System for animal evaluation through image acquisition

• US 5,576,949 A
• Filed: 06/20/1994
• Issued: 11/19/1996
• Est. Priority Date: 08/30/1991
• Status: Expired due to Term

• Current Assignee: Brattain & Asscociates, LLC, Destron Fearing Corporation (Merck & Co., Inc.), GT Acquisition SUB Incorporated (Merck & Co., Inc.)
• Original Assignee: C-Scan, LLC
• Inventors: Engelstad, Mark A., Scofield, Wayne W.
• Primary Examiner(s): Hayes, Gail O.
• Assistant Examiner(s): VERDUN, HAYWARD AUBRY
• Application Number:
  US08/262,234
• Time in Patent Office: 883 Days
• Field of Search: 364/413.13, 364/401, 358/93, 358/107, 358/106, 382/6, 382/162, 382/165, 382/199, 382/201, 382/203, 358/93, 358/107, 358/106, 356/376, 356/374, 356/384
• US Class Current: 702/179
• CPC Class Codes: A01K 1/0613 Devices for immobilisation or restraint of animals, e.g. chutes pens for animals while being shod A01L13/00; appliances for supporting or fettering animals for operative purposes A61D3/00
• CPC Class Codes: A01K 29/005 Monitoring or measuring activity, e.g. detecting heat or mating
• CPC Class Codes: G06T 2207/10012 Stereo images
• CPC Class Codes: G06T 2207/30108 Industrial image inspection
• CPC Class Codes: G06T 7/60 Analysis of geometric attributes

Abstract

The present invention evaluates an animal to determine characteristics or traits thereof. The invention evaluates the animal as it moves through first and second scenes that correspond to different first and second fields of view.

Claims

1. A method for evaluating animals with each such animal during its evaluation moving through first and second scenes corresponding to differing first and second fields of view in a manner so as to partially cover first and second background surfaces provided in each of said first and said scenes, respectively, as viewed in said first and second fields of view, respectively, said first and second background surfaces each having a visual appearance which contrast with a visual appearance of that animal, said method comprising:

acquiring a first scene evaluation image representation of an image of said first scene during movement of an animal undergoing evaluation through said first and second scenes, there being an animal image portion corresponding to that animal in said first scene evaluation image representation and a background image portion in said first scene evaluation image representation corresponding to uncovered portions of said first background surface;

acquiring a second scene evaluation image representation of an image of said second scene during said movement of said animal, there being an animal image portion corresponding to that animal in said second scene evaluation image representation and a background image portion in said second scene evaluation image representation corresponding to uncovered portions of said second background surface;

transforming at least in part one of said first and second scene evaluation image representations by decomposing same into at least a partial spectrum based on selected spectral functions;

locating in that scene evaluation image representation so decomposed a separation location between at least a part of said animal image portion thereof and at least a part of said background image portion thereof; and

forming selected animal indicia for said animal undergoing evaluation based on said locating of said separation location.

2. The method of claim 1 wherein said locating of said separation location in one of said first and second scene evaluation image representations is based on locations in that scene evaluation image representation where corresponding substantial changes of values of spectrum components, found by said transforming, occur.

3. The method of claim 2 wherein said first and second background surfaces each have a periodic pattern thereon.

4. A method for evaluating animals with each such animal during its evaluation moving through first and second scenes corresponding to differing first and second fields of view in a manner so as to partially cover first and second background surfaces provided in each of said first and second scenes, respectively, as viewed in said first and second fields of view, respectively, said first and second background surfaces each having a visual appearance which contrasts with a visual appearance of that animal, said method comprising:

acquiring a first scene evaluation image representation of an image of said first scene at a selected time during movement of an animal undergoing evaluation simultaneously through said first and second scenes, there being an animal image portion corresponding to that animal in said first scene evaluation image representation and a background image portion in said first scene evaluation image representation corresponding to uncovered portions of said first background surface;

acquiring a second scene evaluation image representation of an image of said second scene substantially at said selected time during said movement of said animal, there being an animal image portion corresponding to that animal in said second scene evaluation image representation and a background image portion in said second scene evaluation image representation corresponding to uncovered portions of said second background surface;

locating in said first scene evaluation image representation a first boundary portion between at least part of said first scene image animal image portion and at least part of said first scene image background image portion;

locating in said second scene evaluation image representation a first boundary portion between at least part of said first scene image animal image portion and at least part of said first scene image background image portion;

ascertaining values of a plurality of selected parameters based on said first scene image first boundary portion and said second scene image first boundary portion; and

forming selected animal indicia for said animal undergoing evaluation based on said plurality of selected parameters.

5. The method of claim 4 wherein said method further comprises estimating a value for a selected animal trait based on said selected animal indicia.

6. The method of claim 5 wherein said estimating of said value of said selected animal trait further comprises use of statistical parameters previously determined through performing said method for evaluating animals on other animals and also gauging values of said selected animal trait therefor.

7. The method of claim 6 wherein said ascertaining values of a plurality of selected parameters comprises determining values for those parameters of an animal undergoing evaluation selected from among that animal'"'"'s body length, shoulder width, rump width and rump height.

