WO2005070326A1

WO2005070326A1 - System and process for determining whether an animal is in oestrus

Info

Publication number: WO2005070326A1
Application number: PCT/AU2005/000072
Authority: WO
Inventors: Gregory Lynn Willis
Original assignee: Clarencew Pty Ltd
Priority date: 2004-01-21
Filing date: 2005-01-21
Publication date: 2005-08-04

Abstract

An oestrus determination system, including one or more sensors (118, 120, 122, 124) for generating sensor data representative of one or more behaviours of an animal during confinement; an interface (110) for receiving said sensor data from said one or more sensors; one or more processors (104) for processing said sensor data; non-volatile storage means for storing said sensor data; and one or more oestrus determination modules (114) for causing at least one of said one or more processors to process said sensor data to provide an indication of whether said animal is in oestrus. The sensor data can include activity data representative of activity of said animal during confinement.

Description

SYSTEM AND PROCESS FOR DETERMINING WHETHER AN ANIMAL IS IN OESTRUS

FIELD OF THE INVENTION

The present invention relates to a system and process for determining whether an animal is in oestrus. The present invention also relates to a system and process for predicting whether an animal is in oestrus.

BACKGROUND

To ensure efficient management of breeding animals, it is important to be able to determine or predict when female members of an animal population are entering or are in oestrus so that they can be separated from other members of the herd and inseminated. Existing methods for determining or predicting oestrus are typically invasive, cumbersome and/or unreliable. For example, in the case of dairy cows, it has been discovered that cows grazing in paddocks exhibit increased physical activity before and during oestrus. Such increases in activity can be monitored by hanging a pedometer around a cow's ankle or neck, and reading the pedometer value to provide a measure of the cumulative physical activity of the cow since the pedometer value was last read or zeroed. Such measures are, however, clumsy, time consuming and/or unreliable, often giving false positives. In particular, they are not usually able to detect a cow in "silent oestrus", where at least some indicators of oestrus are either reduced in strength or are not exhibited at all.

It is desired, therefore, to provide a system and process for determining or predicting whether an animal is in oestrus, and an oestrus determination system and process that alleviate one or more difficulties of the prior art, or at least that provide a useful alternative. SUMMARY OF THE INVENTION

In accordance with the present invention, there is provided an oestrus determination system, including: one or more sensors for generating sensor data representative of one or more behaviours of an animal during confinement; an interface for receiving said sensor data from said one or more sensors; one or more processors for processing said sensor data; non-volatile storage means for storing said sensor data; and one or more oestrus determination modules for causing at least one of said one or more processors to process said sensor data to generate an indication of whether said animal is in oestrus.

The present invention also provides a system for determining whether an animal is in oestrus, including: signal generation means for generating one or more signals representative of respective behaviours of said animal during confinement; and signal processing means for processing said signals to determine whether said animal is in oestrus.

The present invention also provides an oestrus determination process in a data processing system, the process including: receiving, at one or more inputs of said data processing system, sensor data generated by one or more sensors, said sensor data representative of one or more behaviours of an animal during confinement; and causing at least one processor of said data processing system to process said sensor data to provide an indication of whether said animal is in oestrus. The present invention also provides a process for determining whether an animal is in oestrus, including: monitoring one or more behaviours of said animal during confinement; and determining whether said animal is in oestrus on the basis of said one or more behaviours.

The present invention also provides a process for determining whether an animal is in oestrus, including: generating one or more signals representative of respective behaviours of said animal during confinement; and processing said one or more signals to determine whether said animal is in oestrus.

The present invention also provides a process for predicting whether an animal is in oestrus, including: monitoring one or more behaviours of said animal during confinement; and predicting whether said animal is in oestrus on the basis of said one or more behaviours.

