Multi-variable model for identifying crop response zones in a field

• US 7,058,197 B1
• Filed: 11/04/1999
• Issued: 06/06/2006
• Est. Priority Date: 11/04/1999
• Status: Expired due to Term

• Current Assignee: Board of Trustees of The University of Illinois
• Original Assignee: Board of Trustees of The University of Illinois
• Inventors: Pearson, Randall Scott, McGuire, John Dennis
• Primary Examiner(s): Couso, Yon J.
• Application Number:
  US09/434,391
• Time in Patent Office: 2,406 Days
• Field of Search: 382/100, 382/103, 382/106, 382/108, 382/109, 382/110, 382/181, 382/191, 382/206, 382/218, 382/224, 382/286, 382/227, 382/228, 382/113, 348/141, 348/144, 358/524, 702/3, 702/13
• US Class Current: 382/100
• CPC Class Codes: G06Q 50/02 Agriculture; Fishing; Forestry; Mining
• CPC Class Codes: G06T 2207/10032 Satellite or aerial image; Remote sensing
• CPC Class Codes: G06T 2207/20064 Wavelet transform [DWT]
• CPC Class Codes: G06T 2207/30188 Vegetation; Agriculture
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06T 7/42 using transform domain methods
• CPC Class Codes: G06V 20/188 Vegetation

Abstract

An computer implemented apparatus and method are disclosed for defining areas of a field in which a crop or other vegetation is grown based on their selective ability to grow such vegetation through a growing season, or some shorter preselected time period. The method includes making a number of temporally separated measurements through air borne imaging of a field, registering the data to the geography of the field and each other, normalizing the data including converting the data to a vegetative index indicative to the presence of vegetation in the field, comparing the data to identify clusters of like value, and classifying the clusters and images to learn how the different field areas responded in growing vegetation through the season. With this method, the field may be segregated into a number of like areas called crop response zones which exhibit similar vegetative growth characteristics as an aid to a grower in his decision making regarding how to maximize yield in his field.

Claims

1. A method for processing multiple initial data sets of unclassified pixel values to enable their comparison on a relative basis, each of said data sets being representative of the vegetative growth in a field, said method comprising the steps of:

converting each unclassified pixel value of each initial data set to a normalized value, thereby creating a new data set of unclassified normalized pixel values corresponding to each initial data set of unclassified pixel values, and

using the data sets of unclassified normalized pixel values as representative of the initial data sets for comparison purposes.

2. The method of claim 1 wherein the converting step comprises the steps of:

for each initial data set of unclassified pixel values, calculating a set of statistical parameters associated therewith, wherein the calculated statistical parameters reflect only the statistical parameters for that initial data set, and

using said calculated statistical parameters in converting only the unclassified pixel values of its associated initial data set.

3. The method of claim 2 wherein the step of calculating comprises the step of calculating the mean and the standard deviation for each initial data set.

4. The method of claim 3 wherein the step of using said calculated statistical parameters comprises the steps of subtracting the pixel value from the mean and dividing the result by the standard deviation.

5. The method of claim 4 wherein each pixel value corresponds to an NDVI value, and further comprising the step of calculating an NDVI value for each original reading taken from the field.

6. The method of claim 5 wherein the original readings are expressed in terms of a binary number and the step of calculating an NDVI value includes the step of converting said binary numbers into a decimal number.

7. A method for scaling multiple data sets of pixel values to enable their comparison on a relative basis, each of said data sets being representative of the vegetative growth in a field, each data set comprising unclassified pixel values, said method comprising the steps of:

statistically normalizing each data set of unclassified pixel values based on its own mean and standard deviation, and

using the data sets of normalized values as representative of the data sets of pixel values for comparison purposes.

8. The method of claim 7 further comprising the step of calculating the pixel values by converting raw data into a vegetative index value for each pixel.

9. The method of claim 8 wherein the step of statistically normalizing includes the steps of subtracting the pixel value from the mean for each data set and dividing the result by the standard deviation of each data set to thereby calculate a set of values representative of the pixel values for comparison purposes.

