WO2012115558A1 - Apparatus and method for tracking a stabled animal - Google Patents

Apparatus and method for tracking a stabled animal Download PDF

Info

Publication number
WO2012115558A1
WO2012115558A1 PCT/SE2011/051550 SE2011051550W WO2012115558A1 WO 2012115558 A1 WO2012115558 A1 WO 2012115558A1 SE 2011051550 W SE2011051550 W SE 2011051550W WO 2012115558 A1 WO2012115558 A1 WO 2012115558A1
Authority
WO
WIPO (PCT)
Prior art keywords
location
pixel
animal
monitored area
indicated location
Prior art date
Application number
PCT/SE2011/051550
Other languages
French (fr)
Inventor
Gösta FORSÉN
Original Assignee
Fotonic I Norden Ab
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Fotonic I Norden Ab filed Critical Fotonic I Norden Ab
Publication of WO2012115558A1 publication Critical patent/WO2012115558A1/en

Links

Classifications

    • AHUMAN NECESSITIES
    • A01AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
    • A01KANIMAL HUSBANDRY; AVICULTURE; APICULTURE; PISCICULTURE; FISHING; REARING OR BREEDING ANIMALS, NOT OTHERWISE PROVIDED FOR; NEW BREEDS OF ANIMALS
    • A01K29/00Other apparatus for animal husbandry
    • A01K29/005Monitoring or measuring activity, e.g. detecting heat or mating
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/20Analysis of motion
    • G06T7/277Analysis of motion involving stochastic approaches, e.g. using Kalman filters
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/10Image acquisition modality
    • G06T2207/10016Video; Image sequence
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/20Special algorithmic details
    • G06T2207/20076Probabilistic image processing
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/30Subject of image; Context of image processing
    • G06T2207/30241Trajectory

