JPS6262391B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a device that automatically reads characters in documents such as arbitrary forms and books.

Conventional devices of this kind have a readable document with a reference mark printed in advance, or a specified frame with characters written correctly in it, etc. However, there was a drawback that it was impossible to read characters from arbitrary documents such as forms and books.

In order to eliminate the above-mentioned conventional drawbacks, the present invention captures a document such as a form or book as a binary image of white pixels and black pixels, scans the entire screen, and when a white pixel is detected, surrounding pixels are scanned. The number of black pixels contained in the image is determined, and processing is performed to change the pixels to black pixels according to the number, creating a patterned blurred image of the character string.The blurred image is scanned to recognize the position, length, etc. of the character string. However, each character is read according to the recognition result, and its purpose is to provide a device that can automatically read characters from any document accurately and in a short time. The invention will now be explained in detail with reference to the drawings.

FIG. 1 is a schematic configuration diagram showing an embodiment of an automatic character reading device of the present invention, in which 1 is a document, 2 is an imaging section, 3 is a blurred image creation section, 4 is a character string recognition section,
Reference numeral 9 denotes a character reading means consisting of a reading head drive section 5, a reading head 6, a character cutting section 7, and a character recognition section 8. The operation of this device is summarized as follows. That is, when a command to capture an image is issued from the blurred image creation section 3 to the imaging section 2 equipped with a TV camera, the imaging section 2 focuses the document 1 on the document 1 with white pixels. The image is captured as a binary image of black pixels, and the entire screen of the image is scanned by the blur image creation unit 3. When a white pixel is detected, the number of black pixels included in the surrounding pixels is checked and the number is calculated. Accordingly, the character string is changed to black pixels to form a patterned blurred image, and the character string recognition unit 4 recognizes the position, length, etc. of the character string, and the reading head drive unit 5 operates based on this recognition result. As a result of the operation, the reading head 6 scans along the character strings of the document 1 and character information for each row is extracted.The character information is cut out character by character in the character cutting section 7 and character recognized in the character recognition section 8. It is becoming like being encoded as a code.

Next, the principle of creating the blurred image will be explained using FIG. 2. Figure 2 shows an image and how its density changes due to focus position control, and shows the density distribution of a portion of a character captured when the TV camera of the imaging unit 2 is focused on the document 1. 21 changes as shown in density distribution 22 when the focus is shifted from the document to create a blurred image. Here, if the density for converting into a white/black binary image is set to the threshold value 23, the pattern 25 of the blurred image corresponding to the density distribution 22
is wider than the original character pattern 24. Incidentally, an image due to a shift in focus of a point light source on the x-y plane becomes a circle of confusion with uniform brightness and a radius p on the imaging plane ξ-η. In other words, the spread function h(ξ, η) is (However, a is a constant). Let the intensity distribution be g(x, y),
If g′(ξ, η), g′(ξ, η)=〓 ^∞ _−∞ g(x, y) h(ξ−x, η−y) dxdy …(2) holds. In order to calculate the concentration of point (ξ, η) from equations (1) and (2), we need to calculate the integrand of equation (2) only for points within a circle with radius p centered at point (ξ, η). All you have to do is calculate. Utilizing this property, a pixel range corresponding to the circle of confusion is considered and blurring processing is performed. An example of a pixel range corresponding to a circle of confusion is shown in FIG. 3a for p=1 pixel and in FIG. 3b for p=2 pixels. In FIG. 3a or b, when the central pixel A is black, it is left as is, and when it is white, the number of black pixels in pixels B ₁ to B ₈ (or B ₁ to B ₂₀ ) surrounding pixel A is checked. Here, a threshold N for the number of black pixels corresponding to the threshold 23 is set according to the value of p, and if the number of black pixels in pixels B ₁ to B ₈ (or B ₁ to B ₂₀ ) is N or more, pixel A is Change to black. A blurred image is created by performing this blurring process over the entire screen of the binary image.

FIG. 4 is a block diagram of the blurred image creation unit 3, and 30 in the figure is an image memory that stores the image of the document 1 sent from the imaging unit 2, with white pixels corresponding to “0” and black pixels corresponding to “1”. , 31 is an address register for scanning the image memory 30, 32 is instruction information of a pixel range, and 33 is an address register for scanning the image memory 30 according to the instruction information 32.
34 is a comparator that compares the pixel read from the image memory 30 with "1"; 35 is a counter that counts the number of black pixels detected by the comparator 34; 36 is a counter 35; threshold value N determined by the numerical value of and the instruction information 32
37 is a memory that stores "1", 38 is a buffer memory that temporarily stores "1" information, and 39 is a read end signal for pixels within the pixel range. The pixel range instruction information 32 is in document 1
The value of p is determined based on the document size, character size, feedback information of the recognition result of the character recognition unit, etc., and is given manually or by a control circuit (not shown).