10. A method for characterizing different areas of a field used for growing vegetation according to the presence of vegetation in said areas over a growing season, said method comprising the steps of:

periodically collecting a set of data characterizing the amount of vegetation present in the field, thereby producing a plurality of data sets,

normalizing the data in each data set,

aggregating the normalized data comprising said normalized data sets, and

grouping said normalized data into a plurality of groups based on their respective normalized values, said plurality of groups being representative of said different field areas.

11. The method of claim 10 wherein the step of periodically collecting a set of data includes the step of collecting a set of data that spatially characterizes the presence of vegetation in the field.

12. The method of claim 11 further comprising the step of registering the data sets so that the data in each data set are substantially correlated with the data in other data sets.

13. The method of claim 12 wherein each data set includes a plurality of pixels, each pixel corresponding to a geographic location in said field, and wherein the step of registering the data includes the step of substantially aligning pixels from data sets corresponding to the same geographic location in said field.

14. The method of claim 12 wherein the step of aggregating the normalized data comprises the step of plotting each normalized data set along an axis to thereby create a multi-dimensional array of said normalized data.

15. The method of claim 12 wherein the step of grouping the normalized data comprises the step of assigning each data point into a cluster based on its Euclidean distance from a calculated mean of each cluster.

16. The method of claim 15 further comprising the step of classifying the normalized data in said array by assigning each data point into a cluster based on its mathematical probability of belonging to said cluster.

17. A method for defining a plurality of crop response zones in a field used for growing vegetation, each of said crop response zones being characterized by having similar temporal growth characteristics, the method comprising the steps of:

periodically collecting data characterizing the vegetative growth in the field, thereby creating a plurality of temporally separated data sets,

normalizing the data of each data set independently of the other data sets, and

grouping the normalized data into clusters, each of said groups being representative of a crop response zone.

18. The method of claim 17 wherein the step of periodically collecting the data includes the step of collecting data that is specific to a particular geographic location in the field.

19. The method of claim 18 further comprising the step of converting the data into corresponding values of a vegetative index.

20. The method of claim 19 further comprising the step of registering the data.

21. The method of claim 20 wherein the step of grouping further comprises the step of classifying the data into clusters based on their statistical probability for belonging to a particular cluster.

22. A pre-programmed digital electronic device for defining a plurality of crop response zones in a field used for growing vegetation, each of said crop response zones being characterized by having similar temporal vegetative growth characteristics, comprising:

means for processing a plurality of data sets taken periodically which characterize the vegetative growth in the field,

means for normalizing the data of each data set independently of the other data sets, and

means for clustering the normalized data into clusters, each of said clusters being representative of a crop response zone.

23. The device of claim 22 further comprising means for converting the data into corresponding values of a vegetative index.

24. The device of claim 23 further comprising means for registering the data taken at different times.

25. The device of claim 24 further comprising means for classifying the data into clusters based on their statistical probability for belonging to a particular cluster.

26. The device of claim 25 further comprising means for displaying the data arranged in clusters.

27. A pre-programmed digital electronic device for processing multiple initial data sets of unclassified pixel values to enable their comparison on a relative basis, each of said data sets being representative of the vegetative growth in a field, comprising:

means for converting each unclassified pixel value of each initial data set to a normalized value based on statistical parameters that are derived solely from the unclassified pixel values of the initial data set to which the unclassified pixel value belongs, thereby creating a new data set of normalized values corresponding to each initial data set, and

means for comparing said data sets of normalized values as representative of the initial data sets.

28. The device of claim 27 wherein said initial data sets each represent a reading taken from the field at different times and under different conditions, and wherein said converting means includes means for registering said initial data sets geographically across the field.

