JPS63106882A - Character font deciding system for printed character recognizing device - Google Patents

Abstract

PURPOSE:To increase the processing speed of character recognition and to improve a recognition rate, by finding the mean value of size differences added at every font from the size difference between a reference character pattern and an input character pattern for each font consisting of a font class, and deciding the character font which supplies a minimum mean value as the font of a read character. CONSTITUTION:When no character font is decided from a histogram data at every font unit obtained in prescribed number of character recognition to which all of the dictionary fonts are used, the histogram of the font class consisting of plural similar fonts is generated, and the font class in which the read character is included is decided. Furthermore, the mean value of the size difference for a decided font class is calculated from the size difference added at every font between the reference character pattern and the input character pattern obtained at time of the character recognition, and the character font which supplies the minimum mean value is decided as the character font of the read character. In such way, it is possible to decide an accurate character font even when a large number of similar fonts are included, and to improve the processing speed and the recognition rate.

Description

Detailed Description of the Invention (Field of Industrial Application) The present invention is a method for recognizing read characters from a plurality of types of one-note dictionaries by identifying the font (characters) A1 to A1 of printed characters. The present invention relates to a character font determination method for a printed character reading device that determines a font dictionary to be used.

(Prior art) Conventionally, a character reading device that optically reads and recognizes printed characters printed on paper (in OCR, multiple types of characters to be read) is provided with a standard character pattern for each character. A dictionary is stored in the memory of each individual phono 1, and character fonts are automatically detected from actually read printed characters to determine the font dictionary to be used for subsequent character recognition.

This automatic character phono 1~ determination process is an important function for shortening recognition time and obtaining accurate recognition results.

Here, to specifically explain the conventional automatic character phono 1~ determination process, first, the characters in the first few dozen lines of the print sheet that are optically read by the reading head are recognized using all font dictionaries. to create a histrum of the font dictionary used in this character 62 recognition. Next, from the histogram created, if the frequency of use is the highest and there is a frequency difference of more than a predetermined value from the second highest frequency, the phono 1~ with the highest frequency of use is read and determined to be the font on the print sheet. A font dictionary corresponding to this determined font is used to perform subsequent character recognition. This automatic font determination method is effective when each of the fonts A1 to A1 are different from each other.

(Problem to be solved by the invention) However, in recent years, the number of types of ants that can be read has increased to hundreds of types, and some of them have the same basic form but only partial parts. There are many fonts with high similarity that are different from each other, and when printing sheets 1 to 1 using any one of these fonts are to be read,
In the conventional method of calculating a histogram for each font and determining the most frequently used font under predetermined conditions, errors in font selection are likely to occur, making it difficult to accurately determine the character phono 1. was difficult.

Furthermore, regarding recognition methods, in order to cope with the increase and diversification of character fonts, it has become common to perform recognition processing after normalizing the character size. For example, Fourier 10
Conventional automatic font processing has little effect on similar fonts with mostly similar fonts, such as ``Fourier 12'' and ``Fourier 12''.

(Means for Solving the Problems) The present invention was made in view of such conventional problems, and even if a large number of similar fonts are included, it is possible to accurately recognize characters based on the results of a predetermined number of character recognition. It is an object of the present invention to provide a character phono 1 determination method for a printed character reading device that can accurately determine the font of printed characters to be read.

In order to achieve this object, the present invention provides a character font determination method for a printed character recognition device that reads and recognizes a group of characters printed on paper using one of multiple types of character fonts having different fonts. It is characterized by performing the following processing.

(A> Equipped with multiple font dictionaries that store standard character patterns for character recognition for each of multiple types of character fonts, and sequentially recognizes a predetermined number of characters, for example, several dozen characters, read from paper using all the font dictionaries. A histogram of the font dictionary is created by adding 1 to the histogram corresponding to the font dictionary used for character-by-character recognition.

('B> After creating a histogram, check whether a similar character font exists in the recognized character font, and if a similar font exists, use the corresponding font calculated by calculating the size difference between the reference character pattern in the font dictionary and the input character pattern. Add the size difference for each.

(C) Sort the histograms created in (A) above to a predetermined rank n in ascending order.

(D) The first histogram (maximum usage frequency) obtained by sorting is greater than or equal to the first predetermined value α, and the second histogram is less than or equal to the second predetermined value β, which is smaller than the first predetermined value α. At this time, the font of the read character is determined to be the character font corresponding to the first histogram.

(E) If the font determination condition of (D> above is not satisfied), the predetermined rank m sorted according to (C) above (however,
The histograms included up to m < n are targeted,
A histogram for each font class consisting of similar fonts in this histogram group is calculated and sorted in descending order of size.

