EP0287027A2
Expired / Lapsed
EP
⏱️ 0.0 years remaining
Method and apparatus for recognising characters printed on a document
Assignee: International Business Machines Corp
•
Filing: 1988-04-12
•
Granted: —
•
Expires: 2008-04-12
Abstract
Abstract A document is scanned by an image scanner and image data representing the image of the document is stored in an image storage means. Rectangles contacting and surrounding outer boundaries of each image of plural character rows in the image storage means are generated, and the positions of four edges of the rectangle in XY coordinates of the image storage means are detected. A size of a rectangle is calculated based upon the detected positions of the edges, and the size of each rectangle is compared with a expected size range for the characters and symbols to be recognized. The positions of the rectangles falling into the size range are stored in a first table as the position data wherein the position data of the rectangles of the characters and symbols over the plural character rows are arranged in the first list in the order from a rectangle at one end to a rectangle at the other end along the direction of the X axis of the XY coordinates. The position data of the rectangles in the first list are sequentially fetched in the arranged order to detect a size of each rectangles to determine an average size of all rectangles stored in the first list. Again, the position data of the rectangles in the first list are sequentially fetched in the arranged order to find out a first rectangle falling into a size range settled based upon the average size. The fetch operations are continued to find out a second rectangle having a bottom left corner located within predetermined distances in the X and Y directions from a bottom left corner of the first rectangle. The fetch operations are continued to find out a third rectangle having a bottom left corner located within the predetermined distances in the X and Y directions from the bottom left corner of the second rectangle. The operations continues to find out a predetermined number of rectangles in one character row satisfying the condition. When the predetermined number of the rectangles have been found, a skew of the character row in the XY coordinates is calculated based upon the positions of the bottom left corners of these rectangles, and this detected skew is treated as the skew of the document.
Inventors
- 👤 Takashi Mano
Bibliographic Details
Application Number
—
Jurisdiction
EP
Data Source
google_patents
Last Synchronized
2026-10-29 19:32
CPC Classifications
G; G06; G06V; G06V30/00; G06V30/10; G06V30/14; G06V30/146; G06V30/1475; G06V30/1478
Claims Analysis
Total Claims: 13
Claim 1
Independent Claim
155 words
A method for recognizing characters printed on a document comprising steps of: (a) storing an image of said document in image storage means; (b) detecting position of rectangles in XY coordinates of said image storage means, which contact outer boundaries of character images, respectively in plural character rows stored in said image storage means, to generate position data representing said positions of said rectangles; (c) forming a list in which said position data of the rectangles are arranged in the order from one end to the other end along the direction of X axis of said XY coordinates; (d) detecting a range in Y axis for one character row based upon Y axis information of said position data of said rectangles in said list; and (e) fetching said character images stored in said image storage means by using said position data of said rectangles which fall into said range, to recognize said fetched character images.
Claim 2
Independent Claim
155 words
A method for recognizing characters printed on a document comprising steps of: (a) storing an image of said document in image storage means; (b) detecting position of rectangles in XY coordinates of said image storage means, which contact outer boundaries of character images, respectively in plural character rows stored in said image storage means, to generate position data representing said positions of said rectangles; (c) forming a list in which said position data of the rectangles are arranged in the order from one end to the other end along the direction of X axis of said XY coordinates; (d) detecting a range in Y axis for one character row based upon Y axis information of said position data of said rectangles in said list; and (e) fetching said character images stored in said image storage means by using said position data of said rectangles which fall into said range, to recognize said fetched character images.
Claim 3
Dependent Claim
214 words
Method for recognizing characters in a document comprising steps of: (a) storing image data representing image of said document in image storage means; (b) detecting positions of rectangles in the XY coordinates of said image storage means, which contact outer boundaries of each character image in plural character rows stored in said image storage means to generate position data representing said positions of said rectangles; (c) storing said position data of said rectangles to form a first list in which plural position data of said rectangles are arranged in the order from one end to the other end in a direction of X axis of said XY coordinates; (d) determining the position data of a predetermined number of rectangles belonging to one character row to detect a skew of said document in said XY coordinates; (e) generating virtual position of each rectangle by correcting the position in Y axis of said XY coordinates of each rectangle in said first list by said skew to transfer said position data of said rectangles belonging to the first character row from said first list to a second list; and (f) supplying from said image storage means to a character recognition means said image data surrounded by said rectangles specified by said position data in said second list.
Claim 4
Dependent Claim
214 words
Method for recognizing characters in a document comprising steps of: (a) storing image data representing image of said document in image storage means; (b) detecting positions of rectangles in the XY coordinates of said image storage means, which contact outer boundaries of each character image in plural character rows stored in said image storage means to generate position data representing said positions of said rectangles; (c) storing said position data of said rectangles to form a first list in which plural position data of said rectangles are arranged in the order from one end to the other end in a direction of X axis of said XY coordinates; (d) determining the position data of a predetermined number of rectangles belonging to one character row to detect a skew of said document in said XY coordinates; (e) generating virtual position of each rectangle by correcting the position in Y axis of said XY coordinates of each rectangle in said first list by said skew to transfer said position data of said rectangles belonging to the first character row from said first list to a second list; and (f) supplying from said image storage means to a character recognition means said image data surrounded by said rectangles specified by said position data in said second list.