BRIEF DESCRIPTION OF THE DRAWINGS

Preferred embodiments of the present invention are hereinafter described, by way of example only, with reference to the accompanying drawings, wherein: Figure 1 is a schematic plan view of a preferred embodiment of an oestrus determination system in use with an animal confined within a milking bail; Figure 2 is a flow diagram of an oestrus determination process of the system; Figures 3 and 4 are graphs of milk production as a function of time before, during, and after oestrus; Figure 5 is a bar graph of food spillage measured for six consecutive days, corresponding to days before, during, and after oestrus; Figures 6 to 9 are graphs of milk production over a one month period, for morning milking, afternoon milking, and total milk production each day; and Figure 10 is a set of four graphs of behaviour variables related to oestrus over a period of four days, corresponding to days before, during, and after oestrus.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

As shown in Figure 1, an oestrus determination system includes a data processing unit 102, and sensors 118 to 124, comprising a motion detector 118, a temperature sensor 120, a milk measurement device 122, and a food spillage detector 124. The data processing unit 102 includes a central processing unit (CPU) or processor 104, random access memory (RAM) 106, oestrus determination modules 114, a database 116, and a sensor interface 110, interconnected by a system bus 112. The oestrus determination system uses an oestrus determination process, as shown in Figure 2, to determine or predict whether an animal is in oestrus, or at least is about to be in oestrus.

In the described embodiment, the data processing unit 102 is a standard computer system such as an Intel® IA-32 personal computer system, and the oestrus determination modules 114 are software modules stored on non-volatile (e.g., hard disk) storage 108 of the personal computer 102 and executed by the CPU 104. However, it will be apparent to those skilled in the art that at least parts of the oestrus determination process can be alternatively implemented by dedicated hardware components, such as application-specific integrated circuits (ASICs).

The temperature sensor 120 is an infrared temperature sensor that is directed towards the rear of the animal 114 to monitor its temperature. Alternatively, a thermal imaging camera can be used, or a digital thermometer can be used to measure the temperature of milk produced by the animal 114. The motion detector 118 is a standard motion detection camera such as a Panasonic WV-BPR550/554 digital video camera with a built-in digital motion detector. However, it will be apparent that the motion detection could alternatively be performed within the data processor unit 102 by the CPU 104 in conjunction with the oestrus determination modules 114. Optionally, the system can also include an infrared beam detector (not shown), such as an FDB030 twin beam break detector, as described at http://www.farnell.com. The infrared beam detector can monitor movement of the animal 114 based on the resulting blocking and/or unblocking of infrared beams generated by infrared sources (light-emitting diodes) and detected by infrared sensors mounted in the infrared beam path. However, it will be apparent that a variety of alternative motion detection sensors, including microwave (Doppler) sensors, could be used, either alternatively, or in conjunction with, any of the sensors 118 to 124 described above.

The food spillage detector 124 includes a food collection area 125 located below a food hopper 130. Any food spilled from the food hopper 130 falls onto the collection area 125 where it is channelled into a collection vial (not shown). The collection vial is mounted on a standard electronic weighing device such as a commercial strain gauge scale (for example, the electronic scales with RS-232 interfaces manufactured by Lutron Electronic Enterprise Co. Ltd. and described at http://www.lutron-electronic.com) so that the weight of spilled food can be determined automatically and communicated to the data processing unit. In an alternative embodiment, the volume of food spilled is determined optically using a video camera or alternatively an array of light omitting diodes (LEDs) arranged on one side of the collection vial and corresponding detectors on the opposite side of the vial.

The sensor interface 110 of the data processing unit 102 is a standard interface card providing analog and digital input ports and an analog-to-digital converter (ADC), such as a National Instruments PCI-6104 16-bit multi-function data acquisition PCI card. This allows the electronic signals from the sensors 118 to 124 to be received and processed by the data processing unit 102. However, if any of the sensors 118 to 124 includes a standard computer interface such as a serial (RS-232), universal serial bus (USB), or IEEE 1394 (Firewire) interface, the sensor can be connected directly to a corresponding interface of the computer system 102, which may be provided on the computer's motherboard, or, if not available on the motherboard, on a standard PCI interface card plugged into the computer's motherboard. As shown in Figure 2, the oestrus determination process begins at step 202 when the cow 114 enters the milking bail 116 in order to be milked and is thus confined within the bail 116. At step 204, the individual cow 114 is identified based on an identification tag 126 attached to the cow's ear. The identification tag 126 provides an identification number that uniquely identifies the particular cow 114 from other cows. The identification tag 126 is a radio frequency identification (RFID) tag that is automatically read by a radio frequency sensor (not shown) as the cow 114 enters the bail 116. The resulting identification number is then sent to the data processing unit 102 via a cable (not shown). In an alternative embodiment, the identification number of the tag 126 is read by a barcode reader passed over the tag 126 by a human. In a further alternative embodiment, the tag 126 is automatically read by a second video camera mounted near the cow's head, and determined using optical character recognition or barcode recognition technology.

