WO2011039112A2

WO2011039112A2 - Animal monitoring

Info

Publication number: WO2011039112A2
Application number: PCT/EP2010/064135
Authority: WO
Inventors: Bohao Liao
Original assignee: DeLaval Holding AB
Current assignee: DeLaval Holding AB
Priority date: 2009-10-01
Filing date: 2010-09-24
Publication date: 2011-04-07
Anticipated expiration: 2012-04-01
Also published as: WO2011039112A3

Abstract

The behavior of livestock animals (A1, A2, A3, A4) located within a predefined area (100) is monitored. To this aim each animal (A1, A2, A3, A4) has a respective identification means (T1, T2, T3, T4) uniquely identifying the animal. Image registering means (C1, C2, C3, C4) repeatedly record image data (d1, d2, d3, d4) representing the animals (A1, A2, A3, A4). A data processing means (120) receives the image data (d1, d2, d3, d4), and based thereon derives a movement pattern (MP) for each animal (A1, A2, A3, A4). Moreover, at least one identification station (110, 111) registers an animal identity (ID1) in response to the passing of one of said identification means (T1) within an operating zone of the identification station (110, 111). The identification station (110, 111) is distinct from the at least one image registering means (C1, C2, C3, C4).

Description

ANIMAL MONITORING

THE BACKGROUND OF THE INVENTION AND PRIOR ART

The present invention relates generally to monitoring and analysis of animal activity. More particularly the invention relates to an arrangement according to the preamble of claim 1 and a method according to the preamble of claim 1 1. The invention also concerns a computer program according to claim 21 and a computer readable medium according to claim 22.

In modern herd management it is important to keep track of the movements and behavior of a given group of livestock animals. For instance, based on parameters like walking time, lying down time, drinking time and eating time it is possible to draw conclusions regarding the animals' health status, their oestrus phases etc. There is yet no example of an automatic monitoring/tracking system for herd management. However, solutions are known, which are primarily intended for tracking and analyzing the behavior of smaller animals in a laboratory environment.

JP 1994-0189766 describes a setup for anafyzing the behavior of experimental animals, whereby the animals are provided with respective color marks. The animals are photographed by a color video camera. A color tracker samples a predefined color part of the video signal and calculates a centroid position for each animal at every 0, 1 seconds. Corresponding coordinates are sent to a computer for analysis.

US 2008/0152192 discloses a system for 3D monitoring and analysis of motion-related behavior of test subjects. Here, a cata- dioptric stereo computer vision system registers video data. Based on this data targets are identified and tracked. However, even though sophisticated tracking software were used it would be comp!icated to maintain an accurate tracking of all the animals in a herd of livestock animals in their typical environment. For example, varying light conditions and the fact that other subjects and objects occasionally obstruct the cameras' view of a given subject, the system may loose track of one or more subjects. Automatic re-identification of the subject is then complex to accomplish, if at all possible.

SUMMARY OF THE INVENTION

The present invention aims at alleviating the above problems, and thus offer an efficient and robust solution for automatic monitoring of the behavior of livestock animais.

According to one aspect of the invention, this object is achieved by the initially described arrangement, wherein the arrangement includes at least one identification station that is distinct from the at least one image registering means. The at least one identification station is configured to register an animal identity in response to the passing of one of said identification means within an operating zone of the identification station. This arrangement is advantageous because thereby accurate identity records can be maintained for all animals even if the tracking of one or more animals is temporarily lost, e.g. due to camera obstruction or poor lighting.

According to one preferred embodiment of this aspect of the in- vention, the movement pattern includes a series of data entries and the processing means is configured to trace the animals within the predefined area based on the series of data entries, where each entry comprises a coordinate reflecting an animal position within the predefined area, and a time stamp indicating a point in time when the image data was recorded based on which image data the coordinate was derived. Here, a trace for a given animal starts at one of the at least one identification sta- tion and extends to an animal position given by a coordinate of a latest data entry for the animal. Thus, the position for each animal is known at all times.

