Method and apparatus for determining the quality of an image of an agricultural field using a plurality of fuzzy logic input membership functions

• US 6,714,662 B1
• Filed: 07/10/2000
• Issued: 03/30/2004
• Est. Priority Date: 07/10/2000
• Status: Expired due to Term

• Current Assignee: Board of Trustees of The University of Illinois, Blue Leaf IP Incorporated (CNH Industrial N.V.), CNH Industrial America LLC (CNH Industrial N.V.)
• Original Assignee: Case Corporation (CNH Industrial N.V.)
• Inventors: Zhang, Qin, Benson, Eric R., Reid, John F.
• Primary Examiner(s): MARIAM, DANIEL G
• Application Number:
  US09/612,851
• Time in Patent Office: 1,359 Days
• Field of Search: 382/100, 382/104, 382/153, 382/160, 382/173, 382/181, 382/190, 382/199, 382/206, 382/307, 56/101, 56/102.F, 56/102.J, 348/113, 348/116, 348/119, 318/580, 318/587, 701/23, 701/28, 701/40, 701/41, 701/50, 701/57, 701/213
• US Class Current: 382/104
• CPC Class Codes: A01D 41/127 Control or measuring arrangements specially adapted for combines
• CPC Class Codes: G06T 2207/30161 Wood; Lumber
• CPC Class Codes: G06T 7/70 Determining position or orientation of objects or cameras camera calibration G06T7/80
• CPC Class Codes: G06V 20/10 Terrestrial scenes scenes under surveillance with static cameras G06V20/52; scenes perceived from the exterior of a vehicle G06V20/56; scenes perceived from the interior of a vehicle G06V20/59
• CPC Class Codes: G06V 20/56 exterior to a vehicle by using sensors mounted on the vehicle

Abstract

A combine having a camera system is disclosed in which pictures of the ground in front of the vehicle are processed to extract the slope and intercept of a characteristic line indicative of the boundary between the cut and uncut portions of the field. A fuzzy logic algorithm compares the slope and intercept using four input membership functions and a single output membership function. If the slope and intercept extracted from the image are deemed acceptable, they are used to navigate the vehicle through the field. If they are not acceptable, the combine continues to use slope and intercept values previously deemed acceptable to guide the vehicle. This process of image data extraction and quality analysis continues repetitively as the combine traverses the field harvesting crops.

Claims

1. A method of processing a sensor signal indicative of an environmental condition comprising the steps of:

receiving an electrical sensor signal indicative of an environmental condition;

deriving from the sensor signal at least one numeric sensor signal value characteristic of an environmental condition;

processing the sensor signal with a plurality of fuzzy logic input membership functions;

combining the outputs of the plurality of fuzzy logic membership functions in at least one fuzzy logic output membership function;

defuzzifying the output of the at least one output membership function to determine whether or not the signal is acceptable or not.

2. The method of claim 1, wherein the sensor signal is a plurality of numeric values indicative of an electronic camera image.

3. The method of claim 2, wherein the at least one numeric sensor value is indicative of the boundary between a cut portion and an uncut portion of an agricultural field, and further wherein the step of deriving includes the step of analyzing the camera image to reduce the plurality of numeric values to the at least one numeric sensor signal value.

4. The method of claim 1, wherein the step of processing the sensor signal includes the steps of:

processing a first of the plurality of numeric sensor values with a first of the plurality of fuzzy logic input membership functions, wherein the first input membership function is indicative of the acceptability/unacceptability of the first numeric sensor value;

processing a second of the plurality of numeric sensor values with the second of the plurality of fuzzy logic membership functions, wherein the second input membership function is indicative of the acceptability/unacceptability of the second numeric sensor value.

5. The method of claim 4, wherein the step of combining the outputs includes the step of combining the outputs of the first and second membership functions in at least one fuzzy logic output membership function.

6. The method of claim 4, wherein the first numeric sensor value is derived from a first electronic image.

7. The method of claim 6, wherein the second numeric sensor value is derived from the first electronic image.

8. A method of processing a sequence of electronic camera images using fuzzy logic comprising the steps of:

taking a first electronic camera image indicative of a scene including cut and uncut vegetation in an agricultural field by an electronic camera;

deriving from the first image a first numeric value indicative of a first characteristic of the first image;

processing the first numeric value with at lest a first fuzzy logic input membership function to provide a first numeric input value;

combining the first numeric input value with at least a first fuzzy logic output membership function to provide a first numeric output value;

defuzzifying the first numeric output value to determine whether or not the first numeric value properly characterizes the first image or not; and

saving the first numeric value.

