Search

CN-117479832-B - Monitoring system for individual growth monitoring of livestock animals

CN117479832BCN 117479832 BCN117479832 BCN 117479832BCN-117479832-B

Abstract

A growth monitoring system and method for livestock animals includes a floor-padded shed environment, a 3D camera having a downward field of view in which an animal may appear during a visit to repeatedly obtain a 3D image of the field of view, and an image processor that extracts a 3D point cloud representation. The image processor detects whether an animal is present in the point cloud representation and if so, selects an animal part of the point cloud representation, and determines position-dependent first height information, i.e. the height of the animal part of the point cloud above the floor, and determines the position of the front and/or rear legs of the animal, and if not, determines position-dependent second height information, i.e. the height of the litter above the floor. The image processor determines the height of the animal at visit based on the first height information, the second height information, and the position of the front leg and/or rear leg. This provides reliable height measurements in the natural environment of the animal and requires no physical effort.

Inventors

  • I. E.M. Deng Youjie Er
  • M. Zuluowa
  • T. sola

Assignees

  • 莱利专利股份有限公司

Dates

Publication Date
20260505
Application Date
20220711
Priority Date
20210716

Claims (17)

  1. 1. A growth monitoring system for monitoring growth of an individual livestock animal having a front leg and a rear leg, comprising: -a shed environment in which one or more such livestock animals are able to walk and on whose floors are padded; -a 3D camera having a downward field of view comprising a portion of the shed environment where individual livestock animals may be present during a visit, and the 3D camera being arranged to repeatedly obtain a 3D image of said field of view; an image processor arranged to process the obtained image to extract a 3D point cloud representation of the field of view from said processed obtained image, -Wherein the image processor is arranged to detect the presence or absence of one of the livestock animals in the 3D point cloud representation, and If so, selecting an animal part point cloud of the point cloud representation and determining position-dependent first height information, i.e. the height of the animal part point cloud of the point cloud representation above the floor, and determining the position of the front and/or rear legs of the individual livestock animal from the animal part point cloud, If not, determining position-dependent second height information, i.e. the height of the mat above the floor, Wherein the image processor is arranged to determine the height of said individual livestock animal at said visit based on said first height information, said second height information and said position of said front leg and/or said rear leg.
  2. 2. The system of claim 1, wherein the livestock animal comprises a cow.
  3. 3. The system of claim 1, wherein the image processor is arranged to determine the position-dependent second height information by subdividing the field of view into a predetermined number of subfields, determining height information for each subfield, and determining the height of the livestock animal based on the second height information at the positions of the front leg and/or the rear leg, respectively, at the subfields.
  4. 4. The system of claim 1, wherein the image processor is arranged to subtract the second height information at the position of the rear leg and/or the front leg from the first height information at the position of the rear leg and/or the front leg, respectively, to obtain a corrected rear leg height and/or front leg height.
  5. 5. The system of any one of claims 1 to 4, wherein the image processor is arranged to determine an interpolated height of the second height information between the position of the front leg and the position of the rear leg and to subtract the interpolated height from the animal part point cloud to form corrected animal height information and to determine said height of the livestock animal from the corrected animal height information.
  6. 6. The system of claim 5, wherein the interpolation height comprises a linear interpolation height.
  7. 7. The system of any one of claims 1 to 4, wherein the 3D camera is arranged to repeatedly obtain 3D images, and wherein the image processor is arranged to determine the second height information from 3D images obtained immediately after the visit and/or from 3D images obtained between the visit and a last visit of the livestock animal.
  8. 8. The system of claim 7, wherein the 3D camera is arranged to repeatedly obtain 3D images each time the livestock animal visits.
  9. 9. The system of claim 7, wherein the image processor is arranged to determine said second height information from a 3D image obtained immediately after said visit and/or from a 3D image obtained between said visit and a last visit of any of said livestock animals.
  10. 10. The system of any one of claims 1 to 4, wherein the image processor is arranged to determine a body contour of the livestock animal from the animal part point cloud and to determine the position of the rear leg and/or the front leg from the body contour.
  11. 11. The system of any one of claims 1 to 4, wherein the image processor is arranged to evaluate an animal pose of the livestock animal from the animal part point cloud and discard the animal part point cloud if the animal part point cloud does not meet a pose usefulness index.
  12. 12. The system of claim 11, wherein the gesture usefulness index comprises that a highest portion of the animal portion point cloud is located in a rear portion of the livestock animal.
  13. 13. The system of claim 12, wherein the rear portion comprises a rear half of the livestock animal.
  14. 14. The system of claim 11, wherein the gesture usefulness index comprises that the animal portion point cloud has a degree of symmetry about its long axis at least as high as a symmetry threshold.
  15. 15. The system of any one of claims 1 to 4, further comprising animal identification means arranged to determine the identity of the individual livestock animal at the time of the visit.
  16. 16. A method for monitoring the growth of individual livestock animals having front and rear legs in a shed environment in which one or more such livestock animals are capable of ambulating and the floor of the shed environment is padded, the method comprising: -repeatedly obtaining a 3D image of a field of view by a 3D camera having said field of view, the field of view comprising a portion of the shed environment where individual livestock animals may be present during a visit; -processing the obtained image by an image processor, wherein the processing comprises Extracting a 3D point cloud representation of the field of view, Detecting whether one of the livestock animals is present in the point cloud representation, If yes, then -Selecting an animal part point cloud of the point cloud representation; determining position-dependent first height information, i.e. the height of the animal part point cloud above the floor, and Determining the position of the front and/or rear legs of the individual livestock animal from the animal part point cloud, If not, then -Determining position-dependent second height information, i.e. the height of the padding above the floor; -determining the height of the individual livestock animal at the visit from the first height information, the second height information and the position of the front leg and/or the rear leg.
  17. 17. The method of claim 16, wherein the livestock animal comprises a cow.