(F) The histogram of the first font class in the sorting of font classes is the third predetermined value 1 or more, and the second
When the histogram of the first rank is less than or equal to the fourth predetermined value δ, it is determined that the read character belongs to the first rank font class. still,
If the font class determination conditions are not satisfied, all information is reset and the process starts over.

(G) Calculate the average value of the added size differences obtained in (B) above for each font belonging to the determined font class.

(H) Determine the character font giving the minimum average value as the font of the read characters.

(Operation) According to the character font determination method of the present invention, it is possible not only to determine character fonts with different fonts, but also to determine similar fonts even if they have the same basic font but are partially different. A character font that determines the included font class, calculates the average value of the size difference added for each font from the size difference between the reference character pattern and the input character pattern for each font that makes up the font class, and gives the minimum average value. To improve the processing speed and recognition rate of character recognition by accurately determining the character font of read printed characters from among multiple types of character fonts including many similar fonts. Can be done.

(Embodiment) FIG. 2 is a block diagram schematically showing a printed character recognition device equipped with the character font determination method of the present invention.

In FIG. 2, reference numeral 1 denotes a printing paper, on which text and the like are printed in type using a predetermined character font. A reading head 2 reads printed characters by optically scanning the printing paper 1 and outputs an analog video signal (reading signal). The read signal from the read head 2 is an A/D
The converter 3 converts the signal into a digital signal having a predetermined number of bits, and the signal is written and stored in the memory 4 as a two-dimensional bit pattern. 5
is a character cutting section, which extracts, for example, one line of bit pattern data from the two-dimensional bit pattern of read characters stored in the memory 4, and extracts bit pattern data for each character based on a preset printing character pitch. 1 - Cut out the pattern data. Reference numeral 6 denotes a recognition section, which compares the character pip I pattern cut out one character at a time by the cutout section 5 with a reference character pattern in a dictionary memory prepared in advance and outputs a recognition result. Reference numeral 7 denotes a dictionary memory group, which stores font dictionaries prepared in advance for each of a plurality of types of character fonts having different fonts, and corresponds to character fonts determined based on the character phono 1 to determination method of the present invention. The character recognition unit 6 performs character recognition using the font dictionary.

Furthermore, 8 is a font determining unit equipped with the character font determining method of the present invention, which stores a predetermined number of characters read from the printing paper 1, for example, several dozen characters, in the dictionary memory group 7, as will be explained later. Character recognition is performed using all the font dictionaries that have been created, and the character font of the printed characters on the printing paper 1 is determined from the result of this character recognition, and the recognition unit 6 specifies the Ant dictionary to be used for character recognition in the dictionary memory group 7.

FIG. 1 is a block diagram showing an embodiment of the character font determination process of the present invention. First, the configuration of the font determination process in FIG. 2 will be explained. 10 is a histogram creation section; 6, font determination unit dictionary memory group 7
Dictionary phones 1 to 1 used to obtain recognition results in the recognition process using all font dictionaries provided in
The font information representing the font is input for each character recognition, and the recognition result is added to the histogram corresponding to the font dictionary with 1q.
Performs histogram creation processing that adds .

Here, the character font to be processed by the character font determination method of the present invention has, for example, the contents as shown in FIG. 2a.

First, in dictionary group 7 in FIG. 2, font unit A1.
A2. A3,81. ...N1. A font dictionary is stored in N2゜N3, and among these font units, font units with the same code of A1-A3, C1-C4,...N1-N3 have a basic font in common and have partial Similar font A1 with different fonts constitutes a similar font A1.
~A3 is set as font class A, and similar fonts C1~C
4 as font class C, and further similar fonts N1 to N
3 is defined as font class N. In addition, font unit B
Since font 1 has no similar font, font class B is the font unit B1 itself.

Referring again to FIG. 1, since the processing target is a character font composed of font units and font classes as shown in FIG. A histogram memory 11 is provided to store histograms for each font unit. The histogram memory 11 stores histogram data 1'a1 to Hn3 using the font units A1 to N3 as addresses, and for example, when the recognition unit 6 obtains a recognition result from the font dictionary of the font unit A1 by recognizing one character, , Based on the font information A1 at this time, the histogram creation unit 10 specifies address A1 of the histogram memory 11, reads out histogram data 1-(al, and adds 1 to this histogram data) lal, and writes 1 to address A1 again. The histogram data Hal obtained by adding .

Creation of histogram data for each font by the histogram creation unit 10 is performed for all of the dozens of characters used in character font determination processing, and when the recognition processing for dozens of characters is completed, the histogram memory 11 stores the font units A1 to N3. Histogram data 1”a1
~Hn3 is now stored.

