Method, apparatus, and computer-readable medium for processing an image with horizontal and vertical text

US10423851B2 · US · B2

Patent metadata
FieldValue
Publication numberUS-10423851-B2
Application numberUS-201815908308-A
CountryUS
Kind codeB2
Filing dateFeb 28, 2018
Priority dateFeb 28, 2018
Publication dateSep 24, 2019
Grant dateSep 24, 2019

How to read this patent

A practical reading order for non-experts. Skip the full description unless you need deep technical detail.

  1. Title

    What the patent document calls the invention.

  2. Abstract

    A short plain-language summary of the technical disclosure.

  3. Assignees and inventors

    Who owns or filed the patent and who is credited as inventor.

  4. Key dates

    Filing, priority, publication, and grant dates set the timeline.

  5. First independent claim

    The legal scope of protection — read this for what is actually claimed.

  6. CPC / IPC classifications

    Technology tags used to group this patent with similar filings.

  7. Citations and related patents

    Prior art links and similar publications in this corpus.

Abstract

Official abstract text for this publication.

Speed and accuracy of character recognition can be improved by isolating text orientation during an early stage of processing an image containing a mixture of horizontal and vertical text. Vertical and horizontal line bounding boxes are defined from characters in the image. In a section of the image containing horizontal text, vertical line bounding boxes may tend to be larger and/or spaced close together due to misalignment of characters. For the same reason, horizontal line bounding boxes may tend to be larger and/or spaced closed together in a section of the image containing vertical text. Such variations in size and/or spacing may be used to identify a division between the horizontal and vertical text. A subsequent character recognition process may take advantage of a known division to conserve computing resources.

First claim

Opening claim text (preview).