Definitions

  • the present invention relates to tracking of stabled animals.
  • the invention relates to an apparatus and a method for monitoring livestock in a monitored area.
  • a livestock monitoring system is known.
  • Animals are provided with RFID animal tags which include GPS capability for real time monitoring of animal location.
  • Animal identifiers and locations are uploaded in a local area wireless network to a receiver, which in turn uploads the data to a Web server linked to a host processor for databases.
  • the system enables recording of movement history of individual animals.
  • the system is very complex, with a multitude of subsystems, and it relies on a multitude of radio interface standards.
  • Animal tags are usually attached to an animal's ear once and for all at birth.
  • RFID animal tags certain rigidity is also built into the system. This may be a problem if a new group of livestock, marked according to a different coding scheme is introduced in an incumbent system of marked livestock. Re-tagging of animals would then include removing the old RFID tag, and adding the appropriate tag to each of the newly introduced animals. Further, GPS has very limited coverage indoors. This means that the system disclosed in GB2387465 may be relatively expensive and time consuming to commission, and may be relatively more suitable for outdoor monitoring of herds of livestock that are bom at the premises rather than moved there from another location. Today however, it is more likely that an animal is moved at least once during its life, and it is likely that an animal is stabled during at least a period of its life.
  • a camera system for detecting livestock in heat is known.
  • Each animal is equipped with a tag marked with an identity code intended to be read by a camera from a distance.
  • Fig 1 shows a camera placed on a side wall. Regardless of placement, the disclosed camera system is unable to distinguish between multiple locations - and therefore further fail to register movements - along the camera's line-of- sight.
  • the camera be placed on rails.
  • the identification tag be placed in an upright and elevated position, on a collar adapted for each individual animal's physiognomy.
  • figure 2 indicates that the disclosed camera system may only detect heat- related "hyperactivity" through pulse-shaped binary readings. Further, in a non-clean environment, such as in a livestock stable, optical reading of an identification code from a distance is likely to be obstructed by dirt or debris covering the tag. Such obstruction will render the disclosed system non-functional, since it relies on an initial
  • FR2759541 much like GB2387465 discloses a rigid system, with the added drawback of being less appropriate in non-clean environments.
  • the disclosed system while it may be sufficient for detecting the occurrence of an animal in heat, may not accurately register movement patterns.
  • the object of the present invention is to obviate at least some of the above disadvantages and provide improved methods, apparatuses and systems for individual monitoring and registering of movement of livestock and other animals that may be kept indoors in periods.
  • it is an object to provide flexibility, scalability and simplicity.
  • a first aspect of the invention is an apparatus for monitoring livestock.
  • the apparatus comprises a light source, a light sensitive pixel grid sensor with a field-of- view of a monitored area, and a processing unit adapted and configured to interpret a reading from an individual pixel in the pixel grid as indicating an equivalent location corresponding to the pixel's field-of-view.
  • the apparatus of the first aspect of the invention is adapted and configured to
  • the apparatus of the first aspect may further be adapted and configured to receive a reading in a pixel indicating an equivalent location in the monitored area; and statistically deduct that the lastly indicated location is not associated with any precedently indicated location. It may further be adapted and configured to receive, from an animal identification station present in a known identification location, an animal identification; statistically deduct that a latest indicated location is associated with the received animal identification; and register the latest indicated location as associated with the first animal identification.
  • the processing unit may comprise Kalman filter functionality applicable for statistical deduction of association.
  • a second aspect of the present invention is a method for monitoring livestock in a monitored area comprising locations corresponding to equivalent location associated with the pixel's field-of-view. The method comprises the steps of
  • the method of the second aspect of the invention may comprise the further steps of receiving a reading in a pixel indicating an equivalent location in the monitored area;
  • the method may comprise the further steps of
  • identification location an animal identification
  • Kalman filter functionality may be applied in the statistical deduction of association comprised in the method steps above.
  • a third aspect of the invention is a computer program comprising program instructions for causing a computer to perform the process of the second aspect of the invention when said product is run on a computer.
  • the computer program of the third aspect of the invention may be embodied on a record medium, stored in a computer memory, embodied on a read-only memory, or carried on an electrical carrier signal.
  • a fourth aspect of the invention is a computer program product comprising a computer readable medium, having thereon: computer program code means, when said program is loaded, to make the computer execute the process of the second aspect of the invention.
  • Each animal X, Y, Z in a monitored area 200 may be equipped with a reflective surface such as reflector tags 1, 2 and 3, as illustrated in Figure la.
  • the identity tag that is usually clipped to a cow's ear may be equipped with a highly reflective surface. This is easily accomplished by sticking highly reflective adhesive film to the identity tag. Reflected light from such a reflector tag may be detected by a light-sensitive apparatus 100 comprising a grid 1 10 of light sensitive pixels.
  • tags 1, 2 and 3 are identical, with the exception of the tags' l, 2, 3 respective locations in the monitored area 200.
  • the light sensor's field-of-view effectively defines the monitored area.
  • the monitored area can consequently be thought of as a uniform grid of locations or patches, which are defined by pixel fields-of-view. Since a reading in a certain pixel P therefore unequivocally signifies a presence of a reflector tag in the pixel's, P's, associated uniform location L, a reading in any one individual pixel P can be said to be equivalent to presence in that particular pixel's P associated location L. This is illustrated in figures 2a and 2b. Hence, in each moment in time a reflector tag, and therefore also the animal to which the reflector tag is attached, can be associated with a location.
  • PRF pulse repetition frequency
  • that is sufficiently high in relation to how fast an animal can move
  • two sequential readings at t beau and t n+1 , within a certain radius R from each other can be deducted to be caused by, and therefore associated with, one and the same reflector tag 1 , 2 or 3, as illustrated in Figure lb.
  • the numbering may signify that the tag marked 1 was detected first by the apparatus 100, and then tag 2, followed by tag 3. Even if the identity of the animal carrying the tag is not known, distinguishing between several unique animals is enabled based on the monitored movement of each animals associated reflector tag. The animal whose tag was discovered as number 3, may have recently been let into the monitored area.
  • the first few markings may be tied only to the tag discovery number.
  • the identity of the animal Z can be associated to the tag that is concurrently detected in that same location 300.
  • identified animals are represented by white numbers on black, and unidentified animals are represented by black numbers on white.