8. The method of claim 7 wherein said forming selected animal indicia comprises calculating ratios selected from among a shoulder muscle-to-bone ratio determined by a ratio of said values of said shoulder width and rump height, a rump muscle-to-bone ratio determined by a ratio of said values of rump width and rump height, a musculoskeletal development per unit length ratio determined by a ratio of a value of a sum of said shoulder muscle-to-bone and rump muscle-to-bone ratios and said body length value, and a musculoskeletal development per unit height ratio determined by a ratio of a value of a sum of said shoulder muscle-to-bone and rump muscle-to-bone ratios and said rump height value.

9. The method of claim 8 wherein said selected animal trait is selected from among quality of carcass meat, quantity of carcass meat, relative leanness of animal carcass, fertility and feed conversion efficiency.

10. The method of claim 6 wherein separate sets of statistical parameters are used for each animal gender.

11. The method of claim 4 wherein said forming of selected animal indicia comprises calculating selected ratios of said values of said selected parameters.

12. The method of claim 4 wherein said ascertaining values of a plurality of selected parameters further comprises determining values for those parameters of an animal undergoing evaluation based on acquiring a value from a further measurement other than said acquiring of said first and second scene evaluation image representations.

13. The method of claim 12 wherein said forming selected animal indicia comprises calculating ratios selected from among a shoulder muscle-to-bone ratio determined by a ratio of said values of said shoulder width and rump height, a rump muscle-to-bone ratio determined by a ratio of said values of rump width and rump height, a musculoskeletal development per unit length ratio determined by a ratio of a value of a sum of said shoulder muscle-to-bone and rump muscle-to-bone ratios and said body length value, and a musculoskeletal development per unit height ratio determined by a ratio of a value of a sum of said shoulder muscle-to-bone and rump muscle-to-bone ratios and said rump height value.

14. The method of claim 12 wherein said ascertaining values of a plurality of selected parameters comprises determining values for those parameters of an animal undergoing evaluation selected from among that animal'"'"'s body length, shoulder width, rump width, rump height and weight.

15. The method of claim 4 wherein said acquiring of said first scene evaluation image representation at said selected time during movement of an animal undergoing evaluation has said selected time selected by detecting said animal undergoing evaluation having reach a selected location in said first and second scenes.

16. The method of claim 4 wherein said locating of said first scene image first boundary portion further comprises acquiring a first scene background reference image representation of an image of said first scene absent any animal moving through said first scene, and wherein said finding of said second scene image first boundary portion further comprises acquiring a second scene background reference image representation of an image of said second scene absent any animal moving through said second scene.

17. The method of claim 16 wherein said locating of said first scene image first boundary portion further comprises locating in said first scene background reference image representation at least one boundary of a background image portion thereof corresponding to said first background surface, and wherein said locating of said second scene image first boundary portion further comprises locating in said second scene background reference image representation at least one boundary of a background image portion thereof corresponding to said second background surface.

18. The method of claim 4 wherein said acquiring of said first scene evaluation image representation at said selected time during movement of an animal undergoing evaluation is accomplished using a first video camera and a storage and computing means such that a binary representation is stored at least temporarily in said storage and computing means for each corresponding image pixel element in said first scene image sensed by a corresponding image sensing element in said first video camera, and wherein said acquiring of said second scene evaluation image representation at substantially said selected time during movement of an animal undergoing evaluation is accomplished using a second video camera and said storage and computing means such that a binary representation is stored at least temporarily in said storage and computing means for each corresponding image pixel element in said second scene image sensed by a corresponding image sensing element in said second video camera.

19. The method of claim 18 wherein said locating of said first scene image first boundary portion and said locating of said second scene image first boundary portion further comprises at least partially transforming said first and second scene evaluation image binary representations, respectively, by decomposing selected portions of each into at least a partial spectrum based on selected spectral functions.

20. The method of claim 19 wherein said locating of said first scene image first boundary portion and said locating of said second scene image first boundary portion is based on locations in said first and second scene evaluation image binary representations, respectively, where corresponding substantial changes in values of spectral components, found by said transforming, occur between said selected portions thereof.

21. The method of claim 20 wherein said first and second background surfaces each have a periodic pattern thereon.

22. The method of claim 21 wherein said periodic pattern is a repeating stripe pattern, and said transforming is based on Hadamard transforms.

23. The method of claim 4 wherein first background surface has a visual appearance which contrasts with those visual appearances of animals being evaluated by being of a color which differs from those colors of said animals.

24. The method of claim 4 wherein first background surface has a visual appearance which contrasts with those visual appearances of animals being evaluated by having a periodic pattern thereon.

25. The method of claim 24 wherein said periodic pattern is a repeating stripe pattern.

26. The method of claim 4 wherein said locating of said first scene image first boundary portion and said locating of said second scene image first boundary portion further comprises at least partially transforming said first and second scene evaluation image representations, respectively, by decomposing each into at least a partial spectrum based on selected spectral functions.

27. The method of claim 26 wherein said locating of said first scene image first boundary portion and said locating of said second scene image first boundary portion is based on locations in said first and second scene evaluation image representations, respectively, where corresponding substantial changes in values of spectral components, found by said transforming, occur between said selected portions thereof.

28. The method of claim 27 wherein said first and second background surfaces each have a periodic pattern thereon.

29. The method of claim 28 wherein said periodic pattern is a repeating stripe pattern, and said transforming is based on Hadamard transforms.

30. An animal evaluation system for evaluating animals with each such animal during its evaluation moving through first and second scenes corresponding to differing first and second fields of view in a maimer so as to partially cover first and second background surfaces provided in each of said first and second scenes, respectively, as viewed in said first and second fields of view, respectively, said first and second background surfaces each having a visual appearance which contrast with a visual appearance of that animal, said system comprising:

an image acquisition means for acquiring a first scene evaluation image representation of an image of said first scene at a selected time during movement of an animal undergoing evaluation simultaneously through said first and second scenes, there being an animal image portion corresponding to that animal in said first scene evaluation image representation and a background image portion in said first scene evaluation image representation corresponding to uncovered portions of said first background surface, and for acquiring a second scene evaluation image representation of an image of said second scene substantially at said selected time during said movement of said animal, there being an animal image portion corresponding to that animal in said second scene evaluation image representation and a background image portion in said second scene evaluation image representation corresponding to uncovered portions of said second background surface; and

a storage and computing means for locating in said first scene evaluation image representation a first boundary portion between at least part of said first scene image animal image portion and at least part of said first scene image background image portion, and of locating in said second scene evaluation image representation a first boundary portion between at least part of said first scene image animal image portion and at least part of said first scene image background image portion, and for ascertaining values of a plurality of selected parameters based on said first scene image first boundary portion and said second scene image first boundary portion and forming animal indicia for said animal undergoing evaluation based on said plurality of selected parameters.

31. The apparatus of claim 30 wherein said acquiring of said first scene evaluation image representation at said selected time during movement of an animal undergoing evaluation has said selected time selected by a detection means capable of detecting said animal undergoing evaluation having reached a selected location in said first and second scenes.

32. The apparatus of claim 30 wherein said image acquisition means comprises a first video camera such that a binary representation is stored at least temporarily in said storage and computing means for each corresponding image pixel element in said first scene image sensed by a corresponding image sensing element in said first video camera, and further comprises a second video camera such that a binary representation is stored at least temporarily in said storage and computing means for each corresponding image pixel element in said second scene image sensed by a corresponding image sensing element in said second video camera.

33. The apparatus of claim 32 wherein an animal undergoing evaluation moving through said first and second scenes passes through a chute means, said chute means having an open top through which one of said fields of view is obtained by said first video camera, and further having a side wall with an opening therein through which that remaining one of said fields of view is obtained by said second video camera but only of a selected portion of said animal undergoing evaluation at said selected time.

34. The apparatus of claim 30 further comprising another measuring apparatus for obtaining a measurement value of an animal undergoing evaluation other than said image acquisition means.

35. A method for evaluating animals with each such animal during its evaluation moving through first and second scenes corresponding to differing first and second fields of view in a manner so as to partially cover first and second background surfaces provided in each of said first and second scenes, respectively, as viewed in said first and second fields of view, respectively, said first and second background surfaces each having a visual appearance which contrast with a visual appearance of that animal, said method comprising:

acquiring a first scene evaluation image representation of an image of said first scene at a selected time during movement of an animal undergoing evaluation simultaneously through said first and second scenes, there being an animal image portion corresponding to that animal in said first scene evaluation image representation and a background image portion in said first scene evaluation image representation corresponding to uncovered portions of said first background surface;

acquiring a second scene evaluation image representation of an image of said second scene substantially at said selected time during said movement of said animal, there being an animal image portion corresponding to that animal in said second scene evaluation image representation and a background image portion in said second scene evaluation image representation corresponding to uncovered portions of said second background surface; and

locating in at least one of said first and second scene evaluation image representations a separation location between at least part of said animal image portion thereof and at least part of said background image portion thereof.

36. The method of claim 35 further comprising locating in each of said first and second scene evaluation image representations a separation location between at least part of said animal image portion thereof and at least part of said background image portion thereof.

37. The method of claim 36 wherein said locating in said second scene evaluation image representation a separation location between at least part of said animal image portion thereof and at least part of said background image portion thereof depends on that location determined for said separation location in said first scene evaluation image representation.

38. The method of claim 35 wherein said locating of said separation location in one of said first and second scene evaluation representations further comprises at least partially transforming that scene evaluation image representation by decomposing same into at least a partial spectrum based on selected spectral functions.

39. The method of claim 38 wherein said locating of said separation location in one of said first and second scene evaluation image representations is based on locations in that scene evaluation image representation where corresponding substantial changes of values of spectral components, found by said transforming, occur.

40. The method of claim 39 wherein said first and second background surfaces each have a periodic pattern thereon.