At step 206, the identification number is used to search the database 116 in order to retrieve previously stored data for this particular cow 114. At step 208, the milking cups are placed on the cow's udder, and the milking process begins. During the milking process, the cow's milking behaviour, i.e., the volume of milk produced, is periodically monitored at step 210. For example, Figure 3 is a graph of the cumulative volume of milk produced by the cow 114 during the morning milking session, sampled at 15 second intervals. The line 302 joining the open circles represents the milk production of the cow on the day before oestrus, whereas the line 304 joining the solid black circles represents the milk production by the same cow on the day after oestrus. In this case, the pre and post oestrus milk production of the cow are substantially similar. In contrast, the milk production of the cow during oestrus, represented by the lower line 306 joining the grey shaded circles, is significantly lower, and the gradient of the line 306, corresponding to the milk production rate, is approximately half the gradient of the pre and post oestrus lines 302, 304. Figure 4 is a graph of the same variables for a different month. Once again, the pre-oestrus data set 402 is almost the same as the post- oestrus data set 404, whereas the milk production during oestrus 406 is once again significantly lower, in this case less than one-third of the rate of milk production at other times of the cow's menstrual cycle. These data sets are typical of the milking behaviour of many cows, and thus the average rate of milk production and/or the total milk production or yield can be used as indicators to predict whether the cow 114 is in oestrus.

Moreover, close inspection of the data in Figures 3 and 4 indicates that other characteristics of milk production are also dependent upon whether the cow 114 is in oestrus. In particular, cows in oestrus typically provide milk at a low, substantially constant rate over the entire duration of the milking session, whereas at other times the rate of milk production is typically begins at a substantially higher rate that is maintained for the majority of the milking session, but decreases near the end of the session. For example, in Figure 3, the pre-oestrus rate of milk production 302 is constant over the first three minutes of milking, and then decreases rather suddenly to a substantially lower rate (similar to the rate at oestrus) that is maintained over the final minute of the session. This trend is also observed in the data of Figure 4, where the pre-oestrus rate of milk production 402 decreases suddenly at around 2.5 minutes, once again to a rate that is close to that observed when the same cow is in oestrus, as shown by the grey shaded line 406. Consequently, a decrease in the rate of milk production towards the end of a milking session suggests that the cow is not in oestrus, particularly when the initial rate is relative high. Conversely, a low, substantially constant rate of milk production over the entire milking session suggests that the cow is in oestrus. Accordingly, a curve joining milk production data points showing this characteristic is referred to as an "oestrus flat curve."

Figure 6 is a graph of the milk produced by a particular cow each day over a time period of one month. The graph shows the variation of morning milking 702, afternoon milking 704, and the resulting total milk production 706 for each day. The total milk production 706 is characterised by several local minima, with the overall minimum occurring on the fourth day of the month, where the total milk production falls from around 11 litres to only 8 litres, suggesting that the cow 114 may be in oestrus on that day. However, in this particular instance, the actual day of oestrus occurred two days later, on the sixth day of the month, as indicated by the vertical dash line 708 labelled "OE". The actual day of oestrus was determined by conventional methods, such as observing that the cow was mounted by a bull on that day. Returning to Figure 2, in addition to monitoring milk production, the cow's movement or activity behaviour during confinement is also monitored (step 212). The movement is monitored using the motion detection camera 118 and, if present, the infrared beam detector 120, to detect muscular-skeletal activity of the cow 114 whilst confined within the bail 116. The motion detector 118 detects motion by intensity changes between adjacent pixels that occurs when the cow 114, which almost fills the field of view of the camera 118, moves about within the bail 116 during the milking process. When movement is detected, the motion detector 118 generates a six volt square-wave pulse signal that is sent to the computer 102 via a cable connecting the motion detector 118 to the interface 110 of the data processing unit 102.

During each milking session, the six-volt pulses generated by the motion detection camera 118 are counted during each milking session. The resulting number of motion detection events for each session is shown adjacent to the corresponding milk production data point in Figure 6. The values shown adjacent to the data points for the total milk production curve 706 are provided by summing the corresponding values for the morning and afternoon milking sessions. Although the motion or activity of the cow 114 in the bail 116 during milking shows some variability from day-to-day, it is characterised by a significant increase on the day prior to oestrus, followed by a significant decrease on the actual day of oestrus. For example, considering the motion data adjacent the morning session data points 702, the activity readings during the morning milking sessions are 72, 63, 64 and 53 on the first four days of the month, respectively, indicating relatively constant activity within the bail from day to day. On the fifth day, the activity increases sharply to a value of 393, followed by a dramatic decrease to a value of only 21 on the day of oestrus. This characteristic change in activity or movement behaviour is referred to herein as "delta motion".

