etc.

The milk flow curve, or rather the milk flow sequence which the curve represents, comprises three particularly interesting phases or segments. The first segment of interest 301 com prises a flow increase. The flow increase 301 may be defined as the period from when the so-called alveoli-milk (i.e. not the so-called cistern milk) starts to flow, up to a second seg ment 302, a peak phase. The flow increase 301 may preferably be defined as/ by the flow increase rate, calculated e.g. as the derivative of the milk flow curve during the flow increase 301. The flow increase could alternatively or in addition be defined as/ by the time period required for the milk flow increase.

The peak phase 302 may be defined as a milk flow level at the top flow of the animal 100a, 100b, 100c, 100d, or e.g. within a percentage of the peak flow level, such as e.g. 5%, 10% etc., which may also be selected in dependence on whether the milk flow information is low pass filtered or not. The peak phase 302 may be defined as a peak flow duration, i.e. the length in time wherein the milk flow of the animal 100a, 100b, 100c, 100d is maintained with the peak phase 302.

For some animals 100a, 100b, 100c, 100d, the peak phase may have different durations for different teats. For example, the rear teats may have a much longer peak period than the front teats. When measuring or looking at a cluster level (milk from all teats added together) this may appear e.g. as two different peak levels, first one with milk from all teats, and then one with milk mostly from the rear teats. When the milk flow of the animal 100a, 100b, 100c, 100d decreases below the milk flow level of the peak phase 302, a transition is made to a third segment 303 comprising a flow decrease. The flow decrease 303 defines a decrease in milk flow, from the peak phase 302, to a second phase. The flow decrease 303 may be defined as a flow decrease rate, i.e. the derivative of the milk flow curve during the flow decrease 303, and/ or the duration of the flow decrease, i.e. the flow decrease time.

After the third segment, a post milking period terminates the milking. The length of the post milking period is configurable by the farmer. Some farmers want to extract as much milk as possible from each animal 100a, 100b, 100c, 100d by applying an extended post milking period while other farmers prefer to keep the post milking period as brief as possible in order to keep the total milking session of all the herd 101 as short as possible, and a compromise between these extremes.

Figure 4 illustrates an example of a method 400 according to an embodiment. The flow chart in Figure 4 shows the method 400 executed in a control unit 120 for animal identification based on an individual milk flow characteristic, determined during milking of the animal 100a, 100b, 100c, 100d.

The method 400 may be performed by using a system and a computing device on the farm in some embodiments. However, the method 400 may alternatively be performed remotely, by providing measurement data to a remote computing device such as a server which may perform computations and store the measurement data in a database, in some embodi ments.

In order to determine the identification of the animal 100a, 100b, 100c, 100d, the method 400 may comprise a number of method steps 401 -405. However, some of these method steps 401-405 may be performed solely in some alternative embodiments, like e.g. method step 401 and/ or method step 404. Further, the described method steps 401-405 may be performed in a somewhat different chronological order than the numbering suggests. The method 400 may comprise the subsequent steps: Step 401 , which is an optional part of the method 400, may be performed/regarded as part of the method 400 in some embodiments, comprises identifying (with some alternative method) an animal 100a, 100b, 100c, 100d in a herd 101 at a farm and associating an identity reference of the animal 100a, 100b, 100c, 10Od with information related to a milk flow derived during milking of the animal 100a, 100b, 100c, 100d.

The identity reference may be e.g. a unique (at least within the herd 101) number, code or name, symbol or similar identity reference.

The identification of the animal 100a, 100b, 100c, 100d may be made manually by the farmer in some embodiments; or automatically by using e.g. any of the previously herein described methods for automatic identification of animals 100a, 100b, 100c, 100d.

The identification of the animal 100a, 100b, 100c, 100d may be made during an initialisation process, during which a reference value is established, associated with the identity of the animal 100a, 100b, 100c, 100d.

Step 402 comprises obtaining information related to a milk flow when milking the animal 100a, 100b, 100c, 100d.