  • the firstly discovered tag 1 which has previously also been identified as animal X, has not been near the watering station W all day. This implies that the animal X has not been drinking all day, and might be sick. It is an abnormal behaviour, here represented by a triangular shape rather than a "normal" circle.
  • the apparatus 100 comprises a sensor comprising a grid 1 10 of light sensitive pixels, a processing unit 130 and a memory unit 140.
  • the apparatus 100 may also comprise a light source 150.
  • the light source 150 may emit light towards a monitored area 200, and may be reflected by a reflector tag T comprised in the monitored area 200.
  • the tag T is attached to an animal.
  • the emitted light may be visible light, near-infrared light or infrared light.
  • PRF pulse repetition frequency
  • the processing unit 130 receives information from the sensor that identifies the pixel P that made the detection.
  • the identifying information may comprise coordinates relating to the pixel grid 110, such as for instance the pixel's column and row number.
  • the processing unit 130 is adapted and configured to interpret a reading from an individual pixel in the pixel grid 110 as indicating an equivalent location corresponding to the pixel's field-of-view.
  • the processing unit 130 may be adapted and configured to store this received information as an equivalent data post of applicable format in the memory unit 140.
  • the relation between the resolution of the sensor, the distance between the sensor and the monitored area and the size of the reflector tag can be adapted such that the size of the reflective surface is of the same magnitude as a location of the above definition.
  • two separate reflector tags T cannot cause simultaneous readings in immediately adjacent pixels. Therefore, if simultaneous light detections are made in a group of adjacent pixels, they are caused by, and therefore associated with, the same reflector tag T.
  • the processing unit 130 may then be adapted to select one location L within the monitored area 200 as the reflector tag's associated pixel equivalent location. Selection may be random, based on relative received light energy in respective pixel, or a pixels geometric distance to the edge of the pixel group.
  • the data posts registered in the memory unit 140 may be multidimensional with retrievable associations between dimensions such as read location L, time t of reading, tag T, movement pattern M.
  • the data posts and the associations can be physically implemented in a number of ways known to a person skilled in the art, as long as the data and the associations are mutually retrievable.
  • Kalman filter functionality 135 may be implemented in the processing unit 130, in such a way that a first reading at t n and a second reading at t n+1 can be deducted to derive from the same tag T, and therefore the same animal. With a Kalman filter functionality 135 implemented in the processing unit 130 this deduction can be made with great reliability.
  • the apparatus 100 may also comprise an output unit 160 that serves as an interface between the apparatus 100 and the environment.
  • the output unit 160 may be configured as a Man-Machine-Interface, for instance a display, or it may be adapted and configured to provide input to another apparatus, such as a computerized control system in a milk production facility.
  • the apparatus 100 may also be operative to control subsystems controlling an animal's access to a certain sub-area of the monitored area 200, or to control provision of e.g. food or water to an individual animal.
  • the processing unit 130 may process the stored dataposts in a number of ways.
  • a registered movement pattern may be assessed against at least one reference criterion of a set of reference criteria. From such an assessment it is possible to deduct if an animal is sick, has eaten too little or too much, has been milked too recently or too long ago.
  • the apparatus 100 may be a compact apparatus where the light source 150, the sensor including sensor grid 110, the processing unit 130 and the memory unit 140 are all comprised within the same housing.
  • the apparatus 100 may also to a certain degree use distributed resources. While tracking of one single animal, no identification is necessary. If there is only one reflection from a tag T to be monitored, the registered movement pattern M obviously belongs to that single animal. While tracking a limited number of animals, it may still be quite sufficient to know that one animal of the limited number of animals show a deviating movement pattern.
  • a distributed apparatus 100 may comprise additional light sources and additional light sensors.
  • the monitored global area is a composite of each local monitored area of each light detector comprised in the distributed apparatus 100.
  • the Kalman filter functionality 135 can be used during system setup, to associate overlapping fringe spaces of a first and a second local monitored area. During such a setup, a reflector tag can be moved around in overlapping fringe locations.
  • the processing unit 130 may be configured to map a location in a first local monitored area to a detected location from the second local monitored area.
  • the local monitored area of each light sensor is mapped to a positioning system that is global relative the apparatus.
  • the global positioning system may be the GPS system or some other system that offers service in the monitored areas.
  • Certain embodiments of a system according to the present invention comprise an identification station ID.
  • the identification station is adapted and configured to identify an animal that comes within a certain range of the station, or that passes the station according to some predefined manner. An animal coming within reach of an identification station may have its reflection path associated with the identity read by the identification station. Identification may be performed according to any available animal identification system. Examples include optical reading of a tag attached to the animal of biometric recognition of various kinds.
  • the apparatus 100 may be adapted and configured to perform a method 400 for monitoring livestock in a monitored area 200 comprising pixel equivalent locations. Each such location corresponds to a field-of-view of a certain pixel comprised in the pixel grid 1 10 of the apparatus 100.
  • the method 400 which is a second aspect of the invention, will be described in relation to Figure 4.
  • the method comprises the steps of receiving 410 a reading in a pixel indicating an equivalent location 31 in the monitored area 200;
  • the method comprises the further steps of receiving 415 a reading in a pixel indicating an equivalent location 40 in the monitored area 200; and deducting 425 statistically that the lastly indicated location 40 is not associated with any precedently indicated location 30, 31. Instead a subsequent location 41 may be associated with the location 40, and thus associated with the location 40.
  • the method 400 comprises the further steps of receiving 430, from an animal identification station present in a known identification location 300, an animal identification;
  • the animal's identification can be sent in or together with the alert control signal in the sending 460 step.
  • All embodiments of the method may apply Kalman filtering functionality in the statistical deduction of association. Kalman filtering may further be used during assessment of movement pattern criteria or other assessment tasks.
  • the advantages offered by embodiments of the present invention are several. First of all the apparatuses 100 and methods 400 are entirely scalable.
  • the highly reflective tags may be identical as far as their reflective properties are concerned.
  • Certain embodiments of the apparatus 100 according to the present invention emit intermittent light flashes. This is an advantage because it lowers the overall system power consumption. In an apparatus 100 that emits visible light it is a further advantage that the light may me emitted during a time period so short that a living creature may not register it with its bare eyes.