This decrease in activity during oestrus may appear surprising, because increases in activity have previously been exhibited by cows in paddocks before and during oestrus. The increased activity of a cow in a paddock can be explained as a behaviour likely to attract a bull so that the cow can be mounted. However, once mounted, the cow needs to stand relatively stationary in order for the bull to successfully mate with the cow. The placement of a cow in a confined space, such as the milking bail 116, simulates the physical confinement of the cow by the bull when being mounted. Thus although a cow in oestrus displays increased activity in an open space such as a paddock, a cow in oestrus exhibits decreased activity when confined. Prior to oestrus, the compulsion to be active is dominant. The cow therefore demonstrates increased activity in the milking bail 116 on the day before oestrus. However, once in oestrus, the cow is ready to be mounted, and the confined cow exhibits a marked decrease in activity. This behaviour has been observed in a number of cows over many months, allowing the reduced activity of a confined cow to be used as a fairly reliable indicator of oestrus. Moreover, this characteristic behaviour is also observed when a cow is confined by any means, and does not require that the cow be simultaneously milked. However, a milking session provides a convenient opportunity for a cow's behaviour and other characteristics to be monitored at a time when the cow is already confined, and thus monitoring during milking provides the most efficient and commercially viable process for determining whether a cow is in oestrus.

However, although in this particular instance, the detection of reduced activity following a day of increased activity coincided with oestrus, it has been found that there is not always a perfect correlation between such behaviour and oestrus in any given cow. Moreover, as described further below, it has also been found that no single behaviour or other monitored characteristic of a cow or other animal provides an absolutely reliable indicator of oestrus. However, the monitoring of more than one behaviour, or the monitoring of at least one behaviour in conjunction with at least one other monitored characteristic (e.g., temperature) of the animal have been found to provide a more reliable indicator of oestrus than any one behaviour or other characteristic alone, as described below.

The muscular-skeletal activity of the cow 114 can also be monitored independently by the infrared beam detector described above, if present. Returning to Figure 6, the number of infrared beam blocking/unblocking events due to motion of the cow 114 during milking is shown beneath the data points for total milk production 706. The number of beam breaking/unbreaking events increases from a value of 60 to a value of 94 on the day prior to oestrus, falling to a value of 84 on the day of oestrus. In general, it was found that a decreased number of beam breaking/unbreaking events was measured on the day of oestrus. Thus the infrared beam detector can be used as an alternative method of monitoring the movement of the cow during milking. However, the blocking and/or unblocking of infrared beams provides a less reliable measure of muscular-skeletal activity than the motion camera 118, presumably because a significant amount of activity by the cow may not be detected while the infrared beam remains in a blocked or unblocked state due in part to the relatively small number of beams. Accordingly, the motion detector camera 118 is preferred for monitoring the activity of the cow during milking.

The decreased activity described above is also reflected in the cow's eating behaviour. During milking, the cow 114 is fed from the food hopper 130. As the cow 114 raises its head from the food hopper 130 during feeding, a portion of the food lifted from the food hopper 130 by the cow 114 is spilled onto the collection area 125, from which it is channelled into a collection vial (not shown). As described above, the collection vial is mounted on an electronic weighing device, allowing the weight of spilled food to be monitored as a measure of muscular-skeletal activity.

After finishing the milking session at step 214, the body temperature of the cow is monitored by the temperature sensor 120 at step 215. The amount of food spillage during the milking session is determined at step 216 by reading the output signal generated by the food spillage detector 124 and sent to the interface 110 of the data processing unit 102 via the connection cable 132. The amount of food spillage during the milking session is then compared with the amount of food spillage by the same cow over the preceding days, the latter having been previously retrieved from the database 116 at step 206. Returning to Figure 6, the measured amount of food spillage for each day around oestrus is shown in the dashed box 710 above the total milk production data 706. From the second to fourth days of the month, the amount of food spillage was measured as 164, 156, 208, and 154, respectively. However, on the day of oestrus, the amount of food spillage decreased dramatically to a value of only 94, followed by an increase to 127 on the day following oestrus.