What is claimed is: 1. A method for processing an image containing horizontal text arranged in a horizontal direction and vertical text arranged in a vertical direction, the method comprising: determining a horizontal score for a plurality of horizontal line bounding boxes, the horizontal score determined from one or both of size and spacing of the horizontal line bounding boxes, the plurality of horizontal line bounding boxes defined from character connected components in the horizontal text and the vertical text, each of the horizontal line bounding boxes occupying a position interval in the vertical direction; determining a vertical score for a plurality of vertical line bounding boxes, the vertical score determined from one or both of size and spacing of the vertical line bounding boxes, the plurality of vertical line bounding boxes defined from the character connected components in the horizontal text and the vertical text, each of the vertical line bounding boxes occupying a position interval in the horizontal direction; comparing the horizontal score and the vertical score; and determining from the comparison of the horizontal and vertical scores, either a position of a horizontal division between the horizontal and vertical texts or a position of a vertical division between the horizontal and vertical texts. 2. The method of claim 1 , wherein the determining of a horizontal score comprises for each of the position intervals of the horizontal line bounding boxes, determining whether the position interval is a positive position interval or a negative position interval, the position interval of the horizontal line bounding box determined to be a positive position interval on condition that the position interval satisfies one or both of a horizontal evaluating size range and a horizontal evaluating spacing minimum, determining an overall proportion of positive position intervals among the horizontal line bounding boxes, and deriving the horizontal score from the overall proportion of positive position intervals among the horizontal line bounding boxes; and the determining of a vertical score comprises for each of the position intervals of the vertical line bounding boxes, determining whether the position interval is a positive position interval or a negative position interval, the position interval of the vertical line bounding box determined to be a positive position interval on condition that the position interval satisfies one or both of a vertical evaluating size range and a vertical evaluating spacing minimum, determining an overall proportion of positive position intervals among the vertical line bounding boxes, and deriving the vertical score from the overall proportion of positive position intervals among the vertical line bounding boxes. 3. The method of claim 2 , further comprising: using a median size of the position intervals of the horizontal line bounding boxes to determine the horizontal evaluating size range; and using a median size of the position intervals of the vertical line bounding boxes to determine the vertical evaluating size range. 4. The method of claim 2 , wherein the determining of a position of a horizontal division between the horizontal and vertical texts comprises: identifying a discordant pair of the position intervals among the horizontal line bounding boxes, the discordant pair having a positive position interval and a negative position interval adjacent to the positive position interval, there being a gap interval between the positive position interval and the negative position interval, there being two interval regions each located on opposite sides of the gap interval; identifying a target interval region from among the two interval regions located on opposite sides of the interval gap, the target interval region having a lower number of positive position intervals than the interval region on the opposite side, determining a reference score being a proportion of positive position intervals among the horizontal line bounding boxes in the target interval region, determining a comparative score being a proportion of positive position intervals among the vertical line bounding boxes in the target interval region, and determining the position of the horizontal division as being within the gap interval when the comparative score is greater than the reference score; and the determining of a position of a vertical division between the horizontal and vertical texts comprises: identifying a discordant pair of the position intervals among the vertical line bounding boxes, each discordant pair having a positive position interval and a negative position interval adjacent to the positive position interval, there being a gap interval between the positive position interval and the negative position interval, there being two interval regions each located on opposite sides of the gap interval, identifying a target interval region from among the two interval regions located on opposite sides of the interval gap, the target interval region having a lower number of positive position intervals than the interval region on the opposite side, determining a reference score being a proportion of positive position intervals among the vertical line bounding boxes in the target interval region, determining a comparative score being a proportion of positive position intervals among the horizontal line bounding boxes in the target interval region, and determining the position of the vertical division as being within the gap interval when the comparative score is greater than the reference score. 5. The method of claim 1 , further comprising: before the determining of the horizontal score, performing a horizontal line finding process to define the plurality of horizontal line bounding boxes, the horizontal line finding process comprising, for each of the character connected components: determining whether a horizontal gap between the character connected component and a target horizontal line bounding box satisfies a horizontal gap maximum, the target horizontal line bounding box being one among the plurality of horizontal line bounding boxes, determining whether a projected vertical overlap of the character connected component and the target horizontal line bounding box satisfies a projected vertical overlap minimum, and redefining the target horizontal line bounding box to include the character connected component on condition that both the horizontal gap maximum and the projected vertical overlap minimum are satisfied; and before the determining of the vertical score, performing a vertical line finding process to define the plurality of vertical line bounding boxes, the vertical line finding process comprising, for each of the character connected components: determining whether a vertical gap between the character connected component and a target vertical line bounding box satisfies a vertical gap maximum, the target vertical line bounding box being one among the plurality of vertical line bounding boxes, determining whether a projected horizontal overlap of the character connected component and the target vertical line bounding box satisfies a projected horizontal overlap minimum, and redefining the target vertical line bounding box to include the character connected component on condition that both the vertical gap maximum and the projected horizontal overlap minimum are satisfied. 6. An apparatus for processing an image with horizontal text arranged in a horizontal direction and vertical text arranged in a vertical direction, the apparatus comprising: a processor; and a memory in communication with the processor, the memory storing instructions, wherein the processor is configured to perform a process on the image according to the stor

Assignees

Inventors

Classifications

  • Orientation detection or correction, e.g. rotation of multiples of 90 degrees · CPC title

  • Character recognition · CPC title

  • using recognition of characters or words · CPC title

  • Bounding box · CPC title

  • Drawing from basic elements · CPC title

Patent family

Related publications grouped by family.

External sources

Frequently asked questions

Answers are generated from the same data shown on this page.

What does patent US10423851B2 cover?
Speed and accuracy of character recognition can be improved by isolating text orientation during an early stage of processing an image containing a mixture of horizontal and vertical text. Vertical and horizontal line bounding boxes are defined from characters in the image. In a section of the image containing horizontal text, vertical line bounding boxes may tend to be larger and/or spaced clo…
Who is the assignee on this patent?
Konica Minolta Laboratory Usa Inc
What technology area does this patent fall under?
Primary CPC classification G06V30/1463. Mapped technology areas include Physics.
When was this patent published?
Publication date Tue Sep 24 2019 00:00:00 GMT+0000 (Coordinated Universal Time) (B2). Legal status and post-grant events are not shown on this page.
What related patents are in patentsdb?
We list 2 related publications on this page (citations in our corpus or others sharing the same primary CPC).