The obtained information may comprise milk flow (in weight or volume) per time unit such as e.g. Kg milk per minute, or milk flow per time unit measured in some other measurement unit, as determined by the sensor 115 at the milk line 1 10.

The obtained information may additionally comprise information of a characteristic, or several characteristics, of the milk derived during the milking, such as milk temperature, levels of progesterone, LDH, BHB, and/ or somatic cell count.

The derived characteristic may be determined by the sensor 115 and provided to the control unit 120.

Step 403 comprises determining a first characteristic of the milk flow based on the obtained 402 information.

The first characteristic of the milk flow may comprise at least one of: flow increase rate, flow increase time, peak flow level, peak flow duration, flow decrease rate, flow decrease time, and/or milk yield. The herein used expression“first characteristic of the milk flow” may thus rather comprise a plurality of distinct characteristics of the milk flow, in some embodiments. An advantage by measuring a plurality of distinct characteristics of the milk flow is that the identity of the animal 100a, 100b, 100c, 100d may be determined with increased certainty.

The first characteristic of the milk flow may in particular, in some embodiments, comprise the flow increase rate and/ or flow increase time, as determined from commencement of the milking, to the peak phase 302. An advantage therewith is that an instant identification of the animal 100a, 100b, 100c, 100d may be made, thereby speeding up the identity process and making it possible to identify the animal 100a, 100b, 100c, 100d already during the ongoing milking session.

Step 404 which is an optional part of the method 400, may be performed/ regarded as part of the method 400 in some embodiments, comprises determining whether the milking is per formed during morning or evening.

The comparison of the determined 403 first characteristic of the milk flow may in some em bodiments be made with previously determined milk flow characteristics determined during the corresponding milkings at either only morning, or only evening, or more generically ex pressed, either only a first milking of the day, or only any subsequent milking of the day.

In some embodiments, a more detailed time division may be made such as e.g. every hour, every second hour, every four-hour time segment, etc.

An advantage with establishing reference values for milking events performed during differ ent time segments is that identification of the animal 100a, 100b, 100c, 10Od may be made with a further increased precision.

Step 405 comprises identifying the animal 100a, 100b, 100c, 100d, based on said deter mined 403 first characteristic of the milk flow.

The identification of the animal 100a, 100b, 100c, 100d may in some embodiments comprise comparing the determined 403 first characteristic of the milk flow with previously determined individual milk flow characteristics being related to individual animals 100a, 100b, 100c, 100d at one or more previous milking events, each previously determined individual milk flow char acteristic being associated with an animal identity. Furthermore, the identification of the ani mal 100a, 100b, 100c, 100d may comprise extracting the animal identity, associated with the previously determined milk flow characteristic having the highest correlation with the deter mined 403 first characteristic.

The correlation may be determined e.g. by a correlation function, a covariance function, etc.

The previously determined individual milk flow characteristics may be based on a mean value over a plurality of milkings. The previously determined individual milk flow characteristics may be determined based on previous milking events within about seven days (such as e.g. three days, or eleven days) preceding the milkings from which information related to the milk flow has been obtained 402.

An advantage by calculating the mean value over a plurality of previously made milkings, e.g. the latest made milking sessions, is that a relevant reference value is achieved which could be used also in case the outcome (in volume of milk) differ over time, depending on when (in the lactation period, which lasts approximately 305 days) the animal 100a, 100b, 100c, 100d currently is situated. Further, in case data is saved between lactation periods, and is used as reference data for a corresponding period in a next lactation, it should be noticed that an animal 100a, 100b, 100c, 100d, such as a cow, may produce more milk for each lactation during e.g. the first three lactations.

In embodiments wherein information of the characteristic of the milk derived during the milk ings has been obtained, such as milk temperature, levels of progesterone, LDH, BHB, and/ or somatic cell count, the identification of the animal 100a, 100b, 100c, 100d may be further based on said determined 403 characteristic of the milk.

