Image collating apparatus, image collating method, image collating program product, and computer readable recording medium recording image collating program product

• US 7,697,733 B2
• Filed: 09/28/2006
• Issued: 04/13/2010
• Est. Priority Date: 09/28/2005
• Status: Expired due to Fees

• Current Assignee: Sharp Kabushiki Kaisha (Hon Hai Precision Industry Co., Ltd.)
• Original Assignee: Sharp Kabushiki Kaisha (Hon Hai Precision Industry Co., Ltd.)
• Inventors: Yumoto, Manabu, Onozaki, Manabu
• Primary Examiner(s): Lu, Tom Y
• Application Number:
  US11/528,393
• Publication Number:
  US 20070071359A1
• Time in Patent Office: 1,293 Days
• Field of Search: 382/115, 382/124
• US Class Current: 382/115
• CPC Class Codes: G06V 10/7515 Shifting the patterns to accommodate for positional errors

Abstract

An image collating apparatus includes a storage device and a processor. The storage device stores data representing a sensing image, data representing a template image, and a program. The processor detects exclusion areas in the sensing image to be excluded from the image collation operation based on classification of image patterns. The processor determines whether or not the sensing image and the template image are derived from the identical object by performing template-matching operations to evaluate similarity between the template image and partial images of the sensing image in different levels with increasing the size of area covered by the partial image to be matched with the template image level by level. At least in one level the processor performs the template-matching operation with respect to the partial image where the exclusion areas are excluded.

Claims

1. An image collating apparatus for performing an image collation operation comprising:

a storage device storing image data representing a sensing image of an object sensed by a sensor, template data representing a template image, and a program; and

a processor receiving commands from said program for

(a) detecting exclusion areas in said sensing image to be excluded from said image collation operation based on classification of image patterns,

(b) determining whether or not said sensing image and said template image are derived from the identical object by performing template-matching operations to evaluate similarity between said template image and partial images of said sensing image in different levels, in said different levels said processor increasing the size of area covered by said partial image to be matched with said template image level by level, and at least in one level said processor performing the template-matching operation with respect to said partial image where said exclusion areas are excluded,

wherein each of said partial images includes a plurality of partial areas of said sensing image, and said processor, in the process for determining, determines whether a first partial image among said partial images includes said exclusion area, said first partial image including said partial areas at predetermined positions in the first level, and when said first partial image includes said exclusion area, skips said first level and then determines in the second level whether or not said sensing image and said template image are derived from the identical object based on the template-matching operation between said template image and a second partial image among said partial images which is larger than said first partial image and other than said exclusion areas.

2. The image collating apparatus according to claim 1, wherein said processor, in the process for determining,

(b-1) determines that said sensing image and said template image are derived from the identical object when a first partial image among said partial images matches with any portion of said template image, said exclusion areas being excluded from said first partial image and

(b-2) when said first partial image does not match with said template image, determines whether or not said sensing image and said template image are derived from the identical object based on the template-matching operation between said template image and a second partial image among said partial images which is larger than said first partial image and other than said exclusion areas.

3. The image collating apparatus according to claim 1, wherein

said processor specifies, in the process for detecting, characteristic information which corresponds to classification of the image patterns in arrangements of pixels of an identical color in said partial image, and detects said excluded areas based on said characteristic information.

4. The image collating apparatus according to claim 3, wherein each of said partial images includes a plurality of partial areas of said sensing image, and

said exclusion area includes a first partial area and second partial areas adjacent to the first partial area, wherein said first and second partial areas correspond to the same classification category.

5. The image collating apparatus according to claim 1, wherein

each of said partial images includes a plurality of partial areas of said sensing image,

said processor searches for a most similar portion in said template image for each said partial area, and

said processor calculates similarity score between at least a part of said sensing image and at least a part of said template image based on positional relation between each said partial area and said searched portion.

6. The image collating apparatus according to claim 5, wherein

said processor, in the first level of said different levels, determines that said partial image matches with any portion of said template image if said similarity score is equal to or greater than a first threshold value, and determines that said partial image is different from any portion of said template image if said similarity score is smaller than a second threshold value.

7. The image collating apparatus according to claim 5, wherein

said processor, in the process for calculating, calculates the number of partial areas which satisfies a prescribed requirement of positional relation between a plurality of said partial areas and said searched portion.

8. The image collating apparatus according to claim 1, wherein

said partial image in the first level includes a partial area located in a center of said sensing image.

9. The image collating apparatus according to claim 1, wherein

said processor determines whether or not said sensing image and said template image are derived from the identical object based on different threshold values in the different levels.

10. The image collating apparatus according to claim 1, wherein

said storage device includes a device storing in advance the data representing said template image.

11. The image collating apparatus according to claim 1, further comprising an input unit receiving data of an image of fingerprint as said image data representing the sensing image.

12. A method of collating an image for performing an image collation operation, comprising the steps of:

storing image data representing a sensing image of an object sensed by a sensor, and template data representing a template image;

detecting exclusion areas in said sensing image to be excluded from said image collation operation based on classification of image patterns; and

determining whether or not said sensing image and said template image are derived from the identical object by performing template-matching operations to evaluate similarity between said template image and partial images of said sensing image in different levels, in said different levels said processor increasing the size of area covered by said partial image to be matched with said template image level by level, wherein said step of determining includes the step performing the template-matching operation of at least in one level with respect to said partial image where said exclusion areas are excluded,

wherein each of said partial images includes a plurality of partial areas of said sensing image, and said processor, in the process for determining, determines whether a first partial image among said partial images includes said exclusion area, said first partial image including said partial areas at predetermined positions in the first level, and when said first partial image includes said exclusion area, skips said first level and then determines in the second level whether or not said sensing image and said template image are derived from the identical object based on the template-matching operation between said template image and a second partial image among said partial images which is larger than said first partial image and other than said exclusion areas.

13. A computer readable recording medium recording an image collating program for performing an image collation operation causing a computer to execute the steps of:

storing image data representing a sensing image of an object sensed by a sensor and template data representing a template image;

detecting exclusion areas in said sensing image to be excluded from said image collation operation based on classification of image patterns; and

determining whether or not said sensing image and said template image are derived from the identical object by performing template-matching operations to evaluate similarity between said template image and partial images of said sensing image in different levels, in said different levels said processor increasing the size of area covered by said partial image to be matched with said template image level by level, wherein said step of determining includes the step performing the template-matching operation of at least in one level with respect to said partial image where said exclusion areas are excluded,

wherein each of said partial images includes a plurality of partial areas of said sensing image, and said processor, in the process for determining, determines whether a first partial image among said partial images includes said exclusion area, said first partial image including said partial areas at predetermined positions in the first level, and when said first partial image includes said exclusion area, skips said first level and then determines in the second level whether or not said sensing image and said template image are derived from the identical object based on the template-matching operation between said template image and a second partial image among said partial images which is larger than said first partial image and other than said exclusion areas.