What is claimed is:
1. A method comprising the steps of:
determining a signature metric for a portion of a first image;
searching a second image for at least one relative match of the signature metric, yielding one or more candidate regions;
in a pixel domain, searching at least one of the one or more candidate regions for the portion of the first image to obtain a motion vector for the portion of the first image.
2. The method of claim 1, wherein the portion of the first image is in signature space.
3. The method of claim 1, wherein the step of determining the signature metric comprises determining a vertical signature metric and a horizontal signature metric.
4. The method of claim 3, wherein the vertical signature metric comprises vertically projected image data.
5. The method of claim 3, wherein the horizontal signature metric comprises horizontally projected image data.
6. The method of claim 1, wherein the step of searching the second image comprises searching less than all of the second image.
7. The method of claim 1, wherein the step of searching in the pixel domain further comprises the step of searching a region defined by a zeroth motion vector, yielding an additional motion vector.
8. The method of claim 7, further comprising the step of comparing the motion vector and the additional motion vector, yielding a final motion vector.
9. A method comprising the steps of:
employing a portion of a signature space to identify a candidate region in an image space of a possible relative match with a current image; and
searching a part of the image space based at least in part on a portion of the candidate region to obtain a motion vector for at least a part of the current image.
10. The method of claim 9, further comprising the step of determining a signature metric.
11. The method of claim 10, wherein the step of determining the signature metric comprises determining a vertical signature metric and a horizontal signature metric.
12. The method of claim 11, wherein the vertical signature metric comprises vertically projected image data.
13. The method of claim 11, wherein the horizontal signature metric comprises horizontally projected image data.
14. The method of claim 9, wherein the step of searching the part of the image space further comprises the step of searching a region defined by a zeroth motion vector, yielding an additional motion vector.
15. The method of claim 14, further comprising the step of comparing the motion vector and the additional motion vector, yielding a final motion vector.
16. A method comprising the steps of:
comparing a signature metric for a portion of a first image to projected data in a search window of a second image, yielding at least one coarse motion vector;
on a pixel basis, searching at least one region related to the at least one coarse motion vector for image data, yielding a fine motion vector.
17. The method of claim 16, further comprising the step of determining a signature metric that comprises a vertical signature metric and a horizontal signature metric.
18. The method of claim 16, wherein the search window is smaller than all of the second image.
19. The method of claim 17, wherein the step of searching further comprises the step of searching a region defined by a zeroth motion vector, yielding a first motion vector.
20. The method of claim 18, wherein the step of searching the at least one region yields a second motion vector, and the first motion vector and the second motion vector are compared to find a better motion vector, yielding a final motion vector.
21. An apparatus comprising:
a projection data comparator, arranged and constructed to compare a signature metric for a portion of a first image to projected data in a search window of a second image, yielding at least one coarse motion vector; and
a pixel domain comparator, arranged and constructed to, on a pixel basis, search at least one region related to the at least one coarse motion vector for image data, yielding a fine motion vector.
22. The apparatus of claim 21, further comprising a signature metric determiner that provides a vertical signature metric and a horizontal signature metric.
23. The apparatus of claim 21, wherein the pixel domain comparator is further arranged and constructed to yield a first motion vector and a second motion vector, and to compare the first motion vector and the second motion vector to find a better motion vector, yielding a final motion vector.
The claims below are in addition to those above.
All refrences to claim(s) which appear below refer to the numbering after this setence.
What is claimed is:
1. An apparatus for judging road surface gradients comprising a speed detecting means for periodically detecting wheel speeds of four wheels of a vehicle, a vehicle body speed calculating means for obtaining a vehicle body speed of the vehicle, a slip rate calculating means for obtaining a slip rate, a means for calculating acceleration of the vehicle body for obtaining an acceleration of the vehicle body of the vehicle, a determining means for determining whether the acceleration of the vehicle body is within a specified range proximate to zero, and a gradient judging means for judging a gradient of a road surface on the basis of the determination of the determining means.
2. The apparatus of claim 1, wherein the gradient judging means judges a road surface gradient, in case the acceleration of the vehicle body is within a specified range which is proximate to zero, on the basis of values of slip rates which are within the specified range.
3. The apparatus of claim 1, wherein the gradient judging means performs 5 liner regression of the slip rates and acceleration of the vehicle body to judge a road surface gradient by using values of slip rates obtained when the acceleration of the vehicle body is zero, in case the acceleration of the vehicle is not within a specified range which is proximate to zero within a specified period of time.
4. The apparatus of any one of claims 2 to 3, wherein the obtained slip rate is corrected in accordance with the vehicle body speed of the vehicle.
5. A method for judging road surface gradients comprising the steps of: periodically detecting wheel speeds of four wheels of a vehicle, obtaining a vehicle body speed of the vehicle, obtaining a slip rate, obtaining an acceleration of the vehicle body of the vehicle, determining whether the acceleration of the vehicle body is within a specified range proximate to zero, and judging a gradient of a road surface on the basis of the determination.
6. The method of claim 5, wherein a road surface gradient is judged, in case the acceleration of the vehicle body is within a specified range which proximate to zero, on the basis of values of slip rates which are within the specified range.
7. The method of claim 5, wherein a road surface gradient is judged by performing linear regression of the slip rates and acceleration of the vehicle body to judge the road surface gradient by using values of slip rates obtained when the acceleration of the vehicle body is zero, in case the acceleration of the vehicle is not within a specified range which is proximate to zero within a specified period of time
8. The method of any one of claims 6 to 7, wherein the obtained slip rate is corrected in accordance with the vehicle body speed of the vehicle.
9. A program for judging gradients which makes a computer function as a vehicle body speed calculating means for obtaining a vehicle body speed of the vehicle, a slip rate calculating means for obtaining a slip rate, a means for calculating acceleration of the vehicle body for obtaining an acceleration of the vehicle body of the vehicle, a determining means for determining whether the acceleration of the vehicle body is within a specified range proximate to zero, and a gradient judging means for judging a gradient of a road surface on the basis of the determination of the determining means.