IMAGE MATCHING APPARATUS, IMAGE MATCHING METHOD, AND IMAGE DATA OUTPUT PROCESSING APPARATUS

• US 20090067727A1
• Filed: 09/08/2008
• Published: 03/12/2009
• Est. Priority Date: 09/10/2007
• Status: Active Grant

• Current Assignee: Sharp Kabushiki Kaisha (Hon Hai Precision Industry Co., Ltd.)
• Original Assignee: Sharp Kabushiki Kaisha (Hon Hai Precision Industry Co., Ltd.)
• Inventors: HIROHATA, Hitoshi
• Granted Patent: US 8,073,262 B2
• Time in Patent Office: Days
• US Class Current: 382/201
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/18076 by analysing connectivity, e.g. edge linking, connected component analysis or slices
• CPC Class Codes: G06V 30/18143 Extracting features based on salient regional features, e.g. scale invariant feature transform [SIFT] keypoints
• CPC Class Codes: H04N 1/00843 based on recognising a copy prohibited original, e.g. a banknote pattern authentication and markers therefor, forgery detection G06V20/95; character recognition G06V30/10; document-oriented image-based pattern recognition G06V30/40; testing paper currency or similar valuable papers for genuineness G07D7/00
• CPC Class Codes: H04N 1/00864 Modifying the reproduction, e.g. outputting a modified copy of a scanned original details of composing or otherwise geometrically modifying originals H04N1/387
• CPC Class Codes: H04N 1/00875 Inhibiting reproduction, e.g. by disabling reading or reproduction apparatus

Abstract

In an image matching apparatus of the present invention, only a connected region in which the number of pixels included therein exceeds a threshold value, among connected regions that are specified by a labeling process section, is sent to a centroid calculation process section from a threshold value processing section, and a centroid (feature point) of the connected region is calculated. When it is determined that a target document to be matched is an N-up document, the threshold value processing section uses, instead of a default threshold value, a variant threshold value that varies depending on the number of images laid out on the N-up document and a document size that are found and detected by an N-up document determination section and a document size detection section. This makes it possible to determine a similarity to a reference document with high accuracy even in a case of an N-up document, i.e., a case where each target image to be matched is reduced in size from an original image.

Claims

1. An image matching apparatus, comprising

a features extraction section for extracting features of an image from input image data, the features extraction section extracting the features based on a connected region in which pixels in a binarized image of the image data are connected to each other and the number of the pixels exceeds a default threshold value,

said apparatus (i) causing the features extraction section to extract the features from input image data of a target document to be matched, and (ii) determining whether or not the target document has a similar image to that of a preliminarily stored reference document based on a comparison of the features thus extracted with features of the image of the reference document;

said apparatus, further comprising;

an N-up document determination section that determines (i) whether or not the target document is an N-up document on which plural documents are laid out, and (ii) the number of images laid out on the N-up document when it is determined that the target document is an N-up document; and

a document size detection section that detects a document size of the target document,

when the N-up document determination section determines that the target document is an N-up document, the features extraction section extracting the features, by use of not the default threshold value but a variant threshold value that varies depending on the number of the images laid out on the N-up document and the document size that are found and detected by the N-up document determination section and the document size detection section, respectively, based on a connected region in which the number of pixels exceeds the variant threshold value.

2. The image matching apparatus as set forth in claim 1, wherein:

the features extraction section includes;

a connected region specifying section that specifies a connected region in which pixels in a binarized image are connected to each other;

a threshold value processing section that extracts a connected region in which the number of pixels exceeds a threshold value, based on a comparison of the number of pixels included in the connected region thus specified by the connected region specifying section with the threshold value;

a feature point calculation process section that calculates a feature point of the connected region extracted by the threshold value processing section; and

a features calculation section that calculates features of an image by use of the feature point calculated by the feature point calculation process section,

when it is determined that the target document is an N-up document, the threshold value processing section using, as the threshold value, the variant threshold value which varies depending on the number of the images laid out on the target document and the document size, instead of the default threshold value.

5. A computer-readable storage medium in which a program, which causes an image matching apparatus as set forth in claim 1 to operate, is stored.

3. An image data output processing apparatus that carries out an output process with respect to image data, the apparatus comprising:

an image matching apparatus; and

an output process control section that controls an output process with respect to image data of a target document to be matched in accordance with a determination result of the image matching apparatus,

said image matching apparatus, including;

a features extraction section for extracting features of an image from input image data, the features extraction section extracting the features based on a connected region in which pixels in a binarized image of the image data are connected to each other and the number of the pixels exceeds a default threshold value,

said image matching apparatus (i) causing the features extraction section to extract the features from input image data of a target document to be matched, and (ii) determining whether or not the target document has a similar image to that of a preliminarily stored reference document based on a comparison of the features thus extracted with features of the image of the reference document;

said image matching apparatus, further including;

an N-up document determination section that determines (i) whether or not the target document is an N-up document on which plural documents are laid out, and (ii) the number of images laid out on the N-up document when it is determined that the target document is an N-up document; and

a document size detection section that detects a document size of the target document,

when the N-up document determination section determines that the target document is an N-up document, the features extraction section extracting the features, by use of not the default threshold value but a variant threshold value that varies depending on the number of the images laid out on the N-up document and the document size that are found and detected by the N-up document determination section and the document size detection section, respectively, based on a connected region in which the number of pixels exceeds the variant threshold value.

4. An image matching method comprising the step of:

(a) extracting features of an image from input image data based on a connected region in which pixels in a binarized image of the image data are connected to each other and the number of the pixels exceeds a default threshold value,

wherein the features are extracted from input image data of a target document to be matched in the step (a), and it is determined whether or not the target document has a similar image to that of a preliminarily stored reference document, based on a comparison of the features thus extracted with features of the image of the reference document;

said method, further comprising the steps of;

(b) determining (i) whether or not the target document is an N-up document on which plural documents are laid out, and (ii) the number of images laid out on the N-up document when it is determined that the target document is an N-up document; and

(c) detecting a document size of the target document,

when it is determined that the target document is an N-up document in the step (b), the step (a) is carried out such that the features are extracted, by use of not the default threshold value but a variant threshold value that varies depending on the number of the images laid out on the N-up document and the document size that are found and detected by the N-up document determination section and the document size detection section, respectively, based on a connected region in which the number of pixels exceeds the variant threshold value.