Character recognition device, character recognition method and program

Patent metadata

Field	Value
Publication number	US-10049309-B2
Application number	US-201515327381-A
Country	US
Kind code	B2
Filing date	Oct 30, 2015
Priority date	Oct 31, 2014
Publication date	Aug 14, 2018
Grant date	Aug 14, 2018

Abstract

A character recognition device comprising: a storage means storing a plurality of dictionary databases corresponding to each of a plurality of degradation levels, said databases being generated from images wherein an original character image has been degraded using a plurality of different degradation levels; an image input means that receives input of images; an extraction means that extracts character images from said images; a usage order determination means that determines an usage order for dictionary databases on the basis of degradation degrees for extracted character images; and a character recognition means that uses dictionary databases in accordance with said usage order, recognizes characters included in the extracted character image and, if recognition results for the character recognition fulfill prescribed conditions, outputs the recognition results without performing recognition using the dictionary database next in said usage order. As a result, accurate and fast recognition of degraded character images can be performed.

First claim

What is claimed is:

1. A character recognition device, comprising:
a memory, configured to store a plurality of dictionary databases corresponding to each of a plurality of degradation levels, said plurality of dictionary databases being generated from images wherein an original character image has been degraded using a plurality of different degradation levels;
a data transmission interface, configured to receive input of images; and
a processor, configured to:
extract character images from said images;
determine an usage order for the plurality of dictionary databases on the basis of degradation degrees for extracted character images comprising:
calculate, according to corresponding relationships between the degradation degrees of the character images and the dictionary databases, a degradation level corresponding to the degradation degree of the character images; and
determine the usage order as first using the dictionary database corresponding to the calculated degradation level and then beginning to sequentially use dictionary databases corresponding to degradation levels approximate to the calculated degradation level; and

use dictionary databases in accordance with said usage order, recognize characters included in the extracted character images and, if recognition results do not fulfil prescribed conditions, perform character recognition using the dictionary database next in said usage order, if recognition results fulfil the prescribed conditions, output the recognition results without performing recognition using the dictionary database next in said usage order.

2. The character recognition device according to claim 1 , wherein
the corresponding relationships are generated based on results obtained by comparing a plurality of test images using all the dictionary databases corresponding to the plurality of degradation levels.

3. The character recognition device according to claim 1 , wherein
the prescribed conditions comprise a condition of setting the highest comparison score in the character recognition using the dictionary databases to be above a prescribed threshold.

4. The character recognition device according to claim 1 , wherein
the prescribed conditions comprise a condition of setting a difference between the highest comparison score and the second highest comparison score in the character recognition using the dictionary databases to be above a prescribed threshold.

5. The character recognition device according to claim 1 , wherein
the degradation degree of the character image is determined based on a difference between a size of the extracted character image by the processor and a size after standardization processing.

6. The character recognition device according to claim 5 , wherein
in the standardization processing, the character image is transformed into a square with a prescribed size, and
the degradation degree of the character image is determined based on a value obtained by subtracting a greater one in a vertical size and a horizontal size of the character image extracted by the processor from a length of one side of the square.

7. The character recognition device according to claim 1 , wherein
the image comprises at least one part of a vehicle, and
the processor extracts, from the image, characters described on a number plate hung on the vehicle as the character image.

8. A character recognition method, executed by a computer, the computer storing a plurality of dictionary databases corresponding to each of a plurality of degradation levels, said plurality of dictionary databases being generated from images wherein an original character image has been degraded using a plurality of different degradation levels, the character recognition method comprising:
an image input step of receiving input of images;
an extraction step of extracting character image from said images;
a usage order determination step of determining an usage order for the plurality of dictionary databases on the basis of degradation degrees for extracted character images, wherein the usage order determination step comprises:
calculating, according to corresponding relationships between the degradation degrees of the character images and the dictionary databases, a degradation level corresponding to the degradation degree of the character images; and
determining the usage order as first using the dictionary database corresponding to the calculated degradation level and then beginning to sequentially use dictionary databases corresponding to degradation levels approximate to the calculated degradation level; and

a character recognition step of using dictionary databases in accordance with said usage order, recognizes characters included in the extracted character image and, if recognition results do not fulfil prescribed conditions, performing character recognition using the dictionary database next in said usage order, if recognition results fulfil the prescribed conditions, outputting the recognition results without performing recognition using the dictionary database next in said usage order.

9. A non-transitory computer-readable recording medium, storing a program causing a computer to perform the steps of:
receiving input of images;
extracting character image from said images;
determining an usage order for the plurality of dictionary databases on the basis of degradation degrees for extracted character images, comprising:
calculating, according to corresponding relationships between the degradation degrees of the character images and the dictionary databases, a degradation level corresponding to the degradation degree of the character images; and
determining the usage order as first using the dictionary database corresponding to the calculated degradation level and then beginning to sequentially use dictionary databases corresponding to degradation levels approximate to the calculated degradation level; and

using dictionary databases in accordance with said usage order, recognizes characters included in the extracted character image and, if recognition results do not fulfil prescribed conditions, performing character recognition using the dictionary database next in said usage order, if recognition results fulfil the prescribed conditions, outputting the recognition results without performing recognition using the dictionary database next in said usage order.

Assignees

• Omron Tateisi Electronics Co

Inventors

• Hamabashiri Hideto
• Aizawa Tomoyoshi
• Hyuga Tadashi

Classifications

• G06V30/242 Primary

  Division of the character sequences into groups prior to recognition; Selection of dictionaries · CPC title

• G06V30/10

  Character recognition · CPC title

• G06F16/50

  of still image data · CPC title

• G06F17/30244

  Physics · mapped topic

• G06K9/68 Primary

  Physics · mapped topic