JPH03214350A - 'bushu' (common part which becomes mark for looking for character in chinese character dictionary, such as left-hand radical or the like) retrieving method - Google Patents

Abstract

PURPOSE:To specify a desired 'BUSHU' (a common part which becomes a mark for looking for a character in a Chinese character dictionary, such as a left-hand radical, etc.) even in the case the name of the 'BUSHU' and the number of strokes of the 'BUSHU' are not known exactly by selecting a position of the 'BUSHU' of its KANJI (Chinese character) from in a table of the position of the 'BUSHU' and outputting one group of 'BUSHU' corresponding to its position. CONSTITUTION:A circumscribed frame circumscribed to a KANJI is derived, and a 'BUSHU' of its KANJI is classified by positions 41A - 41K of the 'BUSHU' of its KANJI. Subsequently, when a prescribed position, for instance, the position 41C is selected from in a table of the positions 41A - 41K of the 'BUSHU' to the circumscribed frame, one group of 'BUSHU' corresponding to its prescribed position is outputted. In such a way, even in the case the name of the 'BUSHU' and the number of strokes of the 'BUSHU' are not known, the 'BUSHU' of a desired KANJI is specified.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a radical search method for Chinese characters suitable for use, for example, in the correction stage of recognition results of a character recognition device, the input stage of a word processor, and the like.

[Summary of the invention]

The present invention provides a Kanji radical search method that is suitable for use, for example, in the stage of correcting the recognition results of a character recognition device and the input stage of a word processor. The radicals of a kanji are classified according to the position of the radical, and when a predetermined position is selected from the list of radical positions relative to the circumscribed frame, a set of radicals corresponding to that predetermined position is classified. By outputting the neck, a desired radical can be specified even if the name of the radical and the number of strokes of the radical are unknown.

[Prior Art] It is necessary to specify a predetermined Chinese character when correcting errors in recognition results in a character recognition device, or when inputting characters from a manuscript in a word processor. In this case, if the sound or precept of the kanji is known, the kanji can be searched relatively easily, but if the pronunciation or precept is unknown, the search is performed using the total number of strokes or the radical of the kanji.

Among these, the method of searching by the total number of strokes requires time to count the total number of strokes if the kanji in question is complex, and there is a risk of specifying the total number of strokes incorrectly.
An effective method is to search by radical. Conventional radical input methods include a method of directly inputting a radical name such as ninben, gandarare, etc., and a method of inputting the number of strokes of the radical.

[Problem to be solved by the invention]

However, the method of directly inputting a radical name cannot be used if the radical name is not known, so it is necessary to use a list of about 200 radicals. There is an inconvenience that requires Further, the method of inputting the number of radical strokes has the disadvantage that it takes time due to an error in counting the number of strokes.

In view of the above, an object of the present invention is to propose a radical search method that can identify a desired radical even if the name of the radical and the number of strokes of the radical are not accurately known.

[Failure to solve the problem]

The radical search method according to the present invention obtains a circumscribing frame that circumscribes a kanji, and the position of the radical of the kanji with respect to this circumscribing frame (41
A) The radical of the kanji is classified according to ~(41K), and a predetermined position (for example, position (41C)) is selected from the list of radical positions (41A) ~ (41K) with respect to the circumscribed frame. Sometimes a string of radicals (
For example, the image shown in FIG. 2) is output.

[Operation] According to the present invention, in order to specify the radical of a certain kanji, the position of the radical of that kanji is selected from the list of positions (414) to (41M) of that kanji. By doing so, a set of radicals corresponding to that position is output, so it is possible to specify the radical of the Kanji character from among the set of radicals.

Therefore, even if the name of the radical and the number of strokes of the radical are not known, the radical of a desired Kanji character can be specified.

〔Example〕

Hereinafter, one embodiment of the present invention will be described with reference to the drawings. In this example, the present invention is applied to the stage of correcting recognition results of a character recognition device.

FIG. 9 shows the system configuration of the character recognition device of this example. In FIG. 9, (13) is a scanner (14) consisting of a document feeder and an image reader, and the scanner (14) is a document sent to this scanner. l3
) is the entire page of the manuscript (l4), for example 400X40
The dots are separated at a reading density of 0 dpi (dots per inch) to generate an original character signal S1 corresponding to the shade of each dot.

(15) is an image data input/output board, (16) is a host computer, (21) is a printer, and the image data input/output board (15) supplies a predetermined portion of the original character signal Sl to the host computer (16). At the same time, a printing signal output from the host computer (16) is supplied to the printer (I6). (17) is a keyboard for operating the host computer (16);
(18) indicates a coordinate input unit for inputting various coordinates to the host computer (16), and (19) indicates a character identification board, in which the host computer (16) extracts the circumscribed frame of one character from the original character signal S1. The basic rectangular cut-out character signal S7 cut out using the cut-out signal that becomes high level "1" internally is sequentially passed to the character identification board (19).
), the character identification board (19) supplies the character code C of the character corresponding to the cutout character signal S7 (or the jump code if the character cannot be recognized) to the host computer (16).