29. A method for defining areas in a field which perform substantially alike in growing vegetation over the course of at least a portion of a growing season, said method comprising the steps of:

generating a plurality of data sets by periodically measuring the vegetative growth across said field, said measurements being associated with a geographic location in said field, each data set comprising a plurality of pixel values and corresponding to a different measurement,

converting said pixel values of each data set into vegetative index values,

normalizing said each converted pixel value of each data set based on statistical parameters that are derived solely from the converted pixel values of the data set to which that converted pixel value belongs,

arranging said normalized pixel values from each data set along an axis of a multi-dimensional coordinate system,

defining, at least partially on the basis of the arranged normalized pixel values, a plurality of field areas in the coordinate system that substantially share the same performance, and

labeling the characteristics of said field areas according to their shared performance.

30. The method of claim 29 wherein the step of periodically measuring includes the step of measuring the reflected electromagnetic energy from the field.

31. The method of claim 30 wherein the step of defining the plurality of field areas includes the step of aggregating a plurality of the normalized pixel values into each of said defined field areas based on their statistical probability of belonging to said defined field area.

32. A method for processing multiple data sets of unclassified pixel values to enable their comparison on a relative basis, at least one of said data sets being representative of the agronomic variability in a field, said method comprising the steps of:

converting each unclassified pixel value of at least one of said data sets to a normalized value, wherein the normalization of each pixel value is based on statistical parameters derived from the unclassified pixel values of only the at least one data set to which that unclassified pixel value belongs, thereby creating a new data set of normalized values corresponding to said at least one of said data sets of pixel values, and

using the data set of normalized values as representative of the data sets of pixel values for comparison purposes.

33. The method of claim 32 wherein the step of converting includes the step of converting the unclassified pixel values of each data set desired to be compared.

34. The method of claim 32 further comprising the step of using at least one data set comprised of remotely sensed data.

35. The method of claim 32 wherein the converting step comprises the steps of calculating a set of statistical parameters for each data set of unclassified pixel values, and using said calculated statistical parameters in converting only its associated data set of pixel values.

36. The method of claim 35 wherein the step of calculating comprises the step of calculating the mean and the standard deviation for each data set of unclassified pixel values desired to be compared.

37. The method of claim 36 wherein each pixel value corresponds to a vegetative index value, and further comprising the step of calculating a vegetative index value for each reading taken from the field.

38. A digital electronic storage medium, said digital electronic storage medium containing a stored digital program for performing the steps of the method recited in claim 32.

39. The digital electronic storage medium of claim 38 wherein said digital electronic medium comprises a computer disk.

40. An electronic digital processing machine, said machine being connected to a network of computers, said machine containing a stored digital program for performing the steps of the method recited in claim 32.

41. A method of scaling multiple data sets of unclassified pixel values to enable their comparison on a relative basis, at least one of said data sets being representative of the agronomic variability in a field, said method comprising the steps of:

statistically normalizing said at least one data set based on its own mean and standard deviation, and

using said at least one data set of normalized values as representative of said at least one data set of pixel values for comparison purposes.

42. The method of claim 41 further comprising the step of calculating the pixel values by converting raw data into a vegetative index value for each pixel.

43. A digital electronic storage medium, said digital electronic storage medium containing a stored digital program for performing the steps of the method recited in claim 41.

44. The digital electronic storage medium of claim 43 wherein said digital electronic medium comprises a computer disk.

45. An electronic digital processing machine, said machine being connected to a network of computers, said machine containing a stored digital program for performing the steps of the method recited in claim 41.

46. A method for characterizing different areas of a field used for growing vegetation according to the presence of vegetation in said areas over time, said method comprising the steps of:

periodically collecting a set of data characterizing the presence of vegetation in the field, thereby producing a plurality of data sets,

for each data set, normalizing the data therein on the basis of statistical parameters derived solely from the data of that data set,

aggregating the normalized data comprising said normalized data sets, and

grouping said normalized data into a plurality of groups based on their respective normalized values, said plurality of groups being representative of said different field areas.

47. The method of claim 46 wherein the step of normalizing comprises normalizing the data in each set according to statistical parameters derived solely from the data in the data set to which the data being normalized belongs.

48. The method of claim 47 wherein the step of periodically collecting the data includes spatially associating said data with the field areas.

49. The method of claim 48 wherein the step of periodically collecting the data includes the step of collecting data which characterizes the health and amount of vegetation in the field areas.