According to another preferred embodiment of this aspect of the invention, the arrangement includes at least two image registering means. At least one first image registering means of said at least two image registering means is configured to record image data representing a first sub-area of the predefined area. At least one second image registering means of said at least two image registering means is likewise configured to record image data representing the first sub-area. Moreover, the data processing means is configured to use the image data recorded by the at least one second image registering means as a complement to the image data recorded by the at least one first image regis- tering means for tracing said animals within the first sub-area. Thus, the responsibility for providing image data in respect of one or more animals can conveniently be handed over from a first to a second image registering means.

Using two or more image registering means in this manner pro- vides additional advantages. For instance, if a first image registering means becomes temporarily obstructed and/or its image data is deteriorated due to poor lighting, a second image registering means may function as a backup.

According to a further preferred embodiment of this aspect of the invention, the processing means is configured to update a trace for a given animal upon reentry of the animal's identification means within the operating zone of an identification station. Thereby, the tracking of that animal is refreshed, and it is guaranteed that the data is accurate. According to yet another preferred embodiment of this aspect of the invention, at least one identification station is arranged at an entry to the predefined area. Namely, it is desirable to identify each animal as early as possible, i.e. in connection with entering the area to be monitored.

According to still another preferred embodiment of this aspect of the invention, the processing means is configured to resolve an ambiguity between the traces of a first and a second animal as follows. In response to a reentry of the first animal's transponder means within the operating zone of an identification station, the processing means is configured to trace back the first animal to an ambiguity zone in which the recorded image data has proven to provide insufficient accuracy to distinguish the first animal from the second animal. Hence, even if one or more animal traces are temporarily lost, this uncertainty may be removed in a straightforward manner.

According to a further preferred embodiment of this aspect of the invention, the identification means includes a transponder, an RFID tag, a bar code tag, a color area and/or a given physical characteristics of the animal body is used for identifying each animal. Consequently, reliable animal identification is rendered possible.

According to another preferred embodiment of this aspect of the invention, the operating zone of each identification station has an extension in space relative to a typical size of each of said animals and a typical location for the identification means on the animal, such that it can be expected that only one identification means at the time may be located within the operating zone. Thereby, the risk becomes very small that an animal trace is associated with an erroneous identity.

According to yet another preferred embodiment of this aspect of the invention, at least one of image registering means includes a camera configured to register three-dimensional (3D) image da- ta in real time. The data processing means, in turn, is configured to derive the movement pattern based on the 3D image data, and each entry in the movement pattern further comprises height data reflecting an animal elevation above a ground surfa- ce of the predefined area. This is advantageous because the 3D data provides the basis for a more detailed analysis of the animals' behavior. Moreover, the 3D data improves the tracking accuracy and reduces the risks that different animal traces are lost and/or mixed up with one another.

According to still another preferred embodiment of this aspect of the invention, the movement pattern for each animal includes a current location, a distance covered during a given interval, a total lying down time during a given interval, a total number of climbing events during a given interval, a total time spent at a feeding station, a number of visits to a feeding station, a total time spent at a drinking station, and/or a number of visits to a drinking station. Hence, the arrangement produces highly valuable herd management data. According to another aspect of the invention, the object is achieved by the initially described method, wherein an animal identity is registered in response to the passing of one of said identification means within an operating zone of an identification station, which is distinct from the at least one image registering means. The advantages of this method, as well as the preferred embodiments thereof, are apparent from the discussion hereinabove with reference to the proposed arrangement.

According to a further aspect of the invention, the object is achieved by a computer program, which is loadable into the internal memory of a computer, and includes software for controlling the above proposed method when the program is run on a computer.

According to another aspect of the invention, the object is achieved by a computer readable medium, having a program recorded thereon, where the program is to control a computer to perform the above-proposed method.

Further advantages, advantageous features and applications of the present invention will be apparent from the following description and the dependent claims. BRIEF DESCRIPTION OF THE DRAWINGS

The invention is now to be explained more closely by means of preferred embodiments, which are disclosed as examples, and with reference to the attached drawings.

Figure 1 schematically shows a monitoring arrangement according to one embodiment of the invention;

Figure 2 illustrates, by means of a flow diagram, the general method according to the invention.

