EP0050234A2 — Method and apparatus for character preprocessing

Claim 1

Independent Claim 138 words

Method for preprocessing video data in an optical character recognition system comprising raster scanning of a document, detection of video data in a video detector (2), transferring the detected data into a memory (13), recognizing characters by means of a control computer (14) and transferring recognized characters from the memory into a user unit (16), characterized by coding digital video bits from the video detector into a compressed run-length code in an encoder (3), transferring video data from the encoder into a memory area (22) for compressed video data, setting up a start-of-scan table (23) comprising address information for start-of-scan data in said area of compressed data (22), expanding compressed video data for one character from the area of compressed data into an expanded data memory area (25) and recognition of the character in the expanded data area.

Claim 2

Dependent Claim 138 words

Method for preprocessing video data in an optical character recognition system comprising raster scanning of a document, detection of video data in a video detector (2), transferring the detected data into a memory (13), recognizing characters by means of a control computer (14) and transferring recognized characters from the memory into a user unit (16), characterized by coding digital video bits from the video detector into a compressed run-length code in an encoder (3), transferring video data from the encoder into a memory area (22) for compressed video data, setting up a start-of-scan table (23) comprising address information for start-of-scan data in said area of compressed data (22), expanding compressed video data for one character from the area of compressed data into an expanded data memory area (25) and recognition of the character in the expanded data area.

Claim 3

Dependent Claim 70 words

Method according to claim 1, characterized in that the transfer of video data from the encoder (3) to the memory area (22) of memory (13) is controlled by a direct memory access unit (10) and that the set up of a start-of-scan table (23), the expansion and the recognition of characters are controlled by the control computer (14), the direct memory access unit (10) and the control computer working asynchronously.

Claim 4

Dependent Claim 103 words

Method according to claim 1, characterized in that the set up of the start-of-scan table (23) comprises the establishment of a frame of constant width and height that surrounds a character through introducing a frame marker (B) in the start-of-scan table, which marker defines the vertical position of the frame, and through introducing a horizontal displacement value in a register area (24), which value defines the horizontal position of the frame, whereby access is achieved to the compressed data within the defined frame through the start-of-scan table, whereupon this data is expanded and stored in the expansion memory area (25) for character recognition.

Claim 5

Dependent Claim 133 words

Method according to claims 1-3, characterized in that the control computer scans data in the compressed memory area (22) position after position and enters during this scan those position addresses that indicate the start of a scan to the start-of-scan table (23), enters flag bits (F) into the start-of-scan table, when all white scans are followed by scans of black and white information and vice versa, registers the maximum value (R2) and the minimum value (Rl) of black information into the register area (24), calculates the position of the frame marker (B) into the scan start table according to the flag bits (F) on the basis of a constant width of the frame, and calculates the horizontal displacement value (R3) from maximum/minimum values on the basis of a constant height of the frame.

Claim 6

Dependent Claim 66 words

Method according to anyone of the claims 1-4, characterized in that the position of the frame marker (B) in the start-of-scan table is shifted one or several steps and/or that the horizontal displacement value (R3) is increased or reduced one or several steps after an unsuccessful character recognition operation, whereafter another expansion operation and character recognition operation is performed by means of the new corrected frame.

Claim 7

Dependent Claim 27 words

Method according to claim 5, characterized in that the correction of the frame is based on statistical experience information and/or upon calculating the amount of black information.

Claim 8

Dependent Claim 43 words

Method according to claim 6, characterized in that the calculation of the amount of black information is performed by means of subtraction pair after pair of position data in the compressed memory area (22) and successive addition of the result of the subtraction.

Claim 9

Dependent Claim 176 words

Device for character recognition in an optical document reader, comprising a video detector (3) for digital detection of video data, a memory (13) for storing video data, a control unit (14) for recognition of characters from stored video data and an output channel (17, 18, 21) for transferring recognized characters from the memory to a user unit (16), characterized by an encoder (3) for coding detected video data from the video detector 2 into a compressed code, means (10) for transferring compressed video data from the encoder into a first memory area (22) for storing the video data in compressed form, a second memory area (23, 24) for storing table data defining a frame around a character sub-area in said first memory area, a third memory area (25) for storing video data in expanded form from said sub-area, whereby the control unit controls setting up table data in the second memory area (23, 24), expanding compressed video data from the sub-area and recognition of a character from expanded video data in the third memory area (25).

Claim 10

Dependent Claim 176 words

Device for character recognition in an optical document reader, comprising a video detector (3) for digital detection of video data, a memory (13) for storing video data, a control unit (14) for recognition of characters from stored video data and an output channel (17, 18, 21) for transferring recognized characters from the memory to a user unit (16), characterized by an encoder (3) for coding detected video data from the video detector 2 into a compressed code, means (10) for transferring compressed video data from the encoder into a first memory area (22) for storing the video data in compressed form, a second memory area (23, 24) for storing table data defining a frame around a character sub-area in said first memory area, a third memory area (25) for storing video data in expanded form from said sub-area, whereby the control unit controls setting up table data in the second memory area (23, 24), expanding compressed video data from the sub-area and recognition of a character from expanded video data in the third memory area (25).

Claim 11

Dependent Claim 153 words

Device according to claim 8, characterized in that the encoder (3) includes a one-bit delay circuit (5) the input of which is connected to the output (7) of the video detector, an EXCLUSIVE OR circuit (6), one input (8) of which is connected to the output of the delay circuit (5), the second input of which is connected to the output (7) of the video detector and the output (9) of which is connected to the data transferring means (10), and a counter (11) that is connected to a clock circuit (15) and that is stepped upwards synchronously with the detection by the video detector of the digital data bits and the output (12) of which is connected to the memory (13), the encoder (3) translating the video data from the video detector into a compressed run length code and transferring these compressed data to the first memory area (22) of memory (13).

Claim 12

Dependent Claim 65 words

Device according to claim 8, characterized in that the control unit (14) comprises a subcontrol unit (26) for general control, a second subcontrol unit (27) for preprocessing and framing compressed video data, a third subcontrol unit (42) for expanding and transferring video data from the first memory area (22) to the third memory area (25) and a fourth subcontrol unit (29) for correcting the preprocessing.

Claim 13

Dependent Claim 50 words

Device according to claim 8, characterized in that the second memory area (23, 24) comprises a start-of-scan table (23) including a frame marker (B) defining the vertical position of the frame, and a register area (24) for storing a horizontal displacement value (R3) defining the horizontal position of the frame.

Claim 14

Dependent Claim 90 words

Device according to claims 10 and 11, characterized in that the fourth subcontrol unit (29) for correcting the preprocessing comprises a first auxiliary control unit (38) for arithmetically calculating the amount of black video data within a chosen region of the first memory area (22), a second auxiliary control unit (39) for shifting the position of the frame marker (B) in the start-of-scan table (23) a third auxiliary control unit (40) for correcting the horizontal displacement value (R3) and the fourth auxiliary control unit (41) for considering statistical video data.

Claim 15

Dependent Claim 27 words

Device according to patent claims 8-12, characterized in that the control unit (14) is a micro computer and that the sub control units (26-29) comprise .micro programs.

Method and apparatus for character preprocessing

Abstract

Inventors

Bibliographic Details

Claims Analysis

References Cited (Backward Citations)

USPTO Annuity & Maintenance Windows

Assigned Categories & Tags

Assign Category Tag