An advantage with measuring one or several additional characteristics is that the identity of the animal 100a, 100b, 100c, 100d may be determined with an increased distinction.

The method 400 may further in some embodiments comprise storing the obtained infor mation related to the milk flow in association with the identity reference of the identified ani mal 100a, 100b, 100c, 100d.

In some cases, e.g. when the animal 100a, 100b, 100c, 100d is ill, and/ or has not been given enough fodder and/ or water etc. The measured characteristics of that animal 100a, 100b, 100c, 100d may then not match the previously stored reference level. Instead, the identity of the deviating animal 100a, 100b, 100c, 100d will need to determined, at least partly, in some other manner, e.g. it may be determined by method of exclusion. Further, in some embodiments, the deviating results may trigger transmission of an alert to the farmer, encouraging him/ her to check the status of the animal 100a, 100b, 100c, 100d.

Figure 5 illustrates an embodiment of a system 500 for identifying an animal 100a, 100b, 100c, 100d.

The system 400 comprises a control unit 120. The control unit 120 is configured to perform at least some of the previously described method steps 401-405 according to the method 400 described above and illustrated in Figure 4. The control unit 120 aims at animal identifi cation, based on an individual milk flow characteristic of an animal 100a, 100b, 100c, 100d. The control unit 120 is configured to obtain information related to a milk flow during milking of the animal 100a, 100b, 100c, 100d. Further, the control unit 120 is additionally configured to determine a first characteristic of the milk flow based on the obtained information. In addi tion, the control unit 120 is configured to identify the animal 100a, 100b, 100c, 100d, based on said determined first characteristic of the milk flow.

In some embodiments, the control unit 120 may be further configured to identify the animal 100a, 100b, 100c, 100d by comparing the determined first characteristic of the milk flow with a previously determined milk flow characteristics being related to individual animals 100a, 100b, 100c, 100d at one or more previous milking events, each previously determined indi vidual milk flow characteristic being associated with an animal identity reference. Further, the control unit 120 may be configured to identify the animal 100a, 100b, 100c, 100d by also extracting the animal identity reference, associated with the previously determined milk flow characteristic having the highest correlation with the determined first characteristic.

The control unit 120 may be additionally configured to determine the identity of the animal 100a, 100b, 100c, 100d by a comparison with the previously determined individual milk flow characteristics, based on a mean value over a plurality of milkings, such as three, five, eight, ten, etc.

In some embodiments, the control unit 120 may be configured to determine the previously determined individual milk flow characteristics, based on previous milking events performed within seven days preceding the milking from which information related to the milk flow has been obtained.

Furthermore, the control unit 120 may be configured to associate the identity of the animal 100a, 100b, 100c, 100d with information related to the milk flow derived during milking of the animal 100a, 100b, 100c, 100d.

The control unit 120 may be configured to determine whether the milking is performed during a first milking event of the day or during a subsequent milking event of the day, e.g. morning or evening. In addition, the control unit 120 may be configured to compare the determined first characteristic of the milk flow with previously determined milk flow characteristics deter mined during the corresponding milkings at either a first milking event of the day or at a subsequent milking event of the day, e.g. only morning, or only evening.

In addition, the control unit 120 may be configured to determine identity of the animal 100a, 100b, 100c, 100d in a context wherein the first characteristic of the milk flow comprises at least one of: flow increase rate, flow increase time, peak flow level, peak flow duration, flow decrease rate, flow decrease time and milk yield. The first characteristic of the milk flow may in some embodiments comprise the flow increase rate and/ or flow increase time.

The control unit 120 may be configured to obtain information of the characteristic of the milk derived during the milking. Also, the control unit 120 may be further configured to identify the animal 100a, 100b, 100c, 100d, further based on said determined characteristic of the milk.

Furthermore, the control unit 120 may be configured to determine identity of the animal 100a, 100b, 100c, 100d in a context wherein the characteristic of the milk comprises any, some or all of conductivity, amount of fat, milk temperature, and/ or somatic cell count.