Landscapes

  • Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Environmental Sciences (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Biophysics (AREA)
  • Animal Husbandry (AREA)
  • Biodiversity & Conservation Biology (AREA)
  • Image Analysis (AREA)

Abstract

The present invention relates to an apparatus for monitoring livestock comprising a light source, a light sensitive pixel grid (1 10) sensor with a field-of-view of a monitored area (200), and a processing unit adapted and configured to interpret a reading from an individual pixel in the pixel grid (1 10) as indicating an equivalent location corresponding to the pixel's field-of-view. The apparatus (100) is adapted and configured to receive a reading in a pixel indicating an equivalent location in the monitored area (200); statistically deduct that the indicated location is associated with a precedently indicated location; convert a set of associated locations to a movement pattern associated with an animal; assess the movement pattern against a set of reference criteria; and send an alert control signal if at least one reference criterion of the reference criteria is met. The invention also relates to a method and a computer program product.

Description

APPARATUS AND METHOD FOR TRACKING A STABLED ANIMAL
Technical Field
The present invention relates to tracking of stabled animals. In particular the invention relates to an apparatus and a method for monitoring livestock in a monitored area.
Background
From GB2387465 a livestock monitoring system is known. Animals are provided with RFID animal tags which include GPS capability for real time monitoring of animal location. Animal identifiers and locations are uploaded in a local area wireless network to a receiver, which in turn uploads the data to a Web server linked to a host processor for databases. The system enables recording of movement history of individual animals. However, the system is very complex, with a multitude of subsystems, and it relies on a multitude of radio interface standards. Animal tags are usually attached to an animal's ear once and for all at birth. With the introduction of
RFID animal tags, certain rigidity is also built into the system. This may be a problem if a new group of livestock, marked according to a different coding scheme is introduced in an incumbent system of marked livestock. Re-tagging of animals would then include removing the old RFID tag, and adding the appropriate tag to each of the newly introduced animals. Further, GPS has very limited coverage indoors. This means that the system disclosed in GB2387465 may be relatively expensive and time consuming to commission, and may be relatively more suitable for outdoor monitoring of herds of livestock that are bom at the premises rather than moved there from another location. Today however, it is more likely that an animal is moved at least once during its life, and it is likely that an animal is stabled during at least a period of its life.
From FR2759541 a camera system for detecting livestock in heat is known. Each animal is equipped with a tag marked with an identity code intended to be read by a camera from a distance. Fig 1 shows a camera placed on a side wall. Regardless of placement, the disclosed camera system is unable to distinguish between multiple locations - and therefore further fail to register movements - along the camera's line-of- sight. As an improvement it is suggested that the camera be placed on rails. It is also suggested that the identification tag be placed in an upright and elevated position, on a collar adapted for each individual animal's physiognomy. Yet, still with these adaptions, figure 2 indicates that the disclosed camera system may only detect heat- related "hyperactivity" through pulse-shaped binary readings. Further, in a non-clean environment, such as in a livestock stable, optical reading of an identification code from a distance is likely to be obstructed by dirt or debris covering the tag. Such obstruction will render the disclosed system non-functional, since it relies on an initial
identification. Hence, FR2759541, much like GB2387465 discloses a rigid system, with the added drawback of being less appropriate in non-clean environments. Lastly, the disclosed system, while it may be sufficient for detecting the occurrence of an animal in heat, may not accurately register movement patterns.
Summary
The object of the present invention is to obviate at least some of the above disadvantages and provide improved methods, apparatuses and systems for individual monitoring and registering of movement of livestock and other animals that may be kept indoors in periods. In particular it is an object to provide flexibility, scalability and simplicity.
A first aspect of the invention is an apparatus for monitoring livestock. The apparatus comprises a light source, a light sensitive pixel grid sensor with a field-of- view of a monitored area, and a processing unit adapted and configured to interpret a reading from an individual pixel in the pixel grid as indicating an equivalent location corresponding to the pixel's field-of-view. The apparatus of the first aspect of the invention is adapted and configured to
receive a reading in a pixel indicating an equivalent location in the monitored area; statistically deduct that the indicated location is associated with a precedently indicated location; convert a set of associated locations to a movement pattern associated with an animal; assess the movement pattern against a set of reference criteria; and send an alert control signal if at least one reference criterion of the reference criteria is met.
The apparatus of the first aspect may further be adapted and configured to receive a reading in a pixel indicating an equivalent location in the monitored area; and statistically deduct that the lastly indicated location is not associated with any precedently indicated location. It may further be adapted and configured to receive, from an animal identification station present in a known identification location, an animal identification; statistically deduct that a latest indicated location is associated with the received animal identification; and register the latest indicated location as associated with the first animal identification. The processing unit may comprise Kalman filter functionality applicable for statistical deduction of association.
A second aspect of the present invention is a method for monitoring livestock in a monitored area comprising locations corresponding to equivalent location associated with the pixel's field-of-view. The method comprises the steps of
receiving a reading in a pixel indicating an equivalent location in the monitored area;
deducting statistically that the indicated location is associated with a precedently indicated location;
converting a set of associated locations to a movement pattern associated with an animal;