9. The method of claim 8, further comprising the steps of:

deriving from the first image a second numeric value indicative of a second characteristic of the first image;

processing the second numeric value with at least a second fuzzy logic input membership function to provide a second numeric input value.

10. The method of claim 9, wherein the first and second characteristics are both indicative of a boundary line defined between a cut and an uncut region of the field.

11. The method of claim 9, wherein the first and second characteristics are indicative of a row of uncut crop.

12. The method of claim 9, wherein the first and second characteristics define a line substantially parallel to a row of uncut crop.

13. The method of claim 9, wherein the step of combining includes the step of combining first numeric input value and the second numeric input value with the at least a first fuzzy logic output membership function to provide the first numeric output value.

14. The method of claim 8, wherein the step of Defuzzifying the first numeric input value includes the step of determining that the first numeric value is acceptable, the method further comprising the steps of:

modifying the first fuzzy logic input membership function with the acceptable first numeric input value;

taking a second electronic camera image indicative of a scene including cut and uncut vegetation in an agricultural field by an electronic camera;

deriving from the second image a second numeric value indicative of the first characteristic of the first image;

processing the second numeric value with the previously modified at least a first fuzzy logic input membership function to provide a second numeric input value;

combining the second numeric input value with the at least first fuzzy logic output membership function to provide a second numeric output value;

defuzzifying the second numeric output value to determine whether or not the second numeric value properly characterizes the first image or not.

15. The method of claim 14, wherein the step of modifying the first fuzzy logic input membership function includes the step of changing the x-axis of the membership function based upon the value of the acceptable first numeric output value.

16. The method of claim 14, wherein the step of modifying the first fuzzy logic input function includes the step of combining the first numeric output value and the second numeric output value to provide a third numeric output value indicative of the difference between the first and second numeric output values.

17. In a work vehicle including a ground engaging implement having a camera for taking successive electronic images of a portion of ground and/or vegetation in front of the vehicle, and a guidance control system responsive to successive images generated by the camera, a method for determining whether each of the successive images transmitted by the camera to the guidance system should be used to guide the vehicle, the method comprising the steps of:

taking successive images of the ground in front of the vehicle with the camera;

successively processing the images to provide, for one image at a time, a first numerical value indicative of a characteristic of the ground and/or vegetation in front of the vehicle with the guidance system, wherein the numerical value is indicative of a line of travel of the vehicle;

determining whether or not the first numerical value is acceptable using at least two fuzzy logic input functions;

combining the results of the at least two fuzzy logic input functions as inputs to a fuzzy logic output function;

based upon a numerical result generated by the fuzzy logic output functions, determining whether the first numerical value should be used to guide the vehicle.

18. The method of claim 17, wherein a first of the at least two fuzzy logic input functions varies in its output both when a previously calculated first numerical value from the processing of a previous image changes and when the first numerical value of the said processing of a previous image changes.

19. The method of claim 18, wherein the previously calculated first numerical value from the processing of a previous image was calculated by:

taking a previous image of the ground in front of the vehicle with the camera; and

processing the previous image to provide the previously acceptable numerical value indicative of a characteristic of the ground and/or vegetation in front of the vehicle with the guidance system, wherein the previously acceptable numerical value is indicative of a line of travel of the vehicle;

determining whether or not the first numerical value is acceptable using the at least two fuzzy logic input functions;

combining the results of the at least two fuzzy logic input functions as inputs to a fuzzy logic output function;

determining if the previously calculated first numerical value should be used to guide the vehicle based upon a numerical result generated by the fuzzy logic output function.

20. The method of claim 18, wherein an output of a second of the at least two fuzzy logic input functions varies when the first numerical value changes.

21. The method of claim 20, further comprising the steps of:

processing the image to provide a second numerical value indicative of a characteristic of the ground and/or vegetation in front of the vehicle with the guidance system, wherein the second numerical value is indicative of a line of travel of the vehicle; and

determining whether or not the second numerical value is acceptable using the at least two fuzzy logic input functions.