Reference numeral 12 denotes a character size difference calculation unit, which inputs font information obtained by recognizing one character in the recognition unit 6 in FIG. 2, a reference character pattern of a dictionary font used for character recognition, and an input character pattern to be recognized. A size difference ΔD between the reference character pattern and the input character pattern is calculated. The character font to be calculated by the character size difference calculation unit 12 is a character font for which there is a similar font to the recognized character font among the character fonts composed of font units and font classes shown in FIG. 2a. .

That is, for the font configuration shown in FIG. 2a, for example, when a recognition result is obtained using the font dictionary for the font unit A1, similar fonts A2. Since A3 exists, the character size difference calculation unit 12 calculates the size difference between the reference character pattern and the input character pattern. On the other hand, regarding the recognition results using the dictionary font of font B1,
Even if there is no similar font, the character size difference calculation unit 12 performs the calculation processing in the same way. However, if the recognition result is that the font cannot be distinguished by character size, the size difference calculation is not performed.

Size difference memory 1 corresponds to character size difference calculation unit 12.
3 is provided. Size difference memory 13 is similar font A
1-A3. Character size difference calculation unit 1 uses each font unit of B'l, C1 to C4, ..., N1 to N3 as an address.
The addition data of the size difference ΔD calculated in step 2 is stored.

To be more specific, for example, if a recognition result using dictionary font A1 is obtained, the character size difference calculation unit 12
The size difference Δ[)al between the reference character pattern obtained from the recognition result of dictionary font A1 and the input character pattern at that time is
is calculated, and the added size difference data ΔDal is read out by addressing the size difference memory 13 based on the font information A1 given at this time, and the calculated size difference is added.
The addition result is written to address A1 again.

As a result, when the recognition process for several dozen characters that is the target of the character font determination process is completed, the size difference memory 13 stores the size difference addition data Δ[)a for each similar font A1-N3.
The state is such that l~Δ[)n3 is stored.

Next, the histogram memory 1 is
The apparatus configuration for font determination processing based on the histogram data for each font unit obtained in 1 will be described.

Reference numeral 14 denotes a first sorting unit, which performs a sorting process of arranging the histogram data l-1al to Hn3 of the font units A1 to N3 stored in the histogram memory 11 in ascending order up to a predetermined rank n. Reference numeral 15 denotes a first and second histogram detection unit, which detects the first and second histogram data from the sorted data arranged in ascending order up to a predetermined rank n in the first sorting unit 14. and is applied to the comparing section 16 as a first character font determining means. Comparison part 1
6 is a first predetermined value α and a second predetermined value α smaller than the first predetermined value α.
A predetermined value β is set as a comparison standard value, and the first place,
It is compared and determined whether the first histogram obtained by the second histogram detection unit 15 is greater than or equal to the reference value α and the second histogram is smaller than β, and if this condition is satisfied, the read character is A first character font determination output Fl (H level output) is generated, which is a character font corresponding to the histogram of 1st position.

On the other hand, if the comparison unit 16 does not satisfy the conditions that the first histogram is greater than or equal to the reference value α and the second histogram is smaller than β, the first character font determination output F1
is not output and the output of the comparison section 16 is at L level at this time, so the L level output of the comparison section 16 is inverted by the inverter 17 to instruct the second sorting section 18 to process.

The second sorting unit 18 targets the histogram data included in the predetermined ranks from 1 to n sorted by the first sorting unit 14, and calculates a histogram for each font class consisting of similar fonts in this histogram data group. A histogram of font classes that are sequentially arranged up to a predetermined rank m is created and the sorting process is performed. 19 is the first
1st, 2nd histogram detection unit, and 2nd sorting unit 1
The first and second font class histograms are detected from among the histograms of the font classes up to the predetermined rank m obtained in step 8, and outputted to the comparing section 20.

A third predetermined value γ and a fourth predetermined value δ smaller than γ are set in the comparison unit 20 as comparison reference values, and if the first font class histogram is the predetermined value 1 or more and the second
When the font class histogram of
Font class determination output Fc assuming that the read character exists in the font class corresponding to the first histogram
is output, and the processing operation of the average value calculation unit 22 is performed.

On the other hand, if the comparison unit 20 does not satisfy the conditions that the first font class histogram is greater than or equal to the predetermined value 1 and the second font class histogram is less than or equal to the predetermined value δ, the histogram memory 11 and the size A reset signal is output to the difference memory 13, data stored in each memory is reset, and the process of creating a histogram and size difference for dozens of read characters is not necessary.