assessing the movement pattern against a set of reference criteria, and;
sending an alert control signal if at least one reference criterion of the reference criteria is met.
The method of the second aspect of the invention may comprise the further steps of receiving a reading in a pixel indicating an equivalent location in the monitored area;
Deducting statistically that the lastly indicated location is not associated with any precedently indicated location.
The method may comprise the further steps of
receiving, from an animal identification station present in a known
identification location, an animal identification;
deducting statistically that a latest indicated location is associated with the received animal identification.
Kalman filter functionality may be applied in the statistical deduction of association comprised in the method steps above.
A third aspect of the invention is a computer program comprising program instructions for causing a computer to perform the process of the second aspect of the invention when said product is run on a computer. The computer program of the third aspect of the invention may be embodied on a record medium, stored in a computer memory, embodied on a read-only memory, or carried on an electrical carrier signal.
A fourth aspect of the invention is a computer program product comprising a computer readable medium, having thereon: computer program code means, when said program is loaded, to make the computer execute the process of the second aspect of the invention. Detailed Description
A solution to this problem is presented below.
Each animal X, Y, Z in a monitored area 200 may be equipped with a reflective surface such as reflector tags 1, 2 and 3, as illustrated in Figure la. As an example the identity tag that is usually clipped to a cow's ear may be equipped with a highly reflective surface. This is easily accomplished by sticking highly reflective adhesive film to the identity tag. Reflected light from such a reflector tag may be detected by a light-sensitive apparatus 100 comprising a grid 1 10 of light sensitive pixels. As far as the grid 110 is concerned, tags 1, 2 and 3 are identical, with the exception of the tags' l, 2, 3 respective locations in the monitored area 200. The light sensor's field-of-view effectively defines the monitored area. The monitored area can consequently be thought of as a uniform grid of locations or patches, which are defined by pixel fields-of-view. Since a reading in a certain pixel P therefore unequivocally signifies a presence of a reflector tag in the pixel's, P's, associated uniform location L, a reading in any one individual pixel P can be said to be equivalent to presence in that particular pixel's P associated location L. This is illustrated in figures 2a and 2b. Hence, in each moment in time a reflector tag, and therefore also the animal to which the reflector tag is attached, can be associated with a location. If the light source emits pulsed light with a pulse repetition frequency PRF = 1/| tn+i- 1„| that is sufficiently high in relation to how fast an animal can move, two sequential readings at t„ and tn+1, within a certain radius R from each other can be deducted to be caused by, and therefore associated with, one and the same reflector tag 1 , 2 or 3, as illustrated in Figure lb. In this particular case the numbering may signify that the tag marked 1 was detected first by the apparatus 100, and then tag 2, followed by tag 3. Even if the identity of the animal carrying the tag is not known, distinguishing between several unique animals is enabled based on the monitored movement of each animals associated reflector tag. The animal whose tag was discovered as number 3, may have recently been let into the monitored area.
Therefore, the first few markings may be tied only to the tag discovery number.
However, as soon as the animal Z may be detected by an identification station ID with a known location 300, the identity of the animal Z can be associated to the tag that is concurrently detected in that same location 300. In Figure lb, as an example, identified animals are represented by white numbers on black, and unidentified animals are represented by black numbers on white. To exemplify further, the firstly discovered tag 1 , which has previously also been identified as animal X, has not been near the watering station W all day. This implies that the animal X has not been drinking all day, and might be sick. It is an abnormal behaviour, here represented by a triangular shape rather than a "normal" circle. An apparatus 100 according to an embodiment of the present invention will now be described in relation to Figure 2a. The apparatus 100 comprises a sensor comprising a grid 1 10 of light sensitive pixels, a processing unit 130 and a memory unit 140. The apparatus 100 may also comprise a light source 150. The light source 150 may emit light towards a monitored area 200, and may be reflected by a reflector tag T comprised in the monitored area 200. The tag T is attached to an animal. The emitted light may be visible light, near-infrared light or infrared light. The light may be emitted in pulses with a pulse repetition frequency PRF=T/| tn+1- 1„|. When an individual pixel P in the pixel grid 110 detects reflected light, the processing unit 130 receives information from the sensor that identifies the pixel P that made the detection. The identifying information may comprise coordinates relating to the pixel grid 110, such as for instance the pixel's column and row number. The processing unit 130 is adapted and configured to interpret a reading from an individual pixel in the pixel grid 110 as indicating an equivalent location corresponding to the pixel's field-of-view. The processing unit 130 may be adapted and configured to store this received information as an equivalent data post of applicable format in the memory unit 140.
The relation between the resolution of the sensor, the distance between the sensor and the monitored area and the size of the reflector tag can be adapted such that the size of the reflective surface is of the same magnitude as a location of the above definition. With an appropriately high system resolution two separate reflector tags T cannot cause simultaneous readings in immediately adjacent pixels. Therefore, if simultaneous light detections are made in a group of adjacent pixels, they are caused by, and therefore associated with, the same reflector tag T. As illustrated in Figure 2b, the processing unit 130 may then be adapted to select one location L within the monitored area 200 as the reflector tag's associated pixel equivalent location. Selection may be random, based on relative received light energy in respective pixel, or a pixels geometric distance to the edge of the pixel group. The data posts registered in the memory unit 140 may be multidimensional with retrievable associations between dimensions such as read location L, time t of reading, tag T, movement pattern M. The data posts and the associations can be physically implemented in a number of ways known to a person skilled in the art, as long as the data and the associations are mutually retrievable. Kalman filter functionality 135 may be implemented in the processing unit 130, in such a way that a first reading at tn and a second reading at tn+1 can be deducted to derive from the same tag T, and therefore the same animal. With a Kalman filter functionality 135 implemented in the processing unit 130 this deduction can be made with great reliability.