(20) indicates a display device made of a cathode ray tube, and a predetermined area on the display screen of this display device (20) displays the results of recognizing characters for one page of the manuscript (14) in a format corresponding to the manuscript (14). Do as shown in . In addition, this display device (2
The display screen 0) is designed to be able to display the donot pattern itself for one page or a predetermined portion of the original (14) as needed.

FIG. 10 shows a more detailed configuration of the printer (21) in the example in FIG. 9 excluding the processing part. In the image data input/output board (15) in FIG. An image data input section (23) having a memory capable of storing dosoto patterns for more than one page;
is also a hack-up memory that can store dot patterns for one or more pages of the original (14), and inputs the original character signal s1 for one page of the original (14) output from the scanner (13) to the image data input section ( 22). Further, a desired portion of the original character signal s1 stored in the image data input section (22) is transferred to the hunkup memory (23) at any time, and the desired portion corresponds to a blank state with no characters ( For example, the original character signal S1 with the desired part replaced is called the original character signal s2, and a predetermined part of this original character signal S2 is replaced with the original character signal s2. It is called S3.

In the host computer (16), (24) is a central processing unit (hereinafter referred to as rcpuJ), (25) is a main memory, and (26) is a video signal RAM (hereinafter referred to as rVRAM) for the display device (2o). , (27
) indicates a font table consisting of a character ROM that inputs a character code and outputs a dot pattern, that is, a font, of a predetermined font corresponding to this character code.
) When various commands, data, and coordinate data are supplied to the CPU (24), the CPU (24) controls the overall operation of the character recognition device of this example in response to these.

Further, the original character signals S2 and S3 are supplied to the main memory (25) and VRAM (26), respectively, as needed. in this case,
The basic rectangular cutout character signal S7, which is a character signal corresponding to the inside of the circumscribed frame of one character read from the main memory (25), is sequentially supplied to the character identification board (19), and the character identification board (19) transmits the character signal S7. The incoming character code C is supplied to the address lot of the font table (27) via the main memory (25), and the font data appearing in the data lot of this font table (27) is sent to a predetermined area of the VRAM (26). Do it as if you were writing it. In addition, the host/computer (16), keyboard (17), coordinate unit (1B), and display device (2) of this example are also included.
The system consisting of 0) also has the function of a word processor.

In the character identification board (19), (28) indicates a character recognition section, (30) indicates a recognition dictionary section that stores font data of various fonts in correspondence with character codes (in this example, JIS codes), The recognition unit (28) and recognition dictionary unit (30) basically perform character identification work. The recognition dictionary section (30) of this example is divided into a large classification dictionary section for large classification characters and a subclassification dictionary section for small classification characters. Generally, the general characteristics of a character are also expressed by the dot patterns near each side of the circumscribing frame of the character, so the dot patterns near the four sides of the circumscribing frame of each character are digitized as four-side data (or peripheral data). The font data of characters (major classification characters) whose four side data fall within a predetermined range may be stored in the major classification dictionary section.

On the other hand, the subdivision dictionary section stores 24 x 24 F y normalized font data of all other characters (subdivision characters) that are not included in the major dictionary section, corresponding to the character codes. There is.

(29) indicates a dictionary creation section, and this dictionary creation section (29)
When the dictionary creation mode is set by the operator, determines whether the font data represented by the supplied basic rectangular cutout character signal for one character corresponds to a major classification character or a subclassification character. When corresponding to major classification characters, the font data is normalized and written into a predetermined character code area of the major classification dictionary of the recognition dictionary section (30), and when corresponding to subclassification characters, the font data is normalized and recognized. It is written in a predetermined character code area of the subdivision dictionary section of the dictionary section (30). This allows the user to easily create a recognition dictionary section (30) that can handle various fonts.

When the character recognition mode is set by the operator, the recognition section (28) of the character identification board (19) changes the size of the recognition dictionary section (30) when the supplied basic rectangular cutout character signal S7 corresponds to a major classification character. The font data of the classified character section is sequentially processed in first-in, first-out (FIFO) format.
When the font data corresponds to subcategory characters, the font data of the subclass character section is sequentially written to the second FIFO register. In addition, in parallel with this operation, the recognition unit (28)
normalizes the dot pattern corresponding to the basic rectangular cutout character signal S7 and sequentially writes it into the third FIFO register. Then, the recognition unit (28) uses the dot pattern of the character to be recognized in the third FIFO register and the first FIFO register.
A series of font data in the FO register and the second FI
By sequentially comparing a series of font data in the FO register, 10 character codes corresponding to the font data closest to the don't pattern of the character to be recognized are generated in descending order of priority, and these character codes are It is written in a predetermined area of the main memory of the host computer (16).