Upon receiving the font class determination output FC from the comparison unit 20, the average value calculation unit 22 calculates the determined font class stored in the size difference memory 13, for example, when the font class A is determined, the similar font class included in the font class A is determined. Addition size difference data Δ[)a1 to Δ[ of phono 1-Al to A3
) a3 and add size difference data Δ1) al-・Δ
Since the number of additions for each of [)a3 is known in advance, the minimum value detection unit 24 calculates the average size difference Δ[)a1 to ΔDa3 for each similar font A1 to A3.
Output to. The minimum value detection unit 24 detects the minimum value from among the average size difference data of similar fonts obtained from the average value calculation unit 22, and selects a second font whose read characters are the character font corresponding to the minimum average value. A decision output F2 is output. That is, the minimum detecting section 24 has a function as a second font determining means that determines any one of the similar fonts in the font class determined by the comparing section 20 as the character font of the read character.

Next, the character font determining means of the present invention will be explained in detail with reference to the flowchart of FIG.

In FIG. 3, when the character font determination process is started,
First, the first character out of several dozen characters read to distinguish between fonts is read into the recognition section, and in block 30, recognition is performed on a character-by-character basis. When the determination blower 32 determines that the recognition result for each character has been obtained, the block 34
Dictionary phono 1 used for character recognition in block 30
Add 1 to the histogram of ~. If the character cannot be recognized, the process returns from the determination block 32 to block 30, reads the next character, and performs the recognition process.

When the histogram 1 of the dictionary font used for character recognition is added in block 34, the next judgment block 36 checks whether or not the recognition result allows font distinction based on character size for the recognized characters, and if possible, proceeds to block 38. Size difference Δ between standard character pattern and input character pattern
Calculate [) and add to each font. Of course, if the font cannot be distinguished by character size, the process of block 38 is not performed.

When the process of block 38 is completed, a check is made in determination block 40 to see if the recognition process for the designated number of characters to be determined as a font has been completed, and if the recognition process for the designated number of characters has not been completed, the process returns to block 30 and the next One character is read and a histogram based on the recognition result and size difference are added in units of fonts.

When the recognition process for the specified number of characters is completed through the processing from block 30 to determination block 40, for example, for a character font having the font unit and font class configuration shown in FIG. 2a, the character font as shown in FIG. A histogram for each font is obtained.

Next, in block 42, the histograms as shown in FIG. 4 obtained for each font are sorted in descending order of size up to the nth position, to obtain the sorting results as shown in FIG.

Regarding the sorting results shown in FIG. 5, in the next determination block 44, the histogram data l-1c3 of the font C3, which ranks first, is equal to or greater than the predetermined value α, and the histogram data 1-( of the font C1, which ranks second) When cl is smaller than the predetermined value β, the process proceeds to block 46, where it is determined that the font C3 that provides the first histogram data l-1c1 is the character font of the read character, and the font determination process is terminated.

On the other hand, if a histogram for each font as shown in FIG.
The sorting result in is as shown in FIG. 7, for example. In this case, the first histogram data is greater than or equal to the predetermined value α, but the second histogram data is greater than the predetermined value β, so the condition of determination block 44 is not satisfied and the process proceeds to block 48.

In the process of block 48, a histogram of the font classes to which the nth fonts sorted in block 42 belong is taken, and the fonts are sorted in ascending order up to the nth font.

That is, regarding the sorting result in FIG. 7, the histogram data of similar fonts C3, C2, C4, and C1 that make up font class C1 are added to obtain histogram data of font class C, and this is used as the histogram data of font class C. The same procedure is performed for font classes, and as a result, a histogram of font classes sorted in ascending order up to the nth rank as shown in FIG. 8 is obtained.

Next, in the determination block 50, the histogram of the font class C, which is the first in the sorting result of the histogram data of the font classes obtained in the block 48, is a predetermined value of 1 or more, and the font class A is the second font class.
If the histogram is less than or equal to the predetermined value δ, the process proceeds to block 52 and the font class C corresponding to the first histogram is determined as the font class of the read character.

On the other hand, as shown in FIG. 9, for example, as shown in FIG. 9, the additional size difference data obtained by adding the size difference between the reference character pattern and the input character pattern for each font has been obtained in advance through the processing in block 38. Similar fonts C1 to C belonging to the determined font class C
Calculate the average value of the 4 added size difference data, for example, the first
Obtain the average value of the size difference shown in Figure 0.

Next, in block 54, the font C2 with the minimum average value is detected from among the size difference average values of the similar fonts C1 to C4 included in the determined font class C, and the font C2 is the character font C2 whose read characters give the minimum average value. It is determined that the process is completed.