The apparatus 100 may also comprise an output unit 160 that serves as an interface between the apparatus 100 and the environment. The output unit 160 may be configured as a Man-Machine-Interface, for instance a display, or it may be adapted and configured to provide input to another apparatus, such as a computerized control system in a milk production facility. The apparatus 100 may also be operative to control subsystems controlling an animal's access to a certain sub-area of the monitored area 200, or to control provision of e.g. food or water to an individual animal.
The processing unit 130 may process the stored dataposts in a number of ways. A registered movement pattern may be assessed against at least one reference criterion of a set of reference criteria. From such an assessment it is possible to deduct if an animal is sick, has eaten too little or too much, has been milked too recently or too long ago.
The apparatus 100 may be a compact apparatus where the light source 150, the sensor including sensor grid 110, the processing unit 130 and the memory unit 140 are all comprised within the same housing. The apparatus 100 may also to a certain degree use distributed resources. While tracking of one single animal, no identification is necessary. If there is only one reflection from a tag T to be monitored, the registered movement pattern M obviously belongs to that single animal. While tracking a limited number of animals, it may still be quite sufficient to know that one animal of the limited number of animals show a deviating movement pattern.
A distributed apparatus 100 according to certain embodiments of the invention may comprise additional light sources and additional light sensors. The monitored global area is a composite of each local monitored area of each light detector comprised in the distributed apparatus 100. The Kalman filter functionality 135 can be used during system setup, to associate overlapping fringe spaces of a first and a second local monitored area. During such a setup, a reflector tag can be moved around in overlapping fringe locations. The processing unit 130 may be configured to map a location in a first local monitored area to a detected location from the second local monitored area. In other embodiments of the apparatus 100, the local monitored area of each light sensor is mapped to a positioning system that is global relative the apparatus. The global positioning system may be the GPS system or some other system that offers service in the monitored areas.
Certain embodiments of a system according to the present invention comprise an identification station ID. The identification station is adapted and configured to identify an animal that comes within a certain range of the station, or that passes the station according to some predefined manner. An animal coming within reach of an identification station may have its reflection path associated with the identity read by the identification station. Identification may be performed according to any available animal identification system. Examples include optical reading of a tag attached to the animal of biometric recognition of various kinds.
The apparatus 100 may be adapted and configured to perform a method 400 for monitoring livestock in a monitored area 200 comprising pixel equivalent locations. Each such location corresponds to a field-of-view of a certain pixel comprised in the pixel grid 1 10 of the apparatus 100. The method 400, which is a second aspect of the invention, will be described in relation to Figure 4.
The method comprises the steps of receiving 410 a reading in a pixel indicating an equivalent location 31 in the monitored area 200;
deducting 420 statistically that the indicated location 31 is associated with a precedently indicated location 30;
converting 440 a set of associated locations 30, 31 to a movement pattern 30; 31 associated with an animal;
assessing 450 the movement pattern 30; 31 against a set of reference criteria, and;
sending 460 an alert control signal if at least one reference criterion of the reference criteria is met. The receiving, deducting and converting steps are illustrated in respective figures 3a-c.
According to certain embodiments of the present invention, the method comprises the further steps of receiving 415 a reading in a pixel indicating an equivalent location 40 in the monitored area 200; and deducting 425 statistically that the lastly indicated location 40 is not associated with any precedently indicated location 30, 31. Instead a subsequent location 41 may be associated with the location 40, and thus associated with the location 40. This is illustrated in figures 3c-d. According to certain embodiments of the present invention, the method 400 comprises the further steps of receiving 430, from an animal identification station present in a known identification location 300, an animal identification;
deducting 435 statistically that a latest indicated location 31, 40 is associated with the received animal identification;
registering the latest indicated location 31 , 40 as associated with the received animal identification. This enables linking a certain animal's identity to its present location and to its current movement pattern. In the event that the movement pattern fulfills a certain reference criterion or a set of reference criteria, the animal's identification can be sent in or together with the alert control signal in the sending 460 step.
All embodiments of the method may apply Kalman filtering functionality in the statistical deduction of association. Kalman filtering may further be used during assessment of movement pattern criteria or other assessment tasks.
The advantages offered by embodiments of the present invention are several. First of all the apparatuses 100 and methods 400 are entirely scalable. The highly reflective tags may be identical as far as their reflective properties are concerned.
Therefore, animals can be added or removed without adaption.
Certain embodiments of the apparatus 100 according to the present invention emit intermittent light flashes. This is an advantage because it lowers the overall system power consumption. In an apparatus 100 that emits visible light it is a further advantage that the light may me emitted during a time period so short that a living creature may not register it with its bare eyes.