To determine the priority, for example, for each dot of 24 x 24 dots, the dot pattern of the character to be recognized is compared with the font data read from the recognition dictionary section (30), and if the values of the two are different, The total sum of the donots that are listed is taken as an evaluation value, and the order of priority is set in descending order of the evaluation value. If the evaluation value of the character code with the highest priority is less than or equal to a predetermined value, it is assumed that character recognition has been performed and the character code with the highest priority is changed to the character code C of the character to be recognized. The data is written in a predetermined area of the main memory (25) corresponding to the original (14). At the same time, the recognition unit (28) stores the character code C, data indicating the size of the character, and data indicating the position of the character in the average circumscribing frame in the main memory (
25). On the other hand, if the evaluation value of the character code with the highest priority exceeds the predetermined value, the recognition unit (
28) assumes that character recognition was not possible, and saves the reject code to the manuscript (I4) in its main memory (25).
The data is written in a predetermined area corresponding to the data. The character recognition operation described above is executed at high speed in a pipeline system.

Figure 1 shows the display screen (2OA) of the display device (20) of this example.
This display screen (20A) is provided with a recognition result display area (31) for displaying the recognition results of one page of characters of the original (14). In this example, the original (14) is written vertically, but the recognition results are displayed horizontally. (
33) indicates a cursor for indicating a character to be corrected, and this cursor (33) can be moved over any character in the recognition result using the keyboard (17) or the coordinate input unit (18).

In the example in Figure 1, the cursor (33) is on the blank (32A) indicating the rejected character, so the display screen (
The area (20A) close to the recognition result display area (31)
34) and (35), the word "target character" and a blank are displayed, respectively. (36) indicates the function button area. In this area (36), characters such as "candidate", "code", "kana-kanji", and "number of radical strokes" are displayed. (18). For example, when "Candidate" is selected, 10 character patterns recognized as being closest to the character to be corrected are displayed on a part of the display screen (20A) in descending order of priority, and "Code" is displayed. When selected, the character to be corrected can be specified directly with the character code, and when "Kana-Kanji" is selected, the character to be corrected can be input by Kana/Kanji conversion, and the character with "Radical Stroke Count" is selected. When the radical stroke number button (36E) indicated by is selected, the function button area (36E) is selected.
In the selection area (40) at the bottom of 6), a list of radical positions of Kanji characters is displayed as described later.

Further, (37) indicates a re-recognition button, and by selecting this button (37) with the coordinate input unit (18), it becomes possible to execute re-recognition of the character to be corrected.

In this example, a peripheral image display area (38) with an area of WxW is provided near the recognition result display area (31), and in this peripheral image display area (38) there is a document (14) corresponding to the character to be corrected. A dot pattern in an area of width 170 dots x height 170 dots centered on the upper character is displayed as is. Specifically, in FIG. 10, C P
U (24) is 170 x 170 dots centered around the character to be corrected in the original character signal S2 for one page (original character signal Sl itself in this example) stored in the image data input section (22). Original character signal S3 in the area of
is read out, and the original character signal S3 is enlarged (interpolated) or reduced (thinned out) and written in a predetermined area of VR A M (26). Therefore, the display screen (20A) in Figure 1
In the peripheral image display area (38) inside, images of the character "fL" on the document (14) corresponding to the blank to be corrected and images of, for example, eight characters around these seven characters are displayed as dot patterns. Further, in the cutout character pattern display area (39), a pattern normalized to 24 x 24 dots of the image of the character to be corrected on the original (14) is displayed.

When the image of the character to be corrected and the surrounding characters is displayed in this way, the operator can correct the character to be corrected according to the characters before and after it without referring to the manuscript (14). , there is a benefit of improving the efficiency of correction.

To explain the operation when correcting recognition results with the character recognition device of this example, the characters to be corrected are blanks (32A) to (32) in the recognition result display area (31) in FIG.
C) i.e. blanks among characters that are unrecognizable and excluded (3
2A).

In this case, move the cursor (33) to its blank (32A)
By moving above and operating the coordinate person Kasuinochi, the character corresponding to that blank (32A) is designated as a correction target, and the blank is displayed in the peripheral image display area (38) and cutout character pattern display area (39). (32A)
The information of the characters on the original document (14) corresponding to is displayed.

In this example in Figure 1, the character that could not be recognized is ``Muki'', and the radical of this kanji ``Muki'' is Amikashira ``口'', so we need to identify this radical ``口'' first. . Therefore, the operator must press the radical stroke number button (
By selecting 36E), a list of 11 positions (41A) to (41K) of radicals of kanji is displayed in the selection area (40).