DESCRIPTION OF PREFERRED EMBODIMENTS OF THE IN- VENTION

We refer initially to Figure 1 , which schematically shows one embodiment of a proposed arrangement for monitoring the behavior of livestock animals, here represented by individual subjects A1 , A2, A3 and A4. Each animal A1 , A2, A3 and A4 has an identification means T1 , T2, T3 and T4 respectively, which uniquely identifies the animal in question. The arrangement includes at least one image registering means, here exemplified by C1 , C2, C3 and C4, a data processing means 120 and at least one identification station 1 10 and 1 1 1. Each image registering means C1 , C2, C3 and C4 is configured to repeatedly record image data d1 , d2, d3 and d4 respectively representing the animals A1 , A2, A3 and A4 when located within a predefined area 100. Typically, the area 100 is a part of, or an entire cowshed, barn, pasture or field. The area 100 may thus include at least one feeding station 130 and at least one drinking station 140.

The data processing means 120 is configured to receive the image data d1 , d2, d3 and d4 from the image registering means C1 , C2, C3 and C4, which in turn may include a respective con- ventional type of video camera, or a video camera for registering real time 3D image data. The latter may be represented by a so- called time-of-flight (TOF) camera. In short, a TOF camera inclu- des a light projection source and a two dimensional array of pixels, where each pixel is capable of returning time of flight information regarding the light from the light projection source as well as light intensity information. By utilizing the time of flight information, the image processing can be simplified. It is possible to define a limited depth of field within which image data shall be regarded, and thus the amount of data to process can be reduced relative to conventional 2D image processing. In any case, based on the image data d1 , d2, d3 and d4, the data processing means 120 is configured to derive a movement pattern MP for each of the animals A1 , A2, A3 and A4.

The identification stations 1 10 and 1 1 1 are distinct from the image registering means C1 , C2, C3 and C4. I.e. none of the identification stations 1 10 and 1 1 1 produces image data that is used by the processing means 120 as a basis for deriving the movement patterns MP. However, one or more of the identification stations 1 10 and 1 1 1 may indeed include image registering means configured to capture image data based upon which animals can be identified. Namely, each identification sta- tion 1 10 and 1 1 1 is configured to register an animal identity, say ID1 , in response to the passing an identification means, say T1 , within an operating zone of the station, say 10.

According to one preferred embodiment of the invention, at least one identification station 1 10 is arranged at an entry to the pre- defined area 100, so that the animals A1 , A2, A3 and A4 can be identified in connection with entering the area 100 to be monitored. However, it is also advantageous to arrange identification stations at the exit(s) from the area 100 (if different from the entry), at the feeding station 130 and/or at the drinking station 140.

According to preferred embodiments of the invention, the identification means T1 , T2, T3 and T4 includes a transponder, an RFID tag, a bar code tag and/or a color area. As an alternative, or a complement thereto, the identification means T1 , T2, T3 and T4 may be represented by a given physical characteristics of the animal body, such as the head shape, various body measures or coat markings/lines. Naturally, the capabilities of the identification station 1 10 and 1 1 are selected depending on the identification means used. Hence, besides image registering means the identification stations 1 1 0 and 1 1 1 may include radio transceivers, magnetic field sensors and/or bar code readers.

Preferably, the movement pattern MP includes a series of data entries (e.g. updated 10 times per second), where each entry contains a coordinate reflecting an animal position within the predefined area 100, and a time stamp indicating a point in time when the image data was recorded based on which image data the coordinate was derived. Thus, if the image data d1 , 62, 63 or d4 represents two-dimensional information, the movement pattern MP may include x-y pairs and a respective time stamp associated with each x-y pair, where each x-y pair designates a point in a Cartesian plane. Furthermore, the processing means 120 is preferably configured to trace the animals A1 , A2, A3 and A4 within the predefined area 100 based on the series of data entries. Here, each trace produced by the processing means 120 represents a given animal A1 , A2, A3 or A4, and the traces always start at one of the identification stations 1 10 and 1 1 1 . A trace then extends to a position given by the coordinate of a latest data entry for the animal in question, i.e. A1 , A2, A3 or A4.