The control unit 120 may be configured to store the obtained information related to the milk flow in association with the identity of the identified animal 100a, 100b, 100c, 100d.

The system 500 furthermore comprises a database 130 configured to store information re lated to the respective milk flow of animals 100a, 100b, 100c, 100d, which information is associated with the identity reference of the respective animal 100a, 100b, 100c, 100d.

In some alternative embodiments, the system 500 furthermore may comprise an identifica tion unit 210 comprising a unique identification of the animal 100a, 100b, 100c, 100d. The identification unit 210 may e.g. comprise a transponder such as RFID, carried by the animal 100a, 100b, 100c, 100d in some embodiments.

The system 500 may also comprise a receiver 215, configured to detect and obtain wireless signals from the identification device 210 of the animal 100a, 100b, 100c, 100d. The reader 215 may then provide an identity of the animal 100a, 100b, 100c, 100d to the control unit 120, over a wired or wireless communication interface.

The system 500 may alternatively, or additionally, comprise a camera 230, configured for capturing an image of the animal 100a, 100b, 100c, 100d and automatically derive an esti mate of the animal identity based on an image of at least a part of the animal 100a, 100b, 100c, 100d. The camera 230 may comprise e.g. a camera, a video camera, a stereo camera, a time-of flight camera or similar entity.

In further addition, the system 500 also may comprise a transceiver 225, configured to trans mit and receive wireless signals to/ from the identification device 210, the reader 215, and/ or the camera 230. Thereby, information concerning the animal identity and/ or milk flow characteristics of the animals 100a, 100b, 100c, 100d may be obtained by the control unit 120, in some embodiments.

The system 500 may also comprise a communication unit, configured to output information concerning the milking and/ or animal identification, to the farmer, such as e.g. a cellular telephone or similar communication device.

The control unit 120 may comprise a receiver 510 configured to receive information from the transceiver 225, from the identification unit 1 10, from the reader 1 15, from the camera 130 and/ or from the database 130.

The control unit 120 also comprises a processing circuit 520 configured for performing vari ous calculations for conducting the method 400 according to at least some of the previously described method steps 401-405.

Such processing circuit 520 may comprise one or more instances of a processing circuit, i.e. a Central Processing Unit (CPU), a processing unit, a processing circuit, a processor, an Application Specific Integrated Circuit (ASIC), a microprocessor, or other processing logic that may interpret and execute instructions. The herein utilised expression“processing cir cuit” may thus represent a processing circuitry comprising a plurality of processing circuits, such as, e.g., any, some or all of the ones enumerated above.

Furthermore, the control unit 120 may comprise a memory 525 in some embodiments. The optional memory 525 may comprise a physical device utilised to store data or programs, i.e., sequences of instructions, on a temporary or permanent basis. According to some embodi ments, the memory 525 may comprise integrated circuits comprising silicon-based transis tors. The memory 525 may comprise e.g. a memory card, a flash memory, a USB memory, a hard disc, or another similar volatile or non-volatile storage unit for storing data such as e.g. ROM (Read-Only Memory), PROM (Programmable Read-Only Memory), EPROM (Erasable PROM), EEPROM (Electrically Erasable PROM), etc. in different embodiments.

Further, the control unit 120 may comprise a signal transmitter 530. The signal transmitter 530 may be configured for transmitting signals via a wired or wireless communication inter face to the transceiver 225 and/ or the database 130.

However, in some alternative embodiments, the system 500 may comprise additional units for performing the method 400 according to method steps 401-405.

The above described method steps 401-405 to be performed in the control unit 120 may be implemented through the one or more processing circuits 520 within the control unit 120, together with a computer program for performing at least some of the functions of the method steps 401-405. Thus, the computer program comprises instructions which, when the com puter program is executed by the control unit 120 in the system 500, cause the control unit 120 to carry out the method 400 according to at least some of the method steps 401-405.