On the other hand, if the histogram of the font class shown as an example in FIG. 8 does not satisfy the conditions of the discrimination block 50 for the first and second histograms, the process proceeds to block 56 and proceeds to block 34. The obtained histogram data in font units and the size difference data in font units obtained in block 38 are reset, and the process returns to block 30 again to determine a similar character phono 1 by reading a new specified number of characters. This is repeated until the character phono 1~determined output of block 46 or 54 is obtained.

Of course, once the character phono 1~ of the read character is determined from the character phono 1~ determination process according to the present invention, the process proceeds to the original character recognition process using only the phono 1~ dictionary corresponding to the determined character font. become.

(Effects of the Invention) As explained above, according to the present invention, when character phone 1~ cannot be determined from histogram data for each font unit obtained by recognizing a predetermined number of characters using all dictionary fonts, To do this, create a histogram of font classes consisting of multiple similar fonts, determine the font class that includes the read character, and then calculate the reference character pattern and input character pattern obtained during character recognition for the determined font class. The minimum average value is given by calculating the size difference average value from the tightened size difference for each 4 nt of the size difference of Even if the type of character font contains many similar characters, it is possible to determine the correct character font 1 and greatly improve the processing speed and recognition rate of character recognition.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an embodiment of the present invention, FIG. 2 is a block diagram showing an outline of a printed character reading device, and FIG.
The figure is an explanatory diagram showing the structure of font units and font classes to be processed by the present invention, Figure 3 is a flowchart showing character phono 1 to determination processing of the present invention, and Figure 4 is a flowchart of the present invention. An explanatory diagram showing an example of hiss 1 hegelum in units of phono 1 obtained by character recognition, Fig. 5 is an explanatory diagram showing the results of the histogram of Fig. 4, and Fig. 6 is an illustration of character recognition according to the present invention. Figure 7 is an explanatory diagram showing another example of the histogram for phono 1~ units obtained in Figure 7 is an explanatory diagram showing the sorting results of Figure 6, Figure 8 is obtained from the sorting results of Figure 7. An explanatory diagram showing a histogram of font classes, FIG. 9 is an explanatory diagram showing an example of size difference addition data obtained by character recognition of the present invention in units of fonts, and FIG. FIG. 9 is an explanatory diagram showing the average value of the size difference and latitude data in FIG. 9; 1: Print paper 2: Reading head 3: A/D converter 4: Memory 5: Character cutout section 6: Recognition section 7: Dictionary memory group 8: Font determination section 10: Histogram creation section 11: Histogram memory 12: Character size Difference calculation unit 13: Size difference memory 14: First sorting unit 15: 1st and 2nd position detection unit 16: Comparison unit (first font determination unit) 17.21: Inverter 18: Second sorting unit 19: First 2nd place detection unit 20: Comparison unit ()A-nt class determination unit) 22: Average value calculation unit

Claims (1)

[Scope of Claims] In a printed character recognition device that reads and recognizes a group of characters printed on paper using any of a plurality of types of character fonts having different fonts, A font in which a plurality of font dictionaries storing character patterns serving as recognition standards and a predetermined number of read characters read from the paper are sequentially recognized using all of the font dictionaries to obtain recognition results for each character. a histogram creation means that adds 1 to a histogram corresponding to a dictionary, and after creating the histogram, if there is a similar character font in the recognized character font, calculates the size difference between the reference character pattern of the font dictionary and the input character pattern, and calculates the size difference for each font. size difference calculation means for adding a size difference to the histograms; first sorting means for arranging the histograms obtained by the histogram creation means in descending order of size up to a predetermined rank; When the histogram is greater than or equal to a first predetermined value and the second histogram is less than or equal to a second predetermined value that is smaller than the first predetermined value, the font of the read character is changed to the character font corresponding to the first histogram. a first character font determining means to determine; and when a determination condition in the first character font determining means is not satisfied, target histograms included up to a predetermined rank sorted by the first sorting means; a second sorting means for calculating histograms for each font class made up of similar fonts in the histogram group and arranging them in ascending order up to a predetermined rank; and a histogram of the first-ranked font class obtained by the second sorting means. is greater than or equal to a third predetermined value and the second histogram is less than or equal to a fourth predetermined value smaller than the third predetermined value, the read character is determined to belong to the font class corresponding to the first histogram. and a font class determining means for calculating the average value of the added size differences obtained by the size difference calculating means for each font belonging to the font class determined by the font class determining means, and determining the font of the read character to the minimum average value. A character font determining method for a printed character reading device, comprising: a second font determining means for determining a character font corresponding to the character font.