Defining image features and using features to monitor image transformations
View Patent ↗For monitoring an image transformation such as aspect ratio conversion, an image feature is defined by identifying a position in the image having a local spatial maximum value and then identifying four other positions in the image having local spatial minimum values such that the four minimum value positions surround the position of the maximum, a first pair of the minimums lie on a first line passing through maximum and a second pair of the minimums lie on a second line passing through the maximum.
1. A method of monitoring in a processor a transformation between a first image and a second image, comprising the steps of identifying at least one image feature in the first image and at least one image feature in the second image by, for each image, identifying a first position in the image having a local spatial extremum value of an image parameter of the said image; identifying at least two other positions in the image having local spatial extremum values of the said image parameter such that no straight line passes through the first position and both said other positions; and storing or outputting data defining said first position and said other positions at least relative to said first position.
2. A method according to claim 1 , wherein the local spatial extremum value at said first position is of opposite sense (maximum or minimum) to that of the local spatial extremum value at said at least two other positions.
3. A method according to claim 1 , wherein said data includes relative values of the said image parameter at the respective positions.
4. A method according to claim 1 ; comprising the steps of identifying at least two image feature in each image and computing transform coefficients from the relative positions of the image features in the respective images.
5. A method according to claim 1 , wherein
a. the step of identifying a first position in the image having a local spatial extremum value of an image parameter comprises the step of identifying a position in the image having a local spatial maximum value of an image parameter of the said image;
b. the step of identifying at least two other positions in the image having local spatial extremum values of the said image parameter such that no straight line passes through the first position and both said other positions comprises the step of identifying four other positions in the image having local spatial minimum values of the said image parameter such that
1. the said four minimum value positions surround the position of the said local spatial maximum value, and
2. a first pair of the said minimum value positions lie on a first line passing through the position of the said local spatial maximum and a second pair of the said minimum value positions lie on a second line passing through the position of the said local spatial maximum.
6. A method according to claim 5 in which the said image is spatially sampled and both of the said lines are aligned with the spatial sampling structure of the said image.
7. A method according to claim 5 in which pixel-value data describing the said maximum and the four said minima is output.
8. A method according to claim 7 in which luminance values are output.
9. A method according to claim 7 in which the said pixel-value data is derived from differences between pixel-values.
10. Image processing apparatus comprising a local maximum finder for identifying a first position in the image having a local spatial maximum value of an image parameter of the said image; and a minimum finder for identifying four other positions in the image having local spatial minimum values of the said image parameter such that
the said four minimum value positions surround the position of the said local spatial maximum value, and
a first pair of the said minimum value positions lie on a first line passing through the position of the said local spatial maximum and a second pair of the said minimum value positions lie on a second line passing through the position of the said local spatial maximum.
11. Image processing apparatus comprising an upstream feature detector for identifying at least N image features in a first image by identifying a first position in the first image having a local spatial extremum value of an image parameter of the said image; identifying at least two other positions in the first image having local spatial extremum values of the said image parameter such that no straight line passes through the first position and both said other positions; a downstream feature detector for identifying at least M image features in a second image by identifying a first position in the second image having a local spatial extremum value of an image parameter of the second image; identifying at least two other positions in the second image having local spatial extremum values of the said image parameter such that no straight line passes through the first position and both said other positions; a corresponding feature selector for selecting at least two detected image features in the second image which correspond respectively with detected image features in the first image and a transform coefficient unit for computing from the location of the selected and corresponding image features coefficients of a transform between the first image and the second image.
12. Apparatus according to claim 11 , wherein the local spatial extremum value at said first position is of opposite sense (maximum or minimum) to that of the local spatial extremum value at said at least two other positions.
13. Apparatus according to claim 12 , wherein
a. the step of identifying a first position in the image having a local spatial extremum value of an image parameter comprises the step of identifying a position in the image having a local spatial maximum value of an image parameter of the said image;
b. the step of identifying at least two other positions in the image having local spatial extremum values of the said image parameter such that no straight line passes through the first position and both said other positions comprises the step of identifying four other positions in the image having local spatial minimum values of the said image parameter such that
1. the said four minimum value positions surround the position of the said local spatial maximum value, and
2. a first pair of the said minimum value positions lie on a first line passing through the position of the said local spatial maximum and a second pair of the said minimum value positions lie on a second line passing through the position of the said local spatial maximum.
14. Apparatus according to claim 13 in which the said image is spatially sampled and both of the said lines are aligned with the spatial sampling structure of the said image.
15. Apparatus according to claim 11 in which pixel-value data describing the said maximum and the four said minima is utilised.
16. Apparatus according to claim 11 in which luminance values are utilised.
17. Apparatus according to claim 11 in which pixel-value data is derived from differences between pixel-values.