Figure 5 is a bar graph of a different set of food spillage data for six consecutive days. The day of oestrus is the fourth day, and the food spillage value of 84 on this day is the lowest value. Note, however, that the food spillage on the day before oestrus is 96, which, although higher than on the day of oestrus, is still significantly lower than the values for the preceding days. If food spillage was the only monitored behaviour, the day before oestrus could have been thought to be the day of oestrus. This highlights the importance of monitoring multiple behaviours or other possible indicators of oestrus.

Figure 7 shows the same variables as Figure 6, but for the following month. In this particular instance, it can be seen that the total milk production data 802 does not identify any particular day as being the day of oestrus, which is indicated by the vertical line 804. Indeed, it can be seen that the total amount of milk production rises slightly on the day of oestrus relative to the day before and the day after oestrus. However, the activity measured by the motion detector camera rises sharply to a value of 815 on the day prior to oestrus, falling to a value of only 120 on the day of oestrus. On the day following oestrus, a value of 450 was measured. Thus the day of oestrus is correctly predicted by a sharp rise followed by a sharp fall in the activity (i.e., delta motion) of the cow during milking in the bail 116. The day of oestrus is also correctly predicted by the food spillage data 808, where the spillage fell from values of 222 and 224 on the two days before oestrus, to a value of 157 on the day of oestrus.

Figure 8 shows the same variables for a different cow, with the day of oestrus falling on the 16^th day of the month, as indicated by the vertical line 902. Once again, the activity of the cow rises from a value of 348 to a value of 955 on the day prior to oestrus, falling to a value of 343 on the day of oestrus. Note also in this case that the total milk production data 904 shows a dip on the day of oestrus, in agreement with the activity or motion data. Figure 9 shows the same variables plotted for a different month for the same (second) cow. However, in this case the line 1104 representing the number of motion detection events indicates a rapid increase in activity (from 60 to 108 motion detection events) on the 19^th day, followed by a rapid decrease in activity (from 108 to 41) on the 20^th day. On its own, this observation would suggest that oestrus is occurring on the 20^th day, whereas in fact it occurred on the 22^nd day, as indicated by the vertical dashed line 902. Similarly, the marked decrease in milk production data 904 also suggests that oestrus may be occurring on the 20^th day. This data set was selected to demonstrate that, although the process described above usually predicts the correct day of oestrus, this may not always be the case for some animals, due to biological variations. However, even in such a worst case scenario, the process identifies days very close to oestrus (in this case 3 days before and 1 day after). Thus the process provides data that, if not correctly determining the precise day of oestrus, at least indicates that the animal should be removed in preparation for mating, or should be closely monitored for other signs of oestrus over the following few days.

At step 218, the milk production data, the total food spillage data, the movement data, and the temperature data are stored in the database 116. At step 220, the temporal characteristics of each data set are analysed by comparing the newly added data for the just completed milking session to the corresponding data for the preceding days. At step 222, a test is performed to determine whether one or more of the data sets shows a significant dip. That is, whether the milk production, motion detection, or food spillage data has decreased substantially below the lowest value determined over the preceding (non-oestrus) time period. This time period can be configured by the user, but is typically around 7 days. The comparison can also be made relative to the data generated in previous months or even previous years, providing that the data relates to the same cow. In the case of motion detection monitored by the motion detector camera 118, this analysis includes determining whether the activity measured for the previous day is substantially greater than the activity measured on both the current day and also two days earlier; i.e., whether the motion is characterised by a maximum followed by a minimum. For example, Figure 10 shows four time-aligned graphs of total milk production 1102, motion events 1104 generated by the motion detection camera 118, beam blocking/unblocking events 1106 generated by the infrared beam detector 120, and weight of food spilled 1108 generated by the food spillage detector 124. The day of oestrus is indicated by the vertical line 1110. On the second day (the day before oestrus), the food spillage data 1108 has decreased significantly, from a value of 240 to a value of 96, suggesting that the cow 114 may be in oestrus. However, none of the other data sets 1102 to 1106 have similar decreases in their respective values. On the third day, however, all four behaviours 1102 to 1108 have decreased significantly below the values measured on the preceding days. Note also that the motion detector data 1104 includes the characteristic increase on the previous (2^nd) day. Consequently, the system can determine with a high degree of accuracy that the cow 114 is in oestrus on the third day.

