Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Character recognition device and method, image display device, and image retrieval device

A character recognition and character technology, applied in character recognition, character and pattern recognition, electronic digital data processing, etc., can solve the problems of increasing influence, analyzing confusion, and reducing translation accuracy.

Active Publication Date: 2016-05-11
KK TOSHIBA +1
View PDF5 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] In the case of using text obtained as a result of character recognition in various text applications, loss of text due to inability to recognize such hidden characters can become a major problem
For example, in the case of translating the text of the recognition result, there is a concern that only a part of the character is missing, causing large confusion in the analysis and greatly reducing the translation accuracy
In addition, in the case of image retrieval based on the text of the recognition result, missing characters will reduce the reproducibility of the retrieval result
Especially in the case of word-based retrievers, this effect becomes large

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Character recognition device and method, image display device, and image retrieval device
  • Character recognition device and method, image display device, and image retrieval device
  • Character recognition device and method, image display device, and image retrieval device

Examples

Experimental program
Comparison scheme
Effect test

no. 1 approach >

[0063]The character recognition device according to the embodiment performs character recognition on a text region including characters (text) in an input image (hereinafter referred to as "input image"). At this time, character recognition is performed by supplementing the characters that actually exist but are hidden and do not appear on the input image, and a text of the recognition result is output. However, in the description of the present embodiment, a text area is assumed to represent an image area in which a single line of text appears, and when text spreads over multiple lines, each line is treated as an independent text area.

[0064] (Classification of masking phenomena)

[0065] Hereinafter, the phenomenon in which actual characters are hidden is referred to as a "masking phenomenon". figure 1 is a diagram illustrating the masking phenomenon. From the point of view of where characters in the text area are hidden, the masking phenomenon is divided into two types:...

no. 2 approach >

[0262] Next, a second embodiment will be described. The second embodiment is an embodiment of an image display device using the character recognition device 100 of the first embodiment.

[0263] Figure 28 It is a block diagram showing a configuration example of the image display device 200 of this embodiment. Such as Figure 28 As shown, the image display device 200 of this embodiment includes a character recognition device 100 , a display screen generator 210 , a reconstructor 220 , and a text region structure selector 230 .

[0264] The image display device 200 of this embodiment displays the text (hereinafter referred to as "recognized text") as the character recognition result Y1 so as to overlap the text region detected from the input image X by the character recognition device 100 . In this case, for example, the text in the visible text area is colored blue and the text in the hidden text area is displayed in a identifiable manner, such as by coloring the text in th...

no. 3 approach >

[0273] Next, a third embodiment will be described. The third embodiment is an embodiment of an image retrieval device using the character recognition device 100 of the first embodiment.

[0274] Figure 30 It is a block diagram showing a configuration example of the image search device 300 of this embodiment. Such as Figure 30 As shown, the image search device 300 of this embodiment includes the character recognition device 100 , an image DB (database) 310 , an index builder 320 , and a ranker 330 .

[0275] The image search device 300 of the present embodiment, upon receiving a text search query Q from the user, returns an image group R matching the search query Q from among image groups stored in advance in the image DB 310 . Among them, in the comparison between the search query Q and the image in the image DB 310 , the recognized text of the character recognition performed on the image by the character recognition device 100 is used. That is, the recognized text outpu...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention provides a character recognition device and method, an image display device, and an image retrieval device. According to an embodiment, the character recognition device includes a detector, first and second recognizers, an estimator, a second recognizer, and an output unit. The detector is configured to detect a visible text area including a visible character from an image. The first recognizer is configured to perform character pattern recognition on the visible text area, and calculate a recognition cost according to a likelihood of a character pattern. The estimator is configured to estimate a partially-hidden text area into which a hidden text area estimated to have a hidden character and the visible text area are integrated. The second recognizer is configured to calculate an integrated cost into which the calculated cost and a linguistic cost corresponding to a linguistic likelihood of a text that fits in the entire partially-hidden text area are integrated. The output unit is configured to output a text selected or ranked based on the integrated cost.

Description

[0001] This application enjoys the benefit of priority of Japanese Patent Application No. 2014-223173 filed on October 31, 2014, and uses the entire content thereof. technical field [0002] Embodiments relate to a character recognition device, an image display device, an image retrieval device, and a character recognition method. Background technique [0003] Conventionally, there is known a technique for recognizing characters from a scene image input from, for example, a camera attached to a smartphone. In a scene captured as a scene image, there are cases where characters that actually exist but are hidden and cannot be seen are included. For example, characters on a signboard behind it may not be visible due to obstructions such as utility poles, or characters in a specific range may not be visible due to overexposure, out-of-focus, or the like. However, in the conventional character recognition technology, only visible characters are recognized, and therefore characte...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/20G06V30/10G06V30/246
CPCG06V10/225G06V30/10G06V30/153G06V30/246G06V30/274G06F40/30G06F40/205G06F40/279
Inventor 浜田伸一郎长田邦男
Owner KK TOSHIBA
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products