Description

Monitoring system for individual growth monitoring of livestock animals The present invention relates in a first aspect to a growth monitoring system for monitoring growth of an individual animal and in a second aspect to a method for monitoring growth of an individual animal. Monitoring growth helps to assess the health of the animal and generally helps to manage the animal. For example, determining when certain actions should be taken may depend on the size of the animal. Examples of calves are weaning, first insemination, whereas for meat animals the slaughter time is usually dependent on weight or size. Monitoring growth requires the ability to consistently determine the size of the animal. For most livestock animals, and of course young animals, growth almost always occurs in three spatial dimensions (and body weight). For this reason, the invention focuses on height as the selected dimension for monitoring growth. While it is not uncommon to use a weighing system to monitor growth, the accuracy and usefulness of weight measurements is not ideal. For example, the content weight of the gastrointestinal tract and bladder may vary significantly. Methods of measuring the size of livestock animals are less common. This almost always requires physical labor to provide useful measurements. Document NL2011952A1 discloses a 3D camera system for monitoring the growth of an animal, which camera system is arranged to determine volume-related parameters, such as various height parameters, and preferably also the weight. The system is used in a venue with a pen to limit movement of animals and is arranged to determine the height of the animals relative to a floor or another horizontal surface and may be used, for example, daily or weekly. In practice, it has been found that these measurements are often not reliable enough, while measuring animals in a confined environment can lead to an agitation situation. Furthermore, especially for young animals, the risk of agitation should be avoided as much as possible, so that often transferring the (young) animal into a confined environment is not a good choice. The object of the present invention is to solve the above-mentioned drawbacks. It is a particular object of the present invention to provide a system for monitoring the growth of livestock animals which provides more accurate measurements while causing less fidgeting on the animals and without causing a lot of physical effort. The invention achieves these objects at least partly by a growth monitoring system, in particular for monitoring the growth of individual livestock animals, such as cows, having front and rear legs, comprising a shed environment in which one or more such livestock animals may walk and which is padded on the floor, a 3D camera having a field of view comprising a portion of the shed environment in which the individual livestock animals may be present during a visit and which is arranged to repeatedly obtain a 3D image of said field of view, an image processor arranged to process the obtained image to extract a 3D point cloud representation of the field of view from said processed obtained image, wherein the image processor is arranged to detect if one of said livestock animals is present in said 3D point cloud representation and, if so, to select a point cloud of said point cloud representation and to determine position-dependent first height information, i.e. a position of said animal part of said point cloud representation above said animal leg and said height information, wherein said position of said animal part of said point cloud representation is determined on the basis of said height of said animal part of said point cloud and said animal, and if so, and said height information is determined on the basis of said height of said animal part of said point cloud and said animal part of said point cloud. Here, the "position correlation height information" indicates that the height (or distance, rough, unprocessed measurement results) of the measurement varies with position. Furthermore, the location may be considered as a location in the image itself, or preferably a location in the real world, especially a location on the animal. In each case, this will be clear in context. The inventors have realized that if there is a system that is able to make measurements more frequently and that is located in an environment that is more natural to the animal, the system may be better, less fidgety and also more accurate. The system is almost uniformly padded on the floor. Typically, such litter is quite flexible and may be compacted when an animal stands thereon. This results in the height and other characteristics of the litter changing over time, resulting in inaccurate measured animal heights. To address this problem, the system according to the present invention corrects this problem by taking an image of the empty field of view (i.e., without the animal), measuring the height of the litter, and the