The image memory 30 is scanned by an address register 31 to scan the pixels of the entire screen. “1” (black pixel) is stored as is, but when “0” (white pixel) is detected, scanning of the address register 31 is stopped, and according to the instruction information 32, the pixel with “0” as the center pixel is The pixels in the range are read out by the address register 33 and compared with "1" in the comparator 34 one after another.
The comparator 34 increments a counter 35 each time the read pixel is "1", and the contents of the counter 35 are transferred to the comparator 36 when the read end signal 39 is issued.
is compared with the threshold value N, and if it is greater than or equal to the threshold value N, the memory 37 is driven and "1" is stored in the buffer memory 38, and if it is smaller than the threshold value N, the address is directly changed.
The register 31 is restarted to perform the next scan. Information in buffer memory 38 is transferred into image memory 30 by ORing with the contents of image memory 30, as appropriate. Such processing is performed over the entire screen, and the character strings of the document in the image memory 30 are patterned and converted into a blurred image, which is sent to the character string recognition section 4.

FIG. 5 is a block diagram of the character string recognition unit 4.
4 is an image memory that stores the blurred image sent from the blurred image creation unit 3; 42 is a character area extraction circuit that extracts only the character area excluding "appendices, drawings", etc. from the blurred image; 43 is an image memory 4
1, a tilt correction circuit that detects the tilt of the image and corrects it by rotating the TV camera, etc., 44 is an m, σ calculation circuit that calculates the average value m of the width of the blur pattern of the character string and its standard deviation σ; A character string line extraction circuit extracts a character string line by averaging the midpoints of the blur pattern of a character string, and 46 is a character string end detection circuit that detects the upper, lower, left, and right ends of the blur pattern of a character string. Character area extraction circuit 42 and tilt correction circuit 43
is unnecessary when the object to be read is a document consisting only of characters, and when the document is not tilted due to a document guide or the like.

FIG. 6 shows an example of a blurred image stored in the image memory 41, in which 101 is a character string blurring pattern, 102 is the width of the blurring pattern 101, 103 is the midpoint of the width 102, and 104 is the blurring pattern. 101 is the break, 105 is the line break of the blur pattern 101, 106 is the blur pattern 101
107 is the unevenness of the edge of the blur pattern 101, 108 is noise patterned from dirt on the paper surface, etc., and 109, 110, 111, 112 are characters obtained by averaging the coordinates of the midpoint 103 of the blur pattern 101. Column lines 113 and 114 are arrows 11 representing x-coordinates parallel to the character strings and y-coordinates perpendicular to them.
5 is a scanning line in the y-coordinate direction.

Next, the operation of character string recognition will be described. first,
The m,σ calculation circuit 44 scans the entire screen in the direction of the scanning line 115, and calculates the width 102 of the blur pattern 101.
The average value m and its standard deviation σ are calculated and sent to the character string line extraction circuit 45. Next, the character string line extraction circuit 45 performs scanning with the y direction as the main scan for each blur pattern, determines the coordinates of the midpoint of the blur pattern according to the flowchart shown in FIG. Determine the coordinates of the column lines. now,
The coordinate l(k-1) of the character string line 109 has already been obtained, and the coordinate l(k) of the character string line 110 is newly calculated.
When trying to find the , the scan start command is used to scan in the y direction with the y coordinate of the character string line 109 as the starting point, and the middle point C(n) of the blur pattern that first intersects (n is the scan in the x direction). ) and the width d are calculated. In judgment 1, a judgment is made to remove the unevenness 107 and noise 108 at the edge based on the widths d and m, the average value m and standard deviation σ obtained by the σ calculation circuit 44, and if yes, the C(n) and d are After erasing, scanning in the y direction is continued, and if NO, proceed to judgment 2. In judgment 2, it is judged whether or not the cut 104 has been passed using the coordinates C(n-1) of the midpoint obtained during the previous scan, and if yes, the coordinates C(n-1) of the midpoint are used.
(n) and d are deleted to start the next scan, and if NO, proceed to decision 3. In judgment 3, it is judged whether or not the line feed part 105 or the constriction 106 has been passed, and if NO, the character string line 11 required by C(n) is
It is determined that it belongs to the blur pattern of 0, this is stored, and the next scan is started, and if Yes, proceed to determination 4. Judgment 4 uses the y-coordinate l(k-1) of the previous character string line 109 to determine whether it is the line break 105 or the constriction 106 that has passed, and if Yes, it is determined that it is the constriction 106. Then that C
(n), erase d and continue scanning in the y direction,
If NO, it is determined that the line feed section 105 has been passed and C
It is determined that (n) is the first midpoint belonging to the desired character string line 110, the coordinates of the previous midpoint are deleted, and the next scan is performed. When scanning in the y direction is completed to the right end of the screen, the midpoint coordinate C(n) is determined to belong to the blur pattern of the character string line 110.
is averaged to calculate the y-coordinate l(k) of the character string line 110. Similarly, the coordinates below the character string line 111 can be determined.