Provided that 3D image data d1 , d2, d3 or d4 is available via at least one of the image registering means C 1 , C2, C3 and/or C4 the data processing means 120 is preferably configured to derive the movement pattern MP based on such 3D image data. As a result, the entries in the movement pattern MP will further include height data reflecting an animal elevation above a ground surface of the predefined area 100. Consequently, the movement pattern MP may include x-y-z triplets and a respective time stamp associated with each x-y-z triplet, where each x-y-z triplet designates a point in a Cartesian space. According to one preferred embodiment of the invention, the arrangement includes at least two image registering means C 1 , C2 , C3 and C4. At least one first image registering means C 1 of these i mage registering means is configured to record image data d 1 representing a first sub-area of the predefined area 1 00.

At least one second image registering means C2 is config ured to record image data 62 representi ng the first sub-area . Hence, in Fig ure 1 , the first sub-area may be given by the intersection of a first region R1 covered by the first image registering means C 1 and a second region R2 covered by the second image reg istering means C2. Here, the data processi ng means 120 is configured to use the image data d2 recorded by the second image registering means C2 as a complement to the image data d 1 recorded by the first image registeri ng means C 1 for tracing said animals A1 , A2, A3 and A4 within the fi rst sub-area. This enables a convenient handover of the responsibi lity for providing image data i n respect of one or more animals A1 , A2, A3 and/or A4 from the first to the second image registering means C 1 and C2 respectively. If necessary, the first and second image Regis- tering means C 1 and C2 may also function as a backup for one another i n case of temporary obstruction and/or deterioration of the image data d 1 or d2 due to poor lighti ng. Thus, different regions R1 , R2 , R3 and R4 covered by different image reg istering means C 1 _t C2, C3 and C4 respectively may overlap. In order to further enhance the reliability and robustness of the system, one or more identification stations 1 1 0 and 1 1 1 may function as bridges between said regions R 1 , R2, R3 and R4. Thereby, the identification station may refresh any animal traces becoming temporarily lost when handing over the responsibility for providing image data from one the image registering means to another.

According to one preferred embodiment of the invention, the processing means 120 is configured to produce the movement pattern MP for an animal A1 , A2, A3 or A4, such that in addition to a current location one or more of the following pieces of data is included: a distance covered by the animal during a given interval, a total lying down time for the animal during a given interval, a total number of climbing events for the animal during a given in- terval, a total time that the animal has spent at a feeding station, e.g. 130, a number of visits that the animal has made to a feeding station, e.g. 130, a total time that the animal has spent at a drinking station, e.g. 140, and a number of visits that the animal has made to a drinking station, e.g. 140. According to one preferred embodiment of the invention, for robustness and data quality, the processing means 120 is configured to update a trace for a given animal upon reentry of the animal's identification means within the operating zone of an identification station. This means that if for example the animal A1 approaches the identification station 1 10 in such a manner that its identification means T1 becomes located within the operating zone of the identification station 1 10, the processing means 120 will update the trace in respect of the animal A1 , and thus confirm this animal's identity ID1 to be located at the iden- tification station 1 10 at a particular point in time.

Preferably, the respective operating zone of the identification stations 1 10 and 1 1 1 has an extension in space relative to a typical size of each of the animals A1 , A2, A3 and A4, and a typical location for the identification means T1 , T2, T3 and T4 on the animal, such that it can be expected that only one identification means at the time may be located within the operating zone. Namely, thereby the risk that the animal identities are confused with one another can be held comparatively low.

To further reduce the risk of data errors, according to one pre- ferred embodiment of the invention, the processing means 120 is configured to resolve an ambiguity between the traces of two animals, say A3 and A4, whose traces have been mixed up. Let us assume that this mix-up occurred in an ambiguity zone AZ. The ambiguity zone AZ is a place where the recorded image da- ta has proven to provide insufficient accuracy to distinguish a first animal A3 from a second animal A4. The processing means 120 is configured to resolve the ambiguity as follows. In response to a reentry of the first A3 animal's transponder means T3 wi- thin the operating zone of an identification station 1 10, the processing means 120 traces back the first animal A3 to the ambiguity zone AZ. Thus, by exclusion, the other trace ending in the ambiguity zone AZ must belong to the second animal A4, and consequently also this animal's trace can be confirmed by the processing means 120. This approach is desirable because it enhances the robustness of the monitoring substantially. It is worth mentioning that it is irrelevant which of the animals A3 and A4 that first reaches an identification zone 1 10 or 1 1 1 . It is also immaterial which identification station that is reached. How- ever, of course, in order to resolve the ambiguity between two animal traces at least one of the animals must be identified at an identification zone. As mentioned earlier, the identification means T1 , T2, T3 and T4 may include a transponder, an RFID tag, a bar code tag and/or a color area. According to embodi- ments of the invention, such an identification means can also be combined with identification via physical characteristics of the animal body, such as the head shape, various body measures or coat markings/lines. This is especially advantageous in connection with ambiguity resolution as described above, since thereby further robustness is attained.