The computer program mentioned above may be provided for instance in the form of a com puter-readable medium, i.e. a data carrier carrying computer program code for performing at least some of the method steps 401-405 according to some embodiments when being loaded into the one or more processing circuits 520 of the control unit 120. The data carrier may be, e.g., a hard disk, a CD ROM disc, a memory stick, an optical storage device, a magnetic storage device or any other appropriate medium such as a disk or tape that may hold ma chine readable data in a non-transitory manner. The computer program may furthermore be provided as computer program code on a server and downloaded to the control unit 120 remotely, e.g. over an Internet or an intranet connection.

The terminology used in the description of the embodiments as illustrated in the accompa nying drawings is not intended to be limiting of the described method 400; the control unit 120; the computer program; the system 500 and/ or the computer-readable medium. Various changes, substitutions and/ or alterations may be made, without departing from invention embodiments as defined by the appended claims. As used herein, the term "and/ or" comprises any and all combinations of one or more of the associated listed items. The term“or” as used herein, is to be interpreted as a mathematical OR, i.e. , as an inclusive disjunction; not as a mathematical exclusive OR (XOR), unless ex pressly stated otherwise. In addition, the singular forms "a", "an" and "the" are to be inter- preted as“at least one”, thus also possibly comprising a plurality of entities of the same kind, unless expressly stated otherwise. It will be further understood that the terms "includes", "comprises", "including" and/ or "comprising", specifies the presence of stated features, ac tions, integers, steps, operations, elements, and/ or components, but do not preclude the presence or addition of one or more other features, actions, integers, steps, operations, ele- ments, components, and/ or groups thereof. A single unit such as e.g. a processor may fulfil the functions of several items recited in the claims. The mere fact that certain measures or features are recited in mutually different dependent claims, illustrated in different figures or discussed in conjunction with different embodiments does not indicate that a combination of these measures or features cannot be used to advantage. A computer program may be stored/ distributed on a suitable medium, such as an optical storage medium or a solid-state medium supplied together with or as part of other hardware, but may also be distributed in other forms such as via Internet or other wired or wireless communication system.

Claims

PATENT CLAIMS

1. A method (400) for animal identification, based on an individual milk flow character istic, wherein the method (400) comprises:

-obtaining (402) information related to a milk flow when milking an animal (100a, 100b, 100c, 100d);

-determining (403) a first characteristic of the milk flow based on the obtained (402) information; and

-identifying (405) the animal (100a, 100b, 100c, 100d), based on said determined (403) first characteristic of the milk flow.

2. The method (400) according to claim 1 , wherein the first characteristic of the milk flow comprises at least one of: flow increase rate, flow increase time, peak flow level, peak flow duration, flow decrease rate, flow decrease time and milk yield.

3. The method (400) according to any one of claims 1-2, wherein the first characteristic of the milk flow comprises the flow increase rate and/ or flow increase time.

4. The method (400) according to any one of claims 1-3, further comprising: obtaining information of a characteristic of the milk derived during the milkings; and wherein the identification (405) of the animal (100a, 100b, 100c, 100d) is further based on said determined (403) characteristic of the milk.

5. The method (400) according to claim 4, wherein the characteristic of the milk com prises one of: milk temperature, levels of progesterone, LDH“Lactate Dehydrogen ase”, BHB“Beta-Hydroxybutyrat”, and/ or somatic cell count.

6. The method (400) according any of the claims 1-5, wherein the identification (405) of the animal (100a, 100b, 100c, 100d) comprises:

-comparing the determined (403) first characteristic of the milk flow with previously determined individual milk flow characteristics being related to individual animals (100a, 100b, 100c, 100d) at one or more previous milking events, each previously determined individual milk flow characteristic being associated with an animal identity; and

-extracting the animal identity, associated with the previously determined milk flow characteristic having the highest correlation with the determined (403) first charac teristic.

7. The method (400) according to claim 6, further comprising:

-determining (404) whether the milking is a first or a subsequent milking of the day of the animal (100a, 100b, 100c, 100d); and

wherein the comparison of the determined (403) first characteristic of the milk flow is made with previously determined milk flow characteristics determined during the corresponding milkings at either a first milking of the day, or a subsequent milking of the day.

