Parameter selection and coarse localization of interest regions for MSER processing

• US 9,183,458 B2
• Filed: 03/12/2013
• Issued: 11/10/2015
• Est. Priority Date: 07/19/2012
• Status: Expired due to Fees

• Current Assignee: Qualcomm, Inc.
• Original Assignee: Qualcomm, Inc.
• Inventors: Baheti, Pawan Kumar, Barman, Kishor K., Gore, Dhananjay Ashok, Sundaram, Senthilkumar
• Primary Examiner(s): Wu, Jingge
• Application Number:
  US13/796,729
• Publication Number:
  US 20140023270A1
• Time in Patent Office: 973 Days
• US Class Current: 1/1
• CPC Class Codes: G06V 20/63 Scene text, e.g. street names
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/147 Determination of region of interest
• CPC Class Codes: G06V 30/18076 by analysing connectivity, e.g. edge linking, connected component analysis or slices
• CPC Class Codes: G06V 30/18086 by performing operations within image blocks or by using histograms

Abstract

An attribute is computed based on pixel intensities in an image of the real world, and thereafter used to identify at least one input for processing the image to identify at least a first maximally stable extremal region (MSER) therein. The at least one input is one of (A) a parameter used in MSER processing or (B) a portion of the image to be subject to MSER processing. The attribute may be a variance of pixel intensities, or computed from a histogram of pixel intensities. The attribute may be used with a look-up table, to identify parameter(s) used in MSER processing. The attribute may be a stroke width of a second MSER of a subsampled version of the image. The attribute may be used in checking whether a portion of the image satisfies a predetermined test, and if so including the portion in a region to be subject to MSER processing.

Claims

1. A method to identify regions in images, the method comprising:

receiving an image of a scene of real world;

with one or more processors, computing an attribute based on pixel intensities in the image;

with the one or more processors, using the attribute with a lookup table or a predetermined test, to identify at least one input to be used in processing the image to identify at least one maximally stable extremal region therein;

wherein the at least one input is one of (A) a parameter Δ

or Max Variation or both obtained by use of at least said attribute from the lookup table and used in said processing or (B) a portion of the image obtained prior to said processing by applying the predetermined test to a subsampled region in the image, the portion to be subject to said processing, or both (A) and (B);

with the one or more processors, performing said processing to identify said at least one maximally stable extremal region based on said at least one input;

wherein said processing comprises at least comparing a difference in intensities of a pair of pixels in the image to a predetermined limit, adding to a list, a pair of coordinates of a pixel in said pair of pixels, in response to finding said predetermined limit is exceeded, and repeating said comparing and said adding; and

with the one or more processors, storing in one or more memories, the list as a representation of the at least one maximally stable extremal region identified by said processing.

2. The method of claim 1 wherein:

the attribute is computed from a histogram of said pixel intensities as a function of number of pixels at each intensity in said pixel intensities.

3. The method of claim 2 wherein:

the attribute is based on a plurality of bins in said histogram with corresponding counts of pixels above a threshold.

4. The method of claim 3 wherein:

the threshold is a fraction of a maximum count among the plurality of bins in said histogram.

5. The method of claim 3 wherein:

the attribute is an area of the histogram above a mean of counts of pixels in the plurality of bins of said histogram.

6. The method of claim 1 wherein:

the attribute is a variance of said pixel intensities.

7. The method of claim 1 wherein:

the attribute is used with a look-up table to identify the parameter.

8. The method of claim 1 wherein:

the attribute is used in checking whether the portion satisfies the predetermined test.

9. The method of claim 8 wherein when the portion is found to satisfy the predetermined test:

prior to said processing with the one or more processors, cropping from the image a rectangular region defined by a smallest rectangle that fits the portion.

10. The method of claim 1 wherein said at least one maximally stable extremal region is hereinafter first maximally stable extremal region, the method further comprising:

subsampling the image to obtain a subsampled version;

processing the subsampled version to identify a second maximally stable extremal region in the subsampled version of said image; and

using a smallest dimension of the second maximally stable extremal region to identify said portion to be subject to said processing.

11. A mobile device to identify regions in images, the mobile device comprising:

one or more memories comprising a plurality of portions of an image of a scene of real world;

one or more processors configured to;

compute an attribute based on pixel intensities in the image;

use the attribute, with a lookup table or a predetermined test, to identify at least one input to be used in processing the image to identify at least one maximally stable extremal region therein;

wherein the at least one input is one of (A) a parameter Δ

or Max Variation or both obtained from the lookup table and used in said processing or (B) a portion of the image to be subject to said processing, the portion being obtained prior to said processing by applying the predetermined test to a subsampled region in the image or both (A) and (B);

perform said processing to identify said at least one maximally stable extremal region based on said at least one input;

wherein said processing comprises at least comparing a difference in intensities of a pair of pixels in the image to a predetermined limit, adding to a list, a pair of coordinates of a pixel in said pair of pixels, in response to finding said predetermined limit is exceeded, and repeating said comparing and said adding; and

store in said one or more memories, the list as a representation of the at least one maximally stable extremal region identified by said processing.

12. The mobile device of claim 11 wherein:

the attribute is computed from a histogram of said pixel intensities as a function of number of pixels at each intensity in said pixel intensities.

13. The mobile device of claim 12 wherein:

the attribute is based on a plurality of bins in said histogram with corresponding counts of pixels above a threshold.

14. The mobile device of claim 13 wherein:

the threshold is a fraction of a maximum count among the plurality of bins in said histogram.