50. The method of claim 49 wherein the step of periodically collecting data includes collecting data no less frequently than once for each growing season desired to be characterized.

51. The method of claim 49 wherein the step of periodically collecting data includes collecting data no less frequently than at least four times during a growing season, for characterizing said field areas during said growing season.

52. The method of claim 49 further comprising the step of registering the data sets so that the data in each data set are substantially correlated with the data in other data sets.

53. The method of claim 52 wherein each data set includes a plurality of pixels, each pixel corresponding to a geographic location in said field, and wherein the step of registering the data includes the step of substantially aligning pixels from data sets corresponding to the same geographic location in said field.

54. The method of claim 49 further comprising the step of clustering said normalized data into a plurality of clusters based on their respective normalized values.

55. The method of claim 54 wherein the step of clustering the normalized data comprises the step of assigning each data point into a cluster based on its Euclidean distance from a calculated mean of each cluster.

56. The method of claim 55 further comprising the step of classifying the normalized data by assigning each data point into a cluster based on its mathematical probability of belonging to said cluster.

57. A method for defining a plurality of crop response zones in a field used for growing vegetation, each of said crop response zones being characterized by having similar temporal vegetative growth characteristics, the method comprising the steps of:

periodically collecting data characterizing the vegetative growth in the field, thereby generating a plurality of vegetative growth data sets, each vegetative growth data set corresponding to a different collection,

normalizing the data in each vegetative growth data set independently of the data in the other vegetative growth data sets, and

grouping the normalized data into groups, each of said groups being representative of a crop response zone.

58. The method of claim 57 wherein the step of periodically collecting the data includes the step of collecting a plurality of temporally separated data sets that are specific to a particular geographic location in the field, wherein the step of normalizing comprises normalizing the data in each data set according to statistical parameters derived solely from the data in the data set to which the data being normalized belongs, and further comprising the step of geographically registering the data.

59. The method of claim 58 wherein the step of collecting the data includes the step of collecting remotely sensed data, and further comprising the step of converting the remotely sensed data into corresponding values of a vegetative index.

60. The method of claim 59 wherein the step of grouping further comprises the step of clustering the data into clusters based on their statistical probability of belonging to a particular cluster.

61. A digital electronic storage medium, said digital electronic storage medium containing a stored digital program for performing the steps of the method recited in claim 57.

62. The digital electronic storage medium of claim 61 wherein said digital electronic medium comprises a computer disk.

63. An electronic digital processing machine, said machine being connected to a network of computers, said machine containing a stored digital program for performing the steps of the method recited in claim 57.

64. A method for defining areas in a field which perform substantially alike in growing vegetation over the course of time, said method comprising the steps of:

generating a plurality of data sets by periodically measuring the health and amount of vegetation throughout said field, said measurements being associated with specific geographic locations in said field, each data set corresponding to a different measurement,

converting said data in said data sets into vegetative index values,

normalizing said converted data of each data set independently of the data of the other data sets,

defining a plurality of field areas that substantially share the same performance throughout the data on the basis of the normalized data, and

identifying the characteristics of said field areas according to their shared performance.

65. The method of claim 64 wherein the step of measuring data includes the step of periodically remotely sensing data, thereby creating a plurality of temporally separated data sets comprising data values indicative of the health and amount of vegetation in specific geographic locations in said field, and wherein the step of normalizing comprises normalizing the converted data in each data set according to statistical parameters derived solely from the converted data in the data set to which the converted data being normalized belongs.

66. The method of claim 65 wherein the defining step includes aggregating a plurality of the data into each of said defined field areas based on their statistically probability of belonging to said defined field area.

67. A digital electronic storage medium, said digital electronic storage medium containing a stored digital program for performing the steps of the method recited in claim 64.

68. The digital electronic storage medium of claim 67 wherein said digital electronic medium comprises a computer disk.

69. An electronic digital processing machine, said machine being connected to a network of computers, said machine containing a stored digital program for performing the steps of the method recited in claim 64.