Claims

1. An apparatus (100) for monitoring livestock comprising a light source, a light sensitive pixel grid (110) sensor with a field-of-view of a monitored area (200), and a processing unit (130) adapted and configured to interpret a reading from an individual pixel in the pixel grid (110) as indicating an equivalent location corresponding to the pixel's field-of-view, the apparatus (100) adapted and configured to receive a reading in a pixel indicating an equivalent location (31) in the monitored area (200); statistically deduct that the indicated location (31) is associated with a precedently indicated location (30); convert a set of associated locations (30, 31) to a movement pattern (30; 31) associated with an animal; assess the movement pattern (30; 31) against a set of reference criteria; and send an alert control signal if at least one reference criterion of the reference criteria is met.
2. The apparatus according to claim 1, further adapted and configured to receive a reading in a pixel indicating an equivalent location (40) in the monitored area (200); and statistically deduct that the lastly indicated location (40) is not associated with any precedently indicated location (30, 31).
3. The apparatus according to claim 1 or 2, further adapted and configured to receive, from an animal identification station present in a known identification location (300), an animal identification; statistically deduct that a latest indicated location (31 , 40) is associated with the received animal identification; and register the latest indicated location (31 , 40) as associated with the first animal identification.
4. The apparatus according to any of the previous claims, in which the processing unit comprises Kalman filter functionality applicable for statistical deduction of association.
5. A method for monitoring livestock in a monitored area (200) comprising locations corresponding to equivalent location associated with the pixel's field-of-view, comprising the steps of
receiving (410) a reading in a pixel indicating an equivalent location (31) in the monitored area (200); deducting (420) statistically that the indicated location (31) is associated with a precedently indicated location (30);
converting (440) a set of associated locations (30, 31) to a movement pattern (30; 31) associated with an animal;
assessing (450) the movement pattern (30; 31) against a set of reference criteria; and
sending (460) an alert control signal if at least one reference criterion of the reference criteria is met.
6. The method according to claim 5 comprising the further steps of receiving (415) a reading in a pixel indicating an equivalent location (40) in the monitored area (200); and
deducting (425) statistically that the lastly indicated location (40) is not associated with any precedently indicated location (30, 31).
7. The method according to claim 5 or 6 comprising the further steps of
receiving (430), from an animal identification station present in a known identification location (300), an animal identification; and
deducting (435) statistically that a latest indicated location (31, 40) is associated with the received animal identification.
8. The method according to any of the previous claims, in which alman filter functionality is applied in the statistical deduction of association.
9. A computer program comprising program instructions for causing a computer to perform the process of any of the claims 5-8 when said program is run on a computer.
10. A computer program according to claim 9, embodied on a record medium, stored in a computer memory, embodied on a read-only memory, or carried on an electrical carrier signal.
1 1. A computer program product comprising a computer readable medium, having thereon: computer program code means, when said program is loaded, to make the computer execute the process of any of the claims 5-8.
PCT/SE2011/051550 2011-02-24 2011-12-21 Apparatus and method for tracking a stabled animal WO2012115558A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
SE1150163-2 2011-02-24
SE1150163 2011-02-24