15. The mobile device of claim 13 wherein:

the attribute is an area of the histogram above a mean of counts of pixels in the plurality of bins of said histogram.

16. The mobile device of claim 11 wherein:

the attribute is a variance of said pixel intensities.

17. The mobile device of claim 11 wherein:

the one or more processors are further configured to use the attribute with a look-up table to identify the parameter.

18. The mobile device of claim 11 wherein:

the one or more processors are further configured to use the attribute in checking whether the portion satisfies the predetermined test.

19. The mobile device of claim 18 wherein the one or more processors are further configured to respond to finding the portion to satisfy the predetermined test by:

prior to said processing, cropping from the image a rectangular region defined by a smallest rectangle that fits the portion.

20. The mobile device of claim 18 wherein said at least one maximally stable extremal region is hereinafter first maximally stable extremal region, and the one or more processors are further configured to:

subsample the image to obtain a subsampled version;

to identify a second maximally stable extremal region in the subsampled version of the image; and

use a smallest dimension of the second maximally stable extremal region to identify said portion to be subject to said processing.

21. One or more non-transitory computer-readable media comprising a plurality of instructions to one or more processors to perform a method, the plurality of instructions comprising:

first instructions to receive an image of a scene of real world;

second instructions to compute an attribute based on pixel intensities in the image;

third instructions to use the attribute, with a lookup table or a predetermined test, to identify at least one input to be used in processing the image to identify at least one maximally stable extremal region therein;

wherein the at least one input is one of (A) a parameter Δ

or Max Variation or both obtained from the lookup table and used in said processing or (B) a portion of the image to be subject to said processing, the portion being obtained prior to said processing by applying the predetermined test to a subsampled region in the image or both (A) and (B);

fourth instructions to perform said processing to identify said at least one maximally stable extremal region based on said at least one input;

wherein said processing comprises at least comparing a difference in intensities of a pair of pixels in the image to a predetermined limit, adding to a list, a pair of coordinates of a pixel in said pair of pixels, in response to finding said predetermined limit is exceeded, and repeating said comparing and said adding; and

fifth instructions to store in one or more memories, the list as a representation of the at least one maximally stable extremal region identified by said processing.

22. The one or more non-transitory computer-readable media of claim 21 wherein:

the attribute is computed from a histogram of said pixel intensities as a function of number of pixels at each intensity in said pixel intensities.

23. The one or more non-transitory computer-readable media of claim 22 wherein:

the attribute is based on a plurality of bins in said histogram with corresponding counts of pixels above a threshold.

24. The one or more non-transitory computer-readable media of claim 23 wherein:

the threshold is a fraction of a maximum count among the plurality of bins in said histogram.

25. The one or more non-transitory computer-readable media of claim 23 wherein:

the attribute is an area of the histogram above a mean of counts of pixels in the plurality of bins of said histogram.

26. The one or more non-transitory computer-readable media of claim 21 wherein:

the attribute is a variance of said pixel intensities.

27. The one or more non-transitory computer-readable media of claim 21 further comprising:

sixth instructions to use the attribute with a look-up table to identify the parameter.

28. The one or more non-transitory computer-readable media of claim 21 further comprising:

sixth instructions to use the attribute in checking whether the portion satisfies the predetermined test.

29. The one or more non-transitory computer-readable media of claim 28 further comprising:

prior to execution of the fourth instructions, seventh instructions to crop from the image a rectangular region defined by a smallest rectangle that fits the portion.

30. The one or more non-transitory computer-readable media of claim 21 wherein said at least one maximally stable extremal region is hereinafter first maximally stable extremal region, the one or more non-transitory computer-readable media further comprising:

sixth instructions to subsample the image to obtain a subsampled version;

seventh instructions to process the subsampled version to identify a second maximally stable extremal region in the subsampled version of the image; and

eighth instructions to use a smallest dimension of the second maximally stable extremal region to identify said portion to be subject to said processing.

31. An apparatus to identify regions in images, the apparatus comprising:

means for receiving an image of a scene of real world;

means for computing an attribute based on pixel intensities in the image;

means for using the attribute, with a lookup table or a predetermined test, to identify at least one input to be used in processing the image to identify at least one maximally stable extremal region therein;

wherein the at least one input is one of (A) a parameter Δ

or Max Variation or both obtained from the lookup table and used in said processing or (B) a portion of the image to be subject to said processing, the portion being obtained prior to said processing by applying the predetermined test to a subsampled region in the image or both (A) and (B);

means for performing said processing to identify said at least one maximally stable extremal region based on said at least one input;

wherein said processing comprises at least comparing a difference in intensities of a pair of pixels in the image to a predetermined limit, adding to a list, a pair of coordinates of a pixel in said pair of pixels, in response to finding said predetermined limit is exceeded, and repeating said comparing and said adding; and

means for storing in one or more memories, the list as a representation of the at least one maximally stable extremal region identified by said processing.

32. The apparatus of claim 31 wherein:

the attribute is computed from a histogram of said pixel intensities as a function of number of pixels at each intensity in said pixel intensities.

33. The apparatus of claim 31 wherein:

the attribute is used with a look-up table to identify the parameter.

34. The apparatus of claim 31 further comprising:

means for subsampling the image to obtain a subsampled version;

means for identifying an additional maximally stable extremal region in the subsampled version of the image; and

means for using a smallest dimension of the additional maximally stable extremal region to identify said portion to be subject to said processing.