As shown in FIG.
Scanning 116, 117 in the vertical direction based on
is performed from the left end to the right end of the character string line 110 to detect the maximum value in the vertical direction of the blur pattern, and the upper end 11
8. Detect the coordinates of the bottom edge 119 and select the character string line 110
Scanning 120,1 in the horizontal direction with a predetermined width centered on
21 is performed to detect the maximum value in the horizontal direction of the blur pattern, and the coordinates of the left end 122 and right end 123 are detected.

The recognition results of the position, length, and width of the character string obtained as described above are sent to the reading head drive unit 5, and the reading head 6 is driven as described above. will be done. In addition,
As for the image pickup section 2, the reading head drive section 5, the reading head 6, the character cutting section 7, and the character recognition section 8 that constitute the character reading means 9, those used in conventional character reading devices can be used as they are. , not specifically mentioned. Furthermore, instead of the reading head driving section 5 and the reading head 6, the TV camera of the imaging section 2 may be configured to be capable of scanning and zooming, and this may be controlled to read each character of the character string.

As explained above, according to the present invention, there is an imaging unit that captures a document such as a form or a book as a binary image of white pixels and black pixels, and an imaging unit that scans the entire screen of the binary image and detects white pixels. a blurring image creation unit that converts the character string into a patterned blurred image by checking the number of black pixels included in the surrounding pixels and converting the white pixel into a black pixel according to the number; Scan the patterned blurred image to determine the position of the character string,
It consists of a character string recognition unit that recognizes the length, etc., and a character reading means that reads each character in the character string according to the recognition results such as the position and length of the character string, so it can read the image of the document as it is. By simply importing a binary image, scanning it, and converting white pixels to black pixels based on the number of black pixels in the surrounding pixels, it is possible to create a blurred image of a character string with a pattern that removes the high frequency components of the original character. Therefore, the processing time can be shortened, character strings of all formats and formats can be correctly recognized, and characters can be read accurately. Furthermore, the cost of the device can be reduced because parts of a conventional character reading device can be used as they are as a document imaging unit and means for reading individual characters. Furthermore, since the black pixels in the binary image of the document are preserved as they are, there is no loss of low-density characters, which occurs when blurring is performed by controlling the focus of the camera in the imaging unit, and there is no change in the field of view. No correction circuit is required, simplifying the device. In addition, conventionally, the contents of documents with arbitrary formats and formats were manually input into various processing devices using keyboards, etc., but with this device, it is possible to read them as is, so the input processing time is reduced. It has the advantage that it can be used as an automatic input device when creating a database from a large amount of existing documents and as an input/output device for a terminal in data communication.

[Brief explanation of the drawing]

The drawings show embodiments of the present invention, and FIG. 1 is a schematic diagram of the automatic character reading device of the present invention, FIG. 2 is an explanatory diagram of an image and its density change by controlling the focal position, and FIG. a, b are explanatory diagrams of pixel ranges corresponding to circles of confusion of p=1 pixel and p=2 pixels, respectively, FIG. 4 is a block diagram of the blurred image creation section, FIG. 5 is a block diagram of the character string recognition section, FIG. 6 is an explanatory diagram showing an example of a blurred image, FIG. 7 is a flowchart of a character string line extraction circuit, and FIG. 8 is an explanatory diagram of detection of the top, bottom, left and right ends of a blur pattern. DESCRIPTION OF SYMBOLS 1...Document, 2...Imaging unit, 3...Blurred image creation unit, 4...Character string recognition unit, 9...Reading head drive unit 5, reading head 6, character cutting unit 7, character recognition unit 8 A means of reading characters.

Claims (1)

[Claims] 1. An imaging unit that captures documents such as forms and books as a binary image of white pixels and black pixels; A blur image creation unit that converts a character string into a patterned blur image by checking the number of pixels and changing the white pixels to black pixels according to the number; It is characterized by comprising a character string recognition unit that recognizes the position, length, etc. of a string, and a character reading means that reads each character in the character string according to the recognition results such as the position, length, etc. of the character string. Automatic character reading device.