8. The method (400) according to any of claims 6-7, wherein the previously determined individual milk flow characteristics are based on a mean value over a plurality of milkings.

9. The method (400) according to any one of claim 6-8, wherein the previously deter- mined individual milk flow characteristics are determined based on previous milking events within seven days preceding the milkings from which information related to the milk flow has been obtained (402).

10. The method (400) according to any one of claims 6-9, further comprising a preced- ing step of, for a number of milkings:

-associating (401) the identity of an animal (100a, 100b, 100c, 100d) with infor mation related to a milk flow derived during milking of the animal (100a, 100b,

100c, 100d).

11. The method (400) according to any one of the preceding claims, further comprising:

-storing the obtained information related to a milk flow in association with an identity of the identified animal (100a, 100b, 100c, 100d).

12. A control unit (120) for animal identification, based on an individual milk flow char- acteristic; wherein the control unit (120) is configured to:

-obtain information related to a milk flow during milking of an animal (100a, 100b, 100c, 100d);

-determine a first characteristic of the milk flow based on the obtained information; and

-identify the animal (100a, 100b, 100c, 100d), based on said determined first char acteristic of the milk flow.

13. The control unit (120) according to claim 12, wherein the first characteristic of the milk flow comprises at least one of: flow increase rate, flow increase time, peak flow level, peak flow duration, flow decrease rate, flow decrease time and milk yield.

14. The control unit (120) according to any one of claims 12-13, wherein the first char acteristic of the milk flow comprises the flow increase rate and/ or flow increase time.

15. The control unit (120) according to any one of claims 12-14, further configured to: obtain information of a characteristic of the milk derived during the milking; and wherein the identification of the animal (100a, 100b, 100c, 100d) is further based on said determined characteristic of the milk.

16. The control unit (120) according to claim 15, wherein the characteristic of the milk comprises: milk temperature, levels of progesterone, LDH“Lactate Dehydrogen ase”, BHB“Beta-Hydroxybutyrat” and/ or somatic cell count.

17. The control unit (120) according to any of claims 12-16, further configured to iden tify the animal (100a, 100b, 100c, 100d) by:

-comparing the determined first characteristic of the milk flow with a previously de termined milk flow characteristics being related to individual animals (100a, 100b, 100c, 100d) at one or more previous milking events, each previously determined individual milk flow characteristic being associated with an animal identity; and; -extracting the animal identity, associated with the previously determined milk flow characteristic having the highest correlation with the determined first characteris tic.

18. The control unit (120) according to claim 17, further configured to:

-determine whether the milking is performed during a first or a subsequent milking of the day; and

-wherein the comparison of the determined first characteristic of the milk flow is made with previously determined milk flow characteristics determined during the corresponding milkings at either a first milking of the day, or a subsequent milking of the day.

19. The control unit (120) according to any of claims 17-18, wherein the previously de- termined individual milk flow characteristics are based on a mean value over a plu rality of milkings.

20. The control unit (120) according to any one of claims 17-19, further configured to determine the previously determined individual milk flow characteristics, based on previous milking events performed within seven days preceding the milking from which information related to the milk flow has been obtained.

21. The control unit (120) according to any one of claims 12-20, further configured to:

-associate the identity of an animal (100a, 100b, 100c, 100d) with information re lated to a milk flow derived during milking of the animal (100a, 100b, 100c, 100d).

22. The control unit (120) according to any one of claims 12-21 , further configured to:

-store the obtained information related to a milk flow in association with an identity of the identified animal (100a, 100b, 100c, 100d).

23. A system (500) for identifying an animal (100a, 100b, 100c, 100d); wherein the sys tem (500) comprises:

a control unit (120) according to any of claims 12-22; and

a database (130) configured to store information related to a respective milk flow of animals (100a, 100b, 100c, 100d), which information is associated with an identity reference of the respective animal (100a, 100b, 100c, 100d).