Claim 5
Dependent Claim
77 words
A method for recognizing characters printed on a document according to Claim 2, wherein said step (b) of detecting positions of rectangles includes a step (g) of generating rectangles each of which contacts outer boundaries of discrete images of said document stored in said image storage means, and a step (h) of determining as to whether each rectangle has a size which falls into a predetermined size range settled for said characters and symbols to be recognized.
Claim 6
Dependent Claim
34 words
A method for recognizing characters printed on a document according to Claim 2, wherein said position data represents the positions of top, bottom, left and right edges of said rectangle in said XY coordinates.
Claim 7
Dependent Claim
91 words
A method for recognizing characters printed on a document according to Claim 2, wherein said step (d) of determining the position data includes a step (i) of searching said first list to detect said predetermined number of rectangles, with a bottom left corner of one rectangle being located within a predetermined distance in said X and Y axes from a bottom left corner of the preceding rectangle, and a step of performing calculations of the positions of bottom left corners of said detected rectangles to detect said skew of said document.
Claim 8
Dependent Claim
175 words
A method for recognizing characters pointed on a document according to Claim 2, wherein said step (e) of generating virtual position includes a step (j) of sequentially reading said position data of said rectangles in said first list in said arranged order, a step (k) of generating virtual position of each rectangle by correcting the position in said Y axis of each rectangle in said first list by said skew, a step (l) of storing in a register the virtual position of said rectangle located at the highest position on said document, a step (m) of sequentially reading again said position data of said rectangles in said first list, a step (n) of generating said virtual position of each rectangle, a (o) step of determining as to whether said generated virtual position falls into a predetermined range from said virtual position stored in said register, a step (p) of transferring said position data of said rectangle in said first list, said generated virtual position of which falls into said predetermined range, to said second list.
Claim 9
Dependent Claim
123 words
Character recognition apparatus for recognising characters printed on a document comprising: image storage means for storing image of a document; means for detecting positions of rectangles each of which contacts outer boundaries of each of plural character images stored in said image storage means to generate position data representing said position; means for arranging said rectangles in the order from one end to the other end in a direction of X axis based upon X axis information of said position data; means for selecting rectangles belonging to a character row based upon Y axis information of said ordered rectangles; and means for fetching character images from said image storage means by using said position data of said rectangles to recognize said character images.
Claim 10
Dependent Claim
123 words
Character recognition apparatus for recognising characters printed on a document comprising: image storage means for storing image of a document; means for detecting positions of rectangles each of which contacts outer boundaries of each of plural character images stored in said image storage means to generate position data representing said position; means for arranging said rectangles in the order from one end to the other end in a direction of X axis based upon X axis information of said position data; means for selecting rectangles belonging to a character row based upon Y axis information of said ordered rectangles; and means for fetching character images from said image storage means by using said position data of said rectangles to recognize said character images.
Claim 11
Dependent Claim
51 words
Character recognition apparatus according to Claim 7, wherein said means for detecting positions of rectangles determines as to whether a size of each rectangles falls into a scope of sizes of said characters and symbols to be recognized, and generates said position data of rectangles falling into said scope of sizes.
Claim 12
Dependent Claim
29 words
Character recognition apparatus according to Claim 7, wherein said position data represent upper, bottom, left and right edges of said rectangles in XY coordinates of said image storage means.
Claim 13
Dependent Claim
247 words
Character recognition apparatus according to Claim 7, wherein said means for arranging said rectangles including a first table memory storing a first list which stores said position data of said rectangles of plural character rows of said document in the order from one end to the other end in said direction of said X axis of said XY coordinates; and wherein said means for selecting rectangles sequentially reads said position data of said rectangles of said first list in said arranged order, detects a predetermined number of rectangles, in which a bottom left corner of one rectangle is located within predetermined distances in said X and Y directions from a bottom left corner of the preceding rectangle, detects a skew of said document based upon the positions of bottom left corners of said detected rectangles, sequentially reads again said position data of said rectangles of said first list in said arranged order to generate a virtual position of each rectangle which is a Y axis position corrected by said skew, storing a virtual position located at a highest position of said document into a register, sequentially reads again said position data of said first list in said arranged order to generate said virtual positions of said rectangles, determines as to whether said generated virtual positions falls into a predetermined range from said virtual position in said register, and transfers said position data of rectangle having said virtual position falling into said predetermined range to a second list.
References Cited (Backward Citations)
No citation records indexed.
USPTO Annuity & Maintenance Windows
Maintenance schedule is calculated for granted US utility patents.