If none of the above criteria are met, then the process ends. Alternatively, a message indicating that the cow 114 is not in oestrus can be displayed if desired. Otherwise, if one or more of the monitored behaviours do meet the appropriate criteria, as described above, then a message is displayed at step 224, indicating that the cow 114 may be in oestrus. The message includes a numeric confidence value generated by comparing the new data with correlations between existing data measured over previous months and years and actual days of oestrus, providing an indication of the likelihood that the cow 114 is in fact in oestrus. Optionally, time-aligned graphs of the stored data, similar to those shown in Figure 10, can be displayed to an operator, providing an opportunity to review the temporal characteristics of each data set and the correlation of any minima. After displaying the message and any graphs, the process then ends. The collection vial is then emptied, and the cow 114 is removed from the bail 116. If the message indicates a high probability that the cow 114 is in oestrus, the cow 114 is taken away for insemination. The cow 114 can be drafted out manually or with an automatic drafting system. The next cow (if any) is then brought into the bail 116 for milking.

The oestrus determination system thus provides a convenient and reliable means for determining when a cow is in oestrus. The system determines this automatically, without requiring manual intervention. In addition to the monitoring of the specific behaviours described above, additional behaviours and/or variables can be used to determine oestrus. For example, the body temperature of the cow 114 was monitored at step 215. Because a cow's temperature typically increases on the day of oestrus, the cow's temperature data can be combined with one or more monitored behaviours to determine whether the cow 114 is in oestrus.

It will be apparent that one or more of the sensors 118 to 124 can be added to an existing milking bail to monitor one or more behaviours or characteristics of the cow 114 as described above.

The system can be used in robotic, rotary, herringbone, and other types of dairy. In a rotary dairy, the motion detector 118 and infrared detector 120 are mounted on a robotic platform in order to track the rotation of the dairy and thereby monitor a particular cow for a fraction of the total milking period for that cow before moving to monitor another cow. Alternatively, individual cameras can be provided for each bail or a number of bails on the rotary platform. Similarly, in a herringbone dairy, the sensors 118, 120 can be mounted on a track to follow each cow after cup placement for a predetermined period of time. In a robotic dairy, the sensors 118 to 124 are provided for each individual cow.

Although the oestrus determination system has been described above in terms of determining oestrus in a cow, the system can also be to determine oestrus in other animals, such as goats, sheep, water buffalos, horses, or alpacas, for example.

Many modifications will be apparent to those skilled in the art without departing from the scope of the present invention as herein described with reference to the accompanying drawings.

Claims

CLAIMS:

1. An oestrus determination system, including: one or more sensors for generating sensor data representative of one or more behaviours of an animal during confinement; an interface for receiving said sensor data from said one or more sensors; one or more processors for processing said sensor data; non-volatile storage means for storing said sensor data; and one or more oestrus determination modules for causing at least one of said one or more processors to process said sensor data to generate an indication of whether said animal is in oestrus.

2. A system as claimed in claim 1, wherein said one or more behaviours include one or more of muscular-skeletal activity, food spillage, and milk production.

3. A system as claimed in claim 1, further including one or more additional sensors for generating additional sensor data representative of one or more characteristics of said animal.

4. A system as claimed in claim 3, wherein the one or more additional sensors includes a temperature sensor for generating temperature data representative of a temperature of said animal, said sensor data including said temperature data.

5. A system as claimed in claim 1, wherein said indication is generated on the basis of a comparison of said sensor data with stored sensor data representative of said one or more behaviours during one or more previous confinements.

6. A system as claimed in claim 1, wherein said indication is generated on the basis of correlation of changes in at least two of said behaviours.

7. A system as claimed in claim 1, wherein said indication is generated on the basis of correlation of changes in at least one of said behaviours and at least one characteristic of said animal.

8. A system for determining whether an animal is in oestrus, including: signal generation means for generating one or more signals representative of respective behaviours of said animal during confinement; and signal processing means for processing said signals to determine whether said animal is in oestrus.

9. A system as claimed in claim 8, wherein said confinement includes confinement during milking of said animal.

10. A system as claimed in claim 8, wherein said signal generation means includes a movement sensor for generating a signal representative of movement of said animal during confinement.

11. A system as claimed in claim 8, wherein said signal generation means includes a food spillage sensor for generating a signal representative of an amount of food spilled by said animal during confinement.