Preferably, the data processing means 120 includes, or is associated with, a computer readable medium M, e.g. in the form of a memory module, such that the data processing means 120 has access to the contents of this medium M. Furthermore, a pro- gram is recorded in the computer readable medium M, and the program is adapted to make a data processor in the data processing means 120 control the process described above when the program is run on the processor.

In order to sum up, we will now describe the general method ac- cording to the invention with reference to the flow diagram in Figure 2. Here, we presume that the behavior of a group of livestock animals located within a predefined area is to be monitored, and that each animal in the group has a respective identification means uniquely identifying the animal. An identification station for determining the animals' identities is also distinct from any image registering means used for recording image data of the animals; based upon which image data the animals are tracked within the area.

A first step 210 checks whether an animal identity has been registered at an identification station, and if not the procedure loops back and stays in step 210. If, however, at least one animal identity is registered in step 210, a step 220 follows and image data is recorded. As mentioned above, this data represents the predefined area. Based on the image data, a step 230 derives a movement pattern for each animal located in said area.

A step 240 checks whether an animal identity has again been registered at an identification station, and if not, the procedure loops back to step 220 for continued recording of image data, and in step 230, updating of the movement patterns. If in step 240 it is found that an animal identity is registered, a subsequent step 250 investigates whether this identity has already been registered within the predefined area. If so, a step 260 follows, and otherwise the procedure continues to a step 270. Step 270 starts a new animal trace. I.e. this represents the case when an animal enters fresh the area to be monitored. Thereafter, the procedure loops back to step 220. Step 260 updates an existing animal trace based on the renewed identity registration in step 250, As mentioned above, this may involve resolving an ambiguity between different animal traces. Thereafter, the procedure loops back to step 220.

All of the process steps, as well as any sub-sequence of steps, described with reference to Figure 2 above may be controlled by means of a programmed computer apparatus. Moreover, although the embodiments of the invention described above with reference to the drawings comprise computer apparatus and processes performed in computer apparatus, the invention thus also extends to computer programs, particularly computer programs on or in a carrier, adapted for putting the invention into practice. The program may be in the form of source code, object code, a code intermediate source and object code such as in partially compiled form, or in any other form suitable for use in the implementation of the process according to the invention. The program may either be a part of an operating system, or be a separate application. The carrier may be any entity or device capable of carrying the program. For example, the carrier may comprise a storage medium, such as a Flash memory, a ROM (Read Only Memory), for example a DVD (Digital Video/Versatile Disk), a CD (Compact Disc) or a semiconductor ROM, an EP- ROM (Erasable Programmable Read-Only Memory), an EEPROM (Electrically Erasable Programmable Read-Only Memory), or a magnetic recording medium, for example a floppy disc or hard disc. Further, the carrier may be a transmissible carrier such as an electrical or optical signal which may be conveyed via electrical or optical cable or by radio or by other means. When the program is embodied in a signal which may be conveyed directly by a cable or other device or means, the carrier may be consti- tuted by such cable or device or means, Alternatively, the carrier may be an integrated circuit in which the program is embedded, the integrated circuit being adapted for performing, or for use in the performance of, the relevant processes.

Although the invention is advantageous in connection with cow milking, the invention is equally well adapted for implementation in milking machines for any other kind of mammals, such as goats, sheep or buffaloes.

The term "comprises/comprising" when used in this specification is taken to specify the presence of stated features, integers, steps or components. However, the term does not preclude the presence or addition of one or more additional features, integers, steps or components or groups thereof.

The reference to any prior art in this specification is not, and should not be taken as, an acknowledgement or any suggestion that the referenced prior art forms part of the common general knowledge in Australia, or any other country.

The invention is not restricted to the described embodiments in the figures, but may be varied freely within the scope of the claims.