Publications (1)

Publication Number Publication Date
WO2012115558A1 true WO2012115558A1 (en) 2012-08-30

Family

ID=46721108

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/SE2011/051550 WO2012115558A1 (en) 2011-02-24 2011-12-21 Apparatus and method for tracking a stabled animal

Country Status (1)

Country Link
WO (1) WO2012115558A1 (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP3336783A1 (en) * 2016-12-19 2018-06-20 Farmic AB Method and light assembly for tracking livestock inside a building
EP3335551A1 (en) * 2016-12-19 2018-06-20 Farmic AB Method and system for monitoring livestock inside a building
WO2018174812A1 (en) 2017-03-24 2018-09-27 Bmp Innovation Ab Systems and methods for identifying individual animals in a group of animals

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
FR2759541A1 (en) * 1997-02-18 1998-08-21 Visio Systemes Automatic monitoring system for farm animals kept within stalls
EP0945060A2 (en) * 1998-03-24 1999-09-29 Timothy P. Dr. Guichon Method and system for monitoring animals
JP2010004861A (en) * 2008-06-24 2010-01-14 Darwin:Kk System and method for animal individual identification

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
FR2759541A1 (en) * 1997-02-18 1998-08-21 Visio Systemes Automatic monitoring system for farm animals kept within stalls
EP0945060A2 (en) * 1998-03-24 1999-09-29 Timothy P. Dr. Guichon Method and system for monitoring animals
JP2010004861A (en) * 2008-06-24 2010-01-14 Darwin:Kk System and method for animal individual identification

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP3336783A1 (en) * 2016-12-19 2018-06-20 Farmic AB Method and light assembly for tracking livestock inside a building
EP3335551A1 (en) * 2016-12-19 2018-06-20 Farmic AB Method and system for monitoring livestock inside a building
WO2018174812A1 (en) 2017-03-24 2018-09-27 Bmp Innovation Ab Systems and methods for identifying individual animals in a group of animals
SE541890C2 (en) * 2017-03-24 2020-01-02 Bmp Innovation Ab Systems and methods for identifying individual animals in a group of animals

Similar Documents

Publication Publication Date Title
US9609848B2 (en) Method and system for localizing and displaying positions of autonomously mobile objects
CN102378996B (en) Energy efficient cascade of sensors for automatic presence detection
US9785806B2 (en) Low-frequency receiving for radio frequency identification
CN1950285B (en) Positioning system and method for moving platform
US7982614B2 (en) Method and apparatus for wireless asset tracking using asset tags with motion sensors
US10380858B2 (en) Method and montoring device for monitoring a tag
JP2014149828A5 (en)
US10706696B1 (en) Security system with distributed sensor units and autonomous camera vehicle
US9235795B2 (en) Optical system and method for monitoring and locating objects
RU2010106264A (en) DEVICE AND METHOD FOR PROVIDING INFORMATION ABOUT ANIMALS AT THEIR PASSAGE THROUGH AN ANIMAL ACCESS
WO2010099470A1 (en) Image processing sensor systems
AU2014290924A1 (en) Methods and systems relating to milking sheds
WO2011154949A2 (en) Optical tracking system and method for herd management therewith
WO2021063046A1 (en) Distributed target monitoring system and method
KR101697546B1 (en) Security control system using beacon
EP3335551A1 (en) Method and system for monitoring livestock inside a building
GB2574114A (en) Reduction of theft in a venue based on a dynamically updated high-shrink item library
WO2012115558A1 (en) Apparatus and method for tracking a stabled animal
SE541890C2 (en) Systems and methods for identifying individual animals in a group of animals
US20220067377A1 (en) Device for managing environment of breeding farm
CN109709546B (en) Pet state monitoring method and device
JP2010133964A (en) Sound source position estimation system
CN104200248A (en) Cow occupation recognition method based on multiple radio frequency identification device (RFID) antennas
US10319204B1 (en) Systems and methods for retracing shrink events
KR20210004184A (en) Domestic Animal Managing Method Based on Identifying Marking Information of And Computing Apparatus and Server of Thereof

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 11859170

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 11859170

Country of ref document: EP

Kind code of ref document: A1