Method and device for analyzing berries

• US 8,406,475 B2
• Filed: 06/29/2006
• Issued: 03/26/2013
• Est. Priority Date: 06/30/2005
• Status: Active Grant

• Current Assignee: D Wine
• Original Assignee: D.Wine
• Inventors: Bernard, Nicolas, Larcheveque, Boris, Zebic, Olivier Bertrand Daniel
• Primary Examiner(s): Lu, Tom Y
• Assistant Examiner(s): CONWAY, THOMAS A
• Application Number:
  US11/994,120
• Publication Number:
  US 20100215221A1
• Time in Patent Office: 2,462 Days
• Field of Search: None
• US Class Current: 382/110
• CPC Class Codes: G01N 2021/8812 Diffuse illumination, e.g. "sky"
• CPC Class Codes: G01N 21/27 using photo-electric detection G01N21/31 takes precedence; circuits for computing concentration logarithmic circuits G06G7/24; photometric circuits in general G01J
• CPC Class Codes: G01N 21/8806 Specially adapted optical and illumination features
• CPC Class Codes: G01N 21/951 Balls

Abstract

Method for analyzing berries, characterized in that: berries are arranged on an analysis plate, an image of the berries is recorded, the image consisting of pixels of at least one component of an elementary color, the contours of the berries are identified, and geometric and/or colorimetric characteristics of the berries are determined. The number of berries can then be determined, or a volume, a surface area, a surface area to volume ratio, and a characteristic concentration for each berry can be determined, in addition to an average, for all of the pixels inside the contour of a defined berry, of the values of a component of an elementary color of a component of interest. A device for implementing the method is also disclosed. The invention especially relates to viticulture and to the analysis of the maturity of the grape.

Claims

1. A method for analyzing berries, comprising:

arranging berries on an analysis plate;

an acquisition of an image of the arranged berries, said image of the berries being composed of pixels of at least one elementary colour component;

an identification, by a processor, of contours of the berries;

a determination, by the processor, of geometric or colorimetric characteristics of the berries;

a reiteration of the method according to the invention on a same type of berry over an extended period of time, and a monitoring over the extended period of time of one of the geometric and colorimetric characteristics determined on the berries of the same type to represent capture and analysis of variation of the determined geometric or colorimetric characteristic values of the berries over the extended period of time; and

a visualisation, on a visualization screen, of a graph illustrating development over the extended period of time of the determined geometric or colorimetric characteristics.

2. The method according to claim 1, further comprises saving at least one of the determined characteristics.

3. The method according to claim 1, wherein the berries of the same type originate from a same plot.

4. The method according to claim 1, further comprises a visualisation of determined characteristics.

5. The method according to claim 1, further comprises a determination of the number of berries in the image of berries.

6. The method according to claim 1, wherein the determination of characteristics comprises a determination of a volume per berry.

7. The method according to claim 1, wherein the determination of characteristics comprises a determination of a surface area per berry.

8. The method according to claim 1, wherein the determination of characteristics comprises a determination of a surface area to volume ratio per berry.

9. The method according to claim 1, wherein the determination of characteristics comprises, for each berry and at least one of the elementary colour components or a colour component of interest, a calculation of an average, for all the pixels of the image of berries inside the contour of a berry, of the values of the considered colour component.

10. The method according to claim 1, wherein the determination of characteristics comprises a calculation of a heterogeneity factor and an average of a determined characteristic for each berry.

11. The method according to claim 1, further comprises a lighting of the berries according to predetermined lighting conditions.

12. The method according to claim 1, further comprises a geometric calibration of the image of the berries.

13. The method according to claim 1, further comprises a correction of the colours of the image of the berries.

14. The method according to claim 1, wherein the monitoring over the extended period of time of at least one of the geometric or colorimetric characteristics determined on the berries of the same type occurs under substantially similar lighting conditions.

15. The method according to claim 9, wherein the method is applied to white grape berries, and in that the considered colour component is yellow or green, or in that it is applied to red grape berries, and in that the considered colour component is red or black.

16. The method according to claim 9, wherein the elementary colour component is situated in the infrared, and in that it also comprises a determination, by infrared spectrophotometry and chemometry, of at least one characteristic concentration per berry.

17. The method according to claim 1, wherein the identification of contours of the berries comprises:

a conversion of the image of the berries to a thresholded image, in which it is possible to distinguish blocks of berries,

a deletion of impurities from the thresholded image, and

a segmentation of the blocks of berries to berries with defined contours.

18. The method according to claim 17, wherein the conversion to a thresholded image comprises a thresholding of the image of the berries according to a threshold for a value of an elementary colour component.

19. The method according to claim 17, wherein the deletion of impurities comprises a succession of erosions and dilations of the blocks of berries.

20. The method according to claim 17, wherein the deletion of impurities comprises an algorithm for filling holes by propagation.

21. The method according to claim 17, wherein the segmentation of the blocks of berries comprises:

a conversion of the thresholded image to a grey-scale image,

a thresholding of the grey-scale image by fuzzy logic,

a use of an erosion operator until a centre of gravity for each berry is obtained, and

an expansion around the centres of gravity, until the contours of the berries are obtained.

22. The method according to claim 1, further comprises an access to a database comprising at least one result of another analysis of berries obtained using any method.

23. The method according to claim 22, wherein the analysis result consists of a concentration.

24. The method according to claim 22, further comprises obtaining an item of data by matching at least one of the determined characteristics with the analysis result of berries obtained using any method.

25. The method according to claim 24, wherein the determination of characteristics comprises a determination of an average volume of the arranged berries, and in that it comprises obtaining a sugar content by matching the determined volume and a concentration of sugars.

26. A device for analyzing berries implementing a method according to claim 1, comprising:

an analysis plate;

means for acquiring an image of berries arranged on the analysis plate, said image being composed of pixels of at least one elementary colour component;

means for determining geometric or colorimetric characteristics of the berries;

a means for carrying out a monitoring over an extended period of time of at least one of the determined characteristics on berries of the same type to represent capture and analysis of variation of the determined geometric or colorimetric characteristic values of the berries over the extended period of time; and

a visualization screen for visualizing a graph illustrating a development over the extended period of time of at least one of the determined geometric or colorimetric characteristics.

27. The device according to claim 26, further comprises means for saving at least one of the determined characteristics.

28. The device according to claim 26, further comprises means for visualizing at least one of the determined characteristics.

29. The device according to claim 26, wherein the analysis plate comprises spaced recesses.

30. The device according to claim 26, wherein the means for acquiring an image of berries comprise a multispectral camera carrying out an acquisition of an image of berries in a visible range or an infrared range.

31. The device according to claim 26, wherein the means for determining geometric or colorimetric characteristics include image-processing means.

32. The device according to claim 26, further comprises means for accessing a database comprising at least one result of another analysis of berries obtained using any device.

33. The device according to claim 32, further comprises means for obtaining an item of data by matching at least one of the determined characteristics with the result of analysis of berries obtained using any device.

34. The device according to claim 26, further comprises means for lighting berries arranged on the analysis plate.

35. The device according to claim 34, further comprises a ground glass surface situated between the lighting means and the analysis plate, and situated upstream of the means for acquiring an image of berries.

36. The device according to claim 26, further comprises means for geometrically calibrating the image of the berries.

37. The device according to claim 36, wherein the means for geometrically calibrating the image of berries include a reference object of known dimensions.

38. The device according to claim 26, further comprises means for correcting the colours of the image of berries.

39. The device according to claim 38, wherein the means for correcting the colours of the image of berries include a colour test pattern.