Claims

1 . An arrangement for monitoring the behavior of livestock animals, each of said animals (A1 , A2, A3, A4) having a respective identification means (T1 , T2, T3, T4) uniquely identifying the animal in question , the arrangement comprising:

at least one image registering means (C 1 , C2, C3, C4) configured to repeatedly record image data (d1 , d2, d3, d4) representing animals (A1 , A2, A3, A4) located within a predefined area (100), and

a data processing means (120) configured to receive the image data (d1 , d2, d3, d4), and based thereon derive a movement pattern (M P) for each of said animals (A1 , A2, A3, A4), characterized in that the arrangement comprises at least one identification station (1 10, 1 1 1 ) distinct from the at least one image registering means (C 1 , C2, C3, C4), and the at least one identification station ( 10, 1 1 1 ) is configured to register an animal identity (I D1 ) in response to the passing of one of said identification means (T1 ) within an operating zone of the identification station (1 10, 1 1 1 ).

2. The arrangement according to claim 1 , wherein the movement pattern (MP) includes a series of data entries, each entry comprising a coordinate reflecting an animal position within the predefined area (100), and a time stamp indicating a point in time when the image data was recorded based on which image data the coordinate was derived; and the processing means (120) is configured to trace said animals (A1 , A2, A3, A4) within the predefined area (100) based on the series of data entries, a trace for a given animal starting at one of the at least one identification station (1 10) and extending to an animal position given by a coordinate of a latest data entry for the animal.

3. The arrangement according to claim 2, wherein the arrangement comprises at least two image registering means (C 1 , C2, C3, C4) of which at least one first image registering means (C 1 ) is configured to record image data (d 1 ) representing a first sub-area of the predefined area (100) and at least one second image registering means (C2) is configured to record image data (d2) representing the first sub-area, and the data processing means (120) is configured to use the image data (d2) recorded by the at least one second image registering means (C2) as a complement to the image data (d1 ) recorded by the at least one first image registering means (C1 ) for tracing said animals (A1 , A2, A3, A4) within the first sub-area.

4. The arrangement according to any one of claims 2 or 3, wherein the processing means (120) is configured to update a trace for a given animal (A1 ) upon reentry of the animal's identification means (T1 ) within the operating zone of one of the at least one identification station ( 10).

5. The arrangement according to any one of the preceding claims, wherein at least one of the at least one identification station (1 10) is arranged at an entry to the predefined area (100).

6. The arrangement according to any one of claims 4 or 5, wherein the processing means (120) is configured to resolve an ambiguity between the traces of a first and a second animal (A3, A4) by, in response to a reentry of the first (A3) animal's transponder means (T3) within the operating zone of one of the at least one identification station (1 10), tracing back the first animal (A3) to an ambiguity zone (AZ) in which the recorded image data has proven to provide insufficient accuracy to distinguish the first animal (A3) from the second animal (A4).

7. The arrangement according to any one of the preceding claims, wherein the identification means (T1 , T2, T3, T4) comprises at least one of a transponder, an RFID tag, a bar code tag, a color area and a given physical characteristics of the ani- mal body.

8. The arrangement according to any one of the preceding claims, wherein the operating zone of each of the at least one identification station (1 10) has an extension in space relative to a typical size of each of said animals (A1 , A2, A3, A4) and a ty- pical location for the identification means (T1 , T2, T3, T4) on the animal such that it can be expected that only one identification means at the time may be located within the operating zone.

9. The arrangement according to any one of claims 2 to 8, wherein

at least one of the at least one image registering means

(C1 , C2, C3, C4) includes a camera configured to register three- dimensional image data in real time, and

the data processing means (120) is configured to derive the movement pattern (MP) based on the three-dimensional ima- ge data, each entry in the movement pattern (MP) further comprising height data reflecting an animal elevation above a ground surface of the predefined area (100).

10. The arrangement according to any one of the preceding claims, wherein the movement pattern (MP) for each animal in- eludes one or more of:

a current location,

a distance covered during a given interval,

a total lying down time during a given interval,

a total number of climbing events during a given interval, a total time spent at a feeding station (130),

a number of visits to a feeding station (130),

a total time spent at a drinking station (140), and

a number of visits to a drinking station (140).

