Image processing apparatus with document similarity processing, and image processing method and storage medium therefor

Patent metadata

Field	Value
Publication number	US-10817559-B2
Application number	US-201816122624-A
Country	US
Kind code	B2
Filing date	Sep 5, 2018
Priority date	Sep 21, 2017
Publication date	Oct 27, 2020
Grant date	Oct 27, 2020

Abstract

To make it possible to search for a document of the same kind as that of a document relating to a scanned image both highly accurately and simply. An image processing apparatus including: a calculation unit configured to calculate a degree of similarity between an input document image and each of a plurality of document images by repeatedly performing calculation of a degree of similarity while changing each range including a specific area, which is a calculation target of a degree of similarity, in the input document image and the plurality of document images; and a determination unit configured to determine a document image whose calculated degree of similarity is the highest of the plurality of document images as a document image that matches with the input document image.

First claim

What is claimed is:

1. An image processing apparatus comprising:
a processor; and
a memory for storing a computer executable program, wherein the processor executes the computer executable program to perform:
extracting at least one block having a text attribute by performing area division processing for the input document image;
calculating a degree of similarity between the input document image and each of a plurality of document images by repeatedly performing calculation of the degree of similarity based on a shape and arrangement of the extracted at least one blocks included in each of a plurality of different ranges, which are calculation targets, with respect to the input document image and each of the plurality of document images; and
determining a document image whose calculated degree of similarity is the highest of the plurality of document images as a document image that matches with the input document image.

2. The image processing apparatus according to claim 1 , wherein
each of the plurality of different ranges includes a specific area having a fixed structure in the format of the input document image and the plurality of document images.

3. The image processing apparatus according to claim 1 , wherein
the processor further executes a computer executable program to perform estimating of an amount of shift between the input document image and each of the plurality of document images by acquiring information on a pair of blocks, which indicates a correspondence relationship between the at least one block having the text attribute in the input document image and at least one block having the text attribute in each of the plurality of document images, and based on the acquired information on a pair of blocks, and
wherein in calculating the degree of similarity, the processor executes the computer executable program to perform:
positioning of the at least one block having the text attribute included in the input document image in accordance with the estimated amount of shift; and
calculation of the degree of similarity based on a shape and arrangement of the at least one block having the text attribute after the positioning.

4. The image processing apparatus according to claim 3 , wherein
the estimating sets a weight to each of the pair of blocks, generates a histogram of an amount of shift in each of pairs of blocks by using the weight, and estimates a final amount of shift between the input document image and each of the plurality of document images based on the histogram.

5. The image processing apparatus according to claim 4 , wherein
the setting of the weight is performed based on an overlap state in each of the pairs of blocks or a position of each of the pairs of blocks.

6. The image processing apparatus according to claim 5 , wherein
in the setting of the weight based on the overlap state in each of the pairs of blocks, the smaller a number of pairs made by the other block having a text attribute of the pair with another block having a text attribute, the higher a weight value that is set is.

7. The image processing apparatus according to claim 5 , wherein
in the setting of the weight based on the position of each of the pairs of blocks, for a pair of blocks included in the specific area, a weight value higher than that of a pair of blocks not included in the specific area is set.

8. The image processing apparatus according to claim 7 , wherein
the position of each of the pairs of blocks is specified by coordinates in a vertical direction in a document image and a different weight value is set in accordance with the coordinates.

9. The image processing apparatus according to claim 1 , wherein
the processor further executes a computer executable program to perform registering of the input document image as the plurality of document images.

10. The image processing apparatus according to claim 1 , wherein
the plurality of document images is registered in advance as image data including position information on a text block within each document image, and
the degree of similarity of the plurality of document images is calculated by using the position information on a text block, which is registered in advance.

11. The image processing apparatus according to claim 1 , wherein, in the calculating, the degree of similarity between the input document image and each of a plurality of document images is calculated by comparing the shape and arrangement of the extracted at least one blocks included in each of the plurality of different ranges of the input document image with a shape and arrangement of blocks included in each of the plurality of different ranges of each of the plurality of document images, and
wherein, the degree of similarity is calculated without using a content of a character strings described in the extracted at least one blocks included each of the plurality of different ranges of the input document image.

12. The image processing apparatus according to claim 1 , wherein, in the calculating, the degree of similarity between the input document image and each of a plurality of document images is calculated by correcting the plurality of degrees of similarity obtained by repetition of the calculation in accordance with a distribution of the plurality of degrees of similarity.

13. An image processing apparatus comprising:
a processor; and
a memory for storing a computer executable program, wherein
the processor executes the computer executable program to perform:
calculating a degree of similarity between the input document image and each of a plurality of document images by repeatedly performing calculation of the degree of similarity for each of a plurality of different ranges, which are calculation targets of the degree of similarity, with respect to the input document image and each of the plurality of document images, wherein at least one of a plurality of degrees of similarity obtained by repetition of the calculation is corrected in accordance with a distribution of the plurality of degrees of similarity, and
determining a document image whose degree of similarity after the correction is the highest as a document image that matches with the input document image.

14. A method comprising:
extracting to extract at least one block having a text attribute by performing area division processing for the input document image;
calculating a degree of similarity between the input document image and each of a plurality of document images by repeatedly performing calculation of the degree of similarity based on a shape and arrangement of the extracted at least one blocks included in each of a plurality of different ranges, which are calculation targets, with respect to the input document image and each of the plurality of document images; and
determining a document image whose calculated degree of similarity is the highest of the plurality of document images as a document image that matches with the input document image.

15. A non-transitory computer readable storage medium storing a program for causing a computer to perform a method comprising:
extracting to extract at least one block having a text attribute by performing area division processing for the input document image;
calculating a degree of similarity between the input document image and each of a plurality of document images by repeatedly performing calculation of the degree of similarity based on a shape and arrangement of the extracted at least one blocks included in each of a plurality of different ranges, which are calculation targets, with respect to the input document image and each of the pl

Assignees

• Canon Kk

Inventors

• Arakawa Junya

Classifications

• G06V30/414 Primary

  Extracting the geometrical structure, e.g. layout tree; Block segmentation, e.g. bounding boxes for graphics or text · CPC title

• G06V10/751

  Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching · CPC title

• G06V10/235

  based on user input or interaction · CPC title

• G06F16/93

  Document management systems · CPC title

• H04N2201/0094

  Multifunctional device, i.e. a device capable of all of reading, reproducing, copying, facsimile transception, file transception · CPC title