For the positions (41A) to (41K) of the radicals of the kanji in this example, set a quadrilateral as a circumscribing frame that circumscribes the kanji, and which side of the radical of the kanji is set relative to the four sides of this quadrilateral. The position of the radical is classified according to whether it exists in the vicinity of the radical. Specifically, the left side, right side, and top side of the quadrilateral as its circumscribing frame. Bottom side, left and right sides, top side, left side and top side. Left side and bottom side, left and right sides and bottom side 5 Position the radicals that exist near the top side and right side, top and bottom sides and left side respectively (41A)
- (41J) are classified as belonging to the radicals, and those for which such classification is difficult are classified as the radicals belonging to the other positions (41K). Note that if the position of a radical is confusing, it may be classified into two or more positions (for example, both positions (41F) and (41J)).

In this example, the radical ``mouth'' of the kanji ``过'' is in position IF (4
1C), the operator must use the coordinate input unit (
18) to move the cross cursor (42) to the position (41).
Move onto the figure showing C) and operate the coordinate input switch. This selects that position (41G),
The selection area (40) has a position (41C) as shown in Figure 2.
Radicals belonging to (Nabebuta, Wakanmuri, <Sakanmuri 1st class) are displayed in order of the number of strokes in the radical. For example, position (41
The number of radical strokes in C) is classified into 2 strokes, 3 strokes, ..., 6 strokes, and 7 or more strokes.

Furthermore, radicals with the same number of radical strokes are arranged from left to right in order of frequency of use.

For example, in the third stroke, the most frequently used flower crown is placed on the far left.

Then, since [OIIj, which is sought in this example, is displayed within the frame (43) of the 5-stroke column in Figure 2, the operator moves the cursor (42) into the frame (43) and sets the Identify the "mouth" by manipulating the human body inch. Accordingly, 12 kanji belonging to the radical ``口'' are displayed in the selection area (40), as shown in Fig. 3, and the kanji character ``ku'', which is the target of this example, is displayed in the frame (44). Displayed. Then, the operator moves the cursor (42) to that frame (
44), and by operating the coordinate input switch, the kanji character ``补'' is specified as an unrecognized character. Specifically, the kanji character ``buk'' is displayed instead of the blank (32A) in FIG. 1, and the kanji ``buk'' is recorded in the corresponding file in the main memory (25).

As mentioned above, according to this example, when performing correction processing after character recognition, even if the name and stroke count of the radical of the kanji to be input are not known, the radical can be identified simply by specifying the position of the radical. This has the advantage that kanji can be input quickly and correction work can be done quickly. Furthermore, even if the name or number of strokes of the radical of the kanji is known, the positions of the radical (41,A) to (41k) are known, as in this example.
) can identify the target radical more quickly by selecting the position of the target radical from the list.

Finally, radicals belonging to positions (41A) to (41J) in FIG. 1 are collectively shown in FIGS. 4 to 8. That is, Figure 4 shows the radicals belonging to position (41A)
, Figure 5 shows the radical belonging to position (41B).
, Sanzu (Richi, etc.), Figure 6 shows the radicals belonging to position (41C), and Figure 7 shows the radicals belonging to position (410) (Renga..., Shitagokoro, =..., etc.). .

In addition, Fig. 8 A-F shows husband h positions (41E) to (41J).
Indicates radicals belonging to .

The present invention is not limited to the above-mentioned embodiments, and may be applied within the scope of the present invention, such as classifying radicals using a polygon other than a quadrilateral, a circumference, etc. as a circumscribing frame circumscribing each kanji. Of course, various configurations can be adopted.

[Effects of the Invention] According to the present invention, even if the name and number of strokes of a radical of a kanji character are unknown, there is an advantage that the radical can be searched.

[Brief explanation of drawings]

FIG. 1 is a front view showing the display screen of an embodiment of the present invention, FIGS. 2 and 3 are front views showing the display contents of the selection area in FIG. 1, and FIGS. 4 to 8 are respectively A line diagram showing examples of radicals belonging to 11 types of positions according to an embodiment, FIG. 9 is a configuration diagram including a partial perspective view showing the system configuration of a character recognition device according to an embodiment of the present invention, and FIG. 10 9 is a configuration diagram showing a more detailed configuration of the main part of the example in FIG. 9. FIG. (31) is the recognition result display area, (33) is the cursor that indicates the character to be corrected, (36E) is the radical stroke number button, (4
0) is the selection area, and (41A) to (41K) are the positions of the radicals, respectively.

Claims (1)

[Claims] Obtain a circumscribing frame that circumscribes a kanji, classify the radical of the kanji according to the position of the radical of the kanji with respect to the circumscribing frame, and select a circumscribed frame from a list of radical positions with respect to the circumscribing frame. A radical search method characterized in that when a position is selected, a set of radicals corresponding to the predetermined position is output.