1 1. A method of monitoring the behavior of livestock animals located within a predefined area (100), each of said animals

(A1 , A2, A3, A4) having a respective identification means (T1 , T2, T3, T4) uniquely identifying the animal in question, the method comprising: repeatedly recording image data (d 1 , 62, d3, d4) representing the animals (A1 , A2, A3, A4), the image data (d 1 , d2, d3, d4) being recorded via at least one image registering means (C 1 , C2, C3_t C4), and

deriving a movement pattern (MP) for each of said animals

(A1 , A2, A3, A4) by processing the image data (d 1 , d2, d3, d4), characterized by

registering an animal identity (ID1 ) in response to the passing of one of said identification means (T1 ) within an operating zone of an identification station (1 10) distinct from the at least one image registering means (C 1 , C2 , C3, C4),

12. The method according to claim 1 1 , wherein the movement pattern (MP) includes a series of data entries, each entry comprising a coordinate reflecting an animal position within the pre- defined area (1 00), and a time stamp indicating a point in time when the image data was recorded based on which image data the coordinate was derived; and the method comprises tracing said animals (A1 , A2, A3, A4) within the predefined area (100) based on the series of data entries, a trace for a given animal starting at one of the at least one identification station (1 10) and extending to an animal position given by a coordinate of a latest data entry for the animal.

13. The method according to claim 12, comprising:

recording image data (d 1 ) representing a first sub-area of the predefined area (100) via at least one first image registering means (C1 ),

recording image data (d2) representing the first sub-area of the predefined area (100) via at least one second image registering means (C2), and

using the image data (d2) recorded by the at least one second image registering means (C2) as a complement to the image data (d1 ) recorded by the at least one first image registering means (C 1 ) to trace said animals (A1 , A2, A3, A4) within the first sub-area.

14. The method according to any one of claims 12 or 13, comprising updating a trace for a given animal (A1 ) upon reentry of the animal's identification means (T1 ) within the operating zone of one of the at least one identification station (1 10),

15. The method according to any one of claims 1 1 to 14, comprising arranging at least one of the at least one identification station (1 10) at an entry to the predefined area (100).

16. The method according to any one of claims 14 or 15, comprising resolving an ambiguity between the traces of a first and a second animal (A3, A4) by

detecting a reentry of the first (A3) animal's transponder means (T3) within the operating zone of one of the at least one identification station (1 10), and in response thereto

tracing back the first animal (A3) to an ambiguity zone (AZ) in which the recorded image data has proven to provide insufficient accuracy to distinguish the first animal (A3) from the second animal (A4).

17. The method according to any one of claims 1 1 to 16, wherein the identification means (T1 , T2, T3, T4) comprises at least one of a transponder, an RFID tag, a bar code tag, a color area and a given physical characteristics of the animal body.

18. The method according to any one of claims 1 1 to 17, wherein the operating zone of each of the at least one identification station (1 10) has an extension in space relative to a typical size of each of said animals (A1 , A2, A3, A4) and a typical location for the identification means (T1 , T2, T3, T4) on the animal such that it can be expected that only one identification means at the time may be located within the operating zone.

19. The method according to any one of claims 11 to 18, whe- rein at least one of the at least one image registering means (C1 ,

C2, C3, C4) includes a camera configured to register three- dimensional image data in real time, and the method comprises:

deriving the movement pattern (MP) based on the three-dimensional image data, each entry in the movement pattern (MP) further comprising height data reflecting an animal elevation abo- ve a ground surface of the predefined area (100).

20. The method according to any one of the claims 1 1 to 19, wherein the movement pattern (MP) for each animal includes one or more of:

a current location,

a distance covered during a given interval,

a total lying down time during a given interval, a total number of climbing events during a given interval, a total time spent at a feeding station (130), a number of visits to a feeding station (130), a total time spent at a drinking station (140), and a number of visits to a drinking station (140).

21. A computer program loadable into the internal memory (M) of a computer, comprising software for controlling the steps of any one of claims 1 1 to 20 when the program is run on the com- puter.

22. A computer readable medium (M) having a program recorded thereon, where the program is to make a computer control the steps of any one of claims 1 1 to 20 when the program is loaded into the computer.