12. A system as claimed in claim 8, wherein said signal generation means includes a milk production sensor for generating a signal representative of milk production by said animal.

13. A system as claimed in claim 8, wherein said signal generation further includes means for generating one or more signals representative of one or more characteristics of said animal.

14. A process as claimed in claim 8, wherein said one or more characteristics includes a temperature of said animal.

15. A system as claimed in claim 8, wherein said animal includes a cow, goat, sheep, water buffalo, horse, or alpaca.

16. An oestrus determination process in a data processing system, the process including: receiving, at one or more inputs of said data processing system, sensor data generated by one or more sensors, said sensor data representative of one or more behaviours of an animal during confinement; and causing at least one processor of said data processing system to process said sensor data to provide an indication of whether said animal is in oestrus.

17. A process as claimed in claim 16, including causing at least one processor of said data processing system to generate display data representing said indication; and displaying said display data on a display of said data processing system.

18. A process as claimed in claim 16, wherein the sensor data includes one or more of activity data representative of muscular-skeletal activity of said animal, milk production data representative of milk production of said animal, and food spillage data representative of food spillage by said animal.

19. A process as claimed in claim 18, wherein the sensor data further includes temperature data representative of a temperature of said animal.

20. A process as claimed in claim 16, wherein the processing of said sensor data includes: retrieving, from non-volatile storage means associated with said data processing system, stored sensor data representative of said one or more behaviours during one or more previous confinements; and wherein the indication of whether said animal is in oestrus is determined on the basis of a comparison of the sensor data with the stored sensor data.

21. A process for determining whether an animal is in oestrus, including: monitoring one or more behaviours of said animal during confinement; and determining whether said animal is in oestrus on the basis of said one or more behaviours.

22. A process as claimed in claim 21, wherein said monitoring includes monitoring said one or more behaviours during milking of said animal.

23. A process as claimed in claim 21, wherein said one or more behaviours include one or more of movement, food spillage, and milk production.

24. A process as claimed in claim 23, wherein said movement includes muscular-skeletal activity.

25. A process as claimed in claim 23, wherein said one or more behaviours includes one or more of a rate of milk production, a change in said rate of milk production, and total milk production during a milking session.

26. A process as claimed in claim 21, wherein said one or more behaviours includes one or more of movement, food spillage, and milk production; and wherein said determining includes determining that said animal is in oestrus on the basis of a substantial decrease in one or more of said behaviours.

27. A process as claimed in claim 21, wherein said determining is based on correlation of changes in at least two of said behaviours.

28. A process as claimed in claim 21, wherein said determining is based on correlation of changes in at least one of said behaviours and at least one characteristic of said animal.

29. A process as claimed in claim 21, wherein said determining is based on correlation of substantial decreases in at least two of movement, food spillage, and milk production.

30. A process as claimed in claim 21, wherein said determining includes determining that said animal is in oestrus on the basis of a substantial decrease in movement of said animal following a substantial increase of said movement.

31. A process as claimed in claim 30, wherein said substantial decrease in movement and said substantial increase in movement are determined relative to a preceding degree of movement.

32. A process as claimed in claim 31, wherein said preceding amount of movement, said substantial increase in movement, and said substantial decrease in movement are determined on successive days.

33. A process as claimed in claim 21, wherein said determining includes determining whether said animal is in oestrus on the basis of said one or more behaviours and one or more other characteristics of said animal.

34. A process as claimed in claim 33, wherein said one or more other characteristics includes a temperature of said animal.

35. A process as claimed in claim 21, wherein said animal includes a cow, goat, sheep, water buffalo, horse, or alpaca.

36. A process for determining whether an animal is in oestrus, including: generating one or more signals representative of respective behaviours of said animal during confinement; and processing said one or more signals to determine whether said animal is in oestrus.

37. A process as claimed in claim 36, wherein said generating includes generating said one or more signals representative of said respective behaviours during milking of said animal.

38. A process as claimed in claim 37, wherein said one or more behaviours includes one or more of movement, food spillage, and milk production.

39. A process for predicting whether an animal is in oestrus, including: monitoring one or more behaviours of said animal during confinement; and predicting whether said animal is in oestrus on the basis of said one or more behaviours.

40. An oestrus determination system having components for executing the steps of any one of claims 16 to 39.

41. A computer-readable storage medium having stored thereon program instructions for executing the steps of any one of claims 16 to 39.