JPS6242289A - Kanji retrieving and input system - Google Patents

Abstract

PURPOSE:To easily enable even an operator having no good knowledge of KANJI (Chinese characters) to perform a KANJI input processing, by inputting a part of KANJI together with the matching data supplied for the remaining part of KANJI when the KANJI supplied through writing is obtained from matching with a standard pattern. CONSTITUTION:A part of KANJI is inputted with divided writing on a tablet by depressing an input key. Then a keyboard control part 6 is switched to a data selection part 11 so as to input the hitherto data to a stroke data storing part. A user depresses 'contour' key on a keyboard 2 to input the remaining part of the KANJI in a symbol. The part 6 fetches the data obtained after depressing the 'contour' key and normalizes the KANJI stroke data through a normalizing part 14. Then the user depresses 'stroke' key after the input of the total number of strokes with the operation of digit keys. Thus, a dictionary selection part 7 selects a dictionary with the prescribed number of strokes. A partial collation part 15 performs the partial collation with the output of the part 14. When the collation is through with all KANJI patterns of the dictionary, the candidate KANJI are arrayed for display by a candidate selection part 10 in the order of KANJI having higher resemblