Apparatus and method for estimation of livestock weight

• US 8,755,570 B2
• Filed: 04/27/2011
• Issued: 06/17/2014
• Est. Priority Date: 04/27/2011
• Status: Active Grant

• Current Assignee: Edward Theil, John Sweeney, Steve Gomas
• Original Assignee: Edward Theil, John Sweeney, Steve Gomas
• Inventors: Gomas, Steve, Sweeney, John, Theil, Edward
• Primary Examiner(s): Koziol, Stephen R
• Assistant Examiner(s): Azima, Shaghayegh
• Application Number:
  US13/095,800
• Publication Number:
  US 20120275659A1
• Time in Patent Office: 1,147 Days
• Field of Search: 382/110, 426/529, 426/575, 426/573, 426/271, 426/577
• US Class Current: 382/110
• CPC Class Codes: A01K 29/005 Monitoring or measuring activity, e.g. detecting heat or mating
• CPC Class Codes: G06T 2207/10004 Still image; Photographic image
• CPC Class Codes: G06T 2207/30128 Food products
• CPC Class Codes: G06T 7/0004 Industrial image inspection
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06T 7/12 Edge-based segmentation
• CPC Class Codes: G06T 7/149 involving deformable models, e.g. active contour models
• CPC Class Codes: G06V 10/46 Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features colour feature extraction G06V10/56
• CPC Class Codes: G06V 10/7553 based on shape, e.g. active shape models [ASM]
• CPC Class Codes: G06V 10/76 based on eigen-space representations, e.g. from pose or different illumination conditions; Shape manifolds

Abstract

A digital image of a target object is received and the image includes a plurality of pixels. The target object is at least a portion of an animal, and the portion of the animal is disposed in the image according to a predetermined disposition. The weight of the animal is determined using the model and at least one characteristic. At least some of the pixels of the image of animal are analyzed to determine at least one characteristic of the image of the animal. The weight of the animal is determined using the at least one characteristic.

Claims

1. A method of determining an animal weight, the method comprising:

receiving a digital image of a target object, the image comprising a plurality of pixels, the target object being at least a portion of an animal, the at least a portion of the animal disposed in the image according to a predetermined disposition;

analyzing at least some of the pixels of the image of animal to determine a perimeter of an outline of the at least a portion of the animal, wherein the perimeter of the outline of the at least a portion of the animal is iteratively determined by obtaining a current outline and subsequently calculating a new outline, the new outline being based upon at least one candidate boundary point from the digital image, the determining being terminated based upon a comparison of the current outline with the new outline;

determining an area enclosed by the perimeter; and

determining the weight of the animal using at least a quotient obtained by dividing a square of the perimeter by the area enclosed by the perimeter.

2. The method of claim 1 further comprising receiving a physical characteristic of the animal and wherein determining the weight comprises determining the weight using the perimeter and the physical characteristic.

3. The method of claim 2 wherein the physical characteristic is selected from a group consisting of an age of the animal, a breed of the animal, animal markings, a previous weight, and identification tag information concerning the animal.

4. The method of claim 1 further comprising notifying a management service of the determined weight.

24. The method of claim 1, wherein the determining the weight of the animal further comprises estimating the weight of the animal using according to at least a following algorithm:

W=ƒ

(P²/A, α

;

{a₁, a₂ . . . , a_m});

where W is the weight of the animal;

P is the perimeter of the outline of the at least a portion of the animal;

A is the area enclosed by the perimeter;

a₁, a₂, and a_m are regression coefficients, and ƒ

is a least-squares line associated with the regression coefficients.

29. The method of claim 1, wherein the receiving a digital image of a target object includes receiving a single digital image of a posterior view of an animal.

5. An apparatus for determining an animal weight, the apparatus comprising:

an interface having an input, the input configured to receive a digital image of a target object, the image comprising a plurality of pixels, the target object being at least a portion of an animal, the at least a portion of the animal disposed in the image according to a predetermined disposition, the interface having an output;

a controller coupled to the interface, the controller configured to identify the target object from the image, the controller further configured to determine at least one characteristic of the image of the animal, the controller further configured to determine the weight of the animal using the at least one characteristic and present the weight at the output of the interface;

wherein the at least one characteristic comprises a perimeter of an outline of the animal and wherein the perimeter of the outline of the animal in iteratively determined by the controller by obtaining a current outline and subsequently calculating a new outline, the new outline being based upon at least one candidate boundary point from the digital image, the determining being terminated based upon a comparison of the current outline with the new outline;

wherein the controller is configured to determine an area enclosed by the perimeter and to determine the weight of the animal using at least a quotient obtained by dividing a square of the perimeter by the area enclosed by the perimeter.

6. The apparatus of claim 5 wherein the interface is configured to receive a physical characteristic of the animal at the input and wherein the controller is configured to determine the weight using the perimeter and the physical characteristic.

7. The apparatus of claim 6 wherein the physical characteristic is selected from a group consisting of an age of the animal, a breed of the animal, animal markings, a previous weight, and identification tag information concerning the animal.

25. The apparatus of claim 5, wherein the controller is configured to determine the weight of the animal by using at least a following algorithm:

W=ƒ

(P²/A, α

;

{a₁, a₂ . . . , a_m});

where W is the weight of the animal;

P is the perimeter of the outline of the at least a portion of the animal;

A is the area enclosed by the perimeter;

a₁, a₂, and a_m are regression coefficients, and ƒ

is a least-squares line associated with the regression coefficients.

30. The apparatus of claim 5, wherein the digital image of a target object is a single digital image of a posterior view of an animal.

8. A method of determining an animal weight, the method comprising:

receiving a digital image of a target object, the image comprising a plurality of pixels, the target object being at least a portion of an animal, the at least a portion of the animal disposed in the image according to a predetermined disposition;

constructing a model, the model representing an expected visual perimeter of the animal in the predetermined disposition;

identifying the target object included in the digital image using the model and according to an analysis of at least some of the pixels in the digital image;

iteratively determining the perimeter of the outline of the animal by obtaining a current outline and subsequently calculating a new outline, the new outline being based upon at least one candidate boundary point from the digital image, the determining being terminated based upon a comparison of the current outline with the new outline;

determining an area enclosed by the perimeter using the identified target object;

receiving at least one characteristic of an animal from a database, the animal being the same species as the animal in the target image;

determining the weight of the animal using at least a quotient obtained by dividing a square of the perimeter by the area enclosed by the perimeter, and the at least one characteristic.

9. The method of claim 8 further comprising comparing the determined weight to an expected weight and taking an action when the expected weight is not met.

10. The method of claim 8 wherein the at least one characteristic is selected from a group consisting of an age of the animal, a breed of the animal, animal markings, a previous weight, and identification tag information.

11. The method of claim 8 wherein determining the perimeter comprises finding at least one edge of the animal.

12. The method of claim 11 wherein determining the perimeter comprises creating a region of interest associated with the image.

13. The method of claim 8 further comprising notifying a management service of the determined weight.

14. The method of claim 8 wherein the target image does not include actual dimensions or measurements.

26. The method of claim 8, wherein the determining the weight of the animal further comprises estimating the weight of the animal using according to at least a following algorithm:

W=ƒ

(P²/A, α

;

{a₁, a₂ . . . , a_m});

where W is the weight of the animal;

P is the perimeter of the outline of the at least a portion of the animal;

A is the area enclosed by the perimeter;

a₁, a₂, and a_m are regression coefficients, and ƒ

is a least-squares line associated with the regression coefficients.

31. The method of claim 8, wherein the receiving a digital image of a target object includes receiving a single digital image of a posterior view of an animal.

15. An apparatus for determining an animal weight, the apparatus comprising:

an interface having an input, the input configured to receive a digital image of a target object, the image comprising a plurality of pixels, the target object being at least a portion of an animal, the at least a portion of the animal disposed in the image according to a predetermined disposition, the interface having an output, the input further configured to receive at least one characteristic of an animal from a database, the animal being the same species as the animal in the target image;

a controller coupled to the interface, the controller configured to construct a model, the model representing an expected visual perimeter of the outline of the animal in the predetermined disposition, the controller further configured to identify the target object from the image using the model and according to an analysis of at least some of the pixels in the image, the controller further configured to determine the perimeter of the outline of the animal, the perimeter being iteratively determined by obtaining a current outline and subsequently calculating a new outline, the new outline being based upon at least one candidate boundary point from the digital image, the determination being terminated based upon a comparison of the current outline with the new outline,

the controller being further configured to determine an area enclosed by the perimeter using the identified target, the controller further configured to determine the weight of the animal using at least a quotient obtained by dividing a square of the perimeter by the area enclosed by the perimeter, and the at least one characteristic, and present the weight at the output of the interface.

16. The apparatus of claim 15 wherein the at least one characteristic is selected from a group consisting of an age of the animal, a breed of the animal, animal markings, a previous weight, and identification tag information.

17. The apparatus of claim 15 wherein the controller is further configured to determine at least one edge of the animal.

18. The apparatus of claim 17 wherein the controller is further configured to determine a region of interest on the image.

19. The apparatus of claim 15 wherein the target image does not include actual dimensions or measurements.

27. The apparatus of claim 15, wherein the controller is configured to determine the weight of the animal by using at least a following algorithm:

W=ƒ

(P²/A, α

;

{a₁, a₂ . . . , a_m});

where W is the weight of the animal;

P is the perimeter of the outline of the at least a portion of the animal;

A is the area enclosed by the perimeter;

a₁, a₂, and a_m are regression coefficients, and ƒ

is a least-squares line associated with the regression coefficients.

32. The apparatus of claim 15, wherein the digital image of a target object is a single digital image of a posterior view of an animal.

20. A method of determining an animal weight, the method comprising:

receiving a digital image of a target object, the image comprising a plurality of pixels, the target object being at least a portion of an animal, the at least a portion of the animal disposed in the image according to a predetermined disposition;

performing an analysis of the digital image by analyzing at least some of the pixels in the image;

iteratively determining the perimeter of the outline of the animal by obtaining a current outline and subsequently calculating a new outline, the new outline being based upon at least one candidate boundary point from the digital image, the determining being terminated based upon a comparison of the current outline with the new outline;

determining an area enclosed by the perimeter according to the analysis;

receiving at least one characteristic of an animal from a database, the animal being the same species as the animal in the image;

determining the weight of the animal using at least a quotient obtained by dividing a square of the perimeter by the area enclosed by the perimeter, and the at least one characteristic.

21. The method of claim 20 further comprising determining whether false edges exist in the perimeter based at least in part from information in the data base.

22. The method of claim 20 further comprising comparing the determined weight to a predetermined threshold and taking an action when the threshold is exceeded.

23. The method of claim 20 wherein the at least one characteristic is selected from a group consisting of an age of the animal, animal markings, a previous weight, and identification tag information.

28. The method of claim 20, wherein the determining the weight of the animal further comprises estimating the weight of the animal using according to at least a following algorithm:

W=ƒ

(P²/A, α

;

{a₁, a₂ . . . , a_m});

where W is the weight of the animal;

P is the perimeter of the outline of the at least a portion of the animal;

A is the area enclosed by the perimeter;

a₁, a₂, and a_m are regression coefficients, and ƒ

is a least-squares line associated with the regression coefficients.

33. The method of claim 20, wherein the receiving a digital image of a target object includes receiving a single digital image of a posterior view of an animal.