Pre-filter for object detection

• US 8,300,973 B2
• Filed: 12/10/2007
• Issued: 10/30/2012
• Est. Priority Date: 01/30/2007
• Status: Expired due to Fees

• Current Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Greig, Darryl
• Primary Examiner(s): Do, Anh Hong
• Application Number:
  US11/953,221
• Publication Number:
  US 20080181455A1
• Time in Patent Office: 1,786 Days
• Field of Search: 382/103, 382/260, 382/118, 382/117, 382/106, 382/190, 382/228, 382/297, 382/296, 382/289, 382/265, 382/274, 250/370.09, 378/17, 715/863, 340/5.2, 340/5.83, 348/208.13, 358/462, 358/448
• US Class Current: 382/260
• CPC Class Codes: G06V 10/443 by matching or filtering
• CPC Class Codes: G06V 40/161 Detection; Localisation; Normalisation

Abstract

A method of pre-processing an image to identify processes for subsequent processing of the image, comprising the steps of:

• a) investigating portions of the image using a spatial filter;
• b) calculating for a first plurality of regions within a portion of the image under investigation respective metrics as a function of intensity within those regions;
• c) selecting combinations of regions within the portion of the image under investigation and processing them to obtain a second plurality of filter values, where the second plurality is greater than the first plurality; and
• d) comparing the filter values with process thresholds for subsequent processes so as to identify subsequent processes that can be skipped.

Claims

1. A method, comprising the steps of:

applying, by the processor, a spatial filter to a portion of an image, the spatial filter comprising a plurality of test regions;

calculating, by the processor, for each of the test regions within the portion of the image a metric as a function of intensity within the test region;

determining, by the processor, for the portion of the image a filter value based on the metrics of each of the test regions; and

comparing, by the processor, the filter value with a threshold to identify a subsequent process that can be skipped.

2. The method of claim 1, the spatial filter comprising at least first and second overlapping test regions, the second test region enclosed by the first test region.

3. The method of claim 2, the spatial filter comprising a third test region, the third region enclosed by the second region.

4. The method of claim 2, wherein the overlapping test regions are concentric.

5. The method of claim 1, the spatial filter comprising a centrally disposed first test region and at least four other test regions disposed symmetrically around the central region.

6. The method of claim 5, the at least four other test regions are equidistant from the first region.

7. The method of claim 6, each of the other test regions is rectangular and the centres of the rectangles are equally displaced from the centre of the first test region.

8. The method of claim 1, comprising varying, by the processor, the sizes of one or more of the test regions within the spatial filter during the determination of the filter value.

9. The method of claim 1, comprising applying, by the processor, the spatial filter to a plurality of portions of the image.

10. The method claim 3, the first test region comprising a first square having a width of W^o, the second test region comprising a second square having a width of W^m, and the third test region comprising a third square having a width of W^I, and the method comprising:

calculating an average intensity value for the area enclosed by each square.

11. The method of claim 1, the spatial filter comprising three test regions, a first square, a second square and a third square, and the method comprising:

calculating, by the processor, a first average intensity value for the area bounded between the first and second squares,

calculating, by the processor, a second average intensity value for the area bounded between the second and third squares; and

calculating, by the processor, a third average intensity value is calculated for the area bounded by the third square.

12. The method of claim 10, the average intensity values are identified as A(W^o) for the first square, A(W^m) for the second square and A(W^I) for the third square, and the filter value for the area under investigation is calculated by:

D=A(W^o)−

2A(W^m)+A(W^I).

13. The method of claim 1, comprising calculating, by the processor, a plurality of filter values for a plurality of different combinations of test regions for the portion of the image, each filter value for each combination is based on an average image intensity for each test region in the combination.

14. The method of claim 10, comprising calculating, by the processor, a plurality of filter values based on a plurality of average intensity values for a plurality of square sizes.

15. The method of claim 13, comprising determining, by the processor, for each of the calculated filter values when an associated threshold for each of one or more subsequent processes is exceeded.

16. The method of claim 15, comprising executing, by the processor, the subsequent processes corresponding with each of the calculated filter values exceeding the associated threshold.

17. The method of claim 1, the subsequent process comprising an orientation specific image recognition process.

18. The method of claim 1, the subsequent process comprising a face detection process.

19. The method of claim 14, executing, by the processor, subsequent processes when one or more of the filter values exceed an associated threshold.

20. The method of claim 1, comprising comparing, by the processor, one of average intensity of the image within a plurality portions of the spatial filter and a variance within or between the portions of the spatial filter.

21. A method as claimed in claim 1, the spatial filter exhibits at least four degrees of rotational symmetry.

22. A method as claimed in claim 1, comprising varying, by the processor, threshold values within the calculation of the filter values depending on a selected combination of test regions.

23. A system comprising:

a memory; and

a data processor to;

apply a spatial filter to a portion of an image, the spatial filter comprising a plurality of test regions;

calculate for each of the test regions within the portion of the image a metric as a function of intensity within the test region;

determine for the portion of the image a filter value based on the metrics of each of the test regions; and

compare, the filter value with a threshold to identify a subsequent process that can be skipped.

24. A computer program stored in a computer readable medium that, when executed, enable a data processor to:

apply a spatial filter to a portion of an image, the spatial filter comprising a plurality of test regions;

calculate for each of the test regions within the portion of the image a metric as a function of intensity within the test region;

determine for the portion of the image a filter value based on the metrics of each of the test regions; and

compare, the filter value with a threshold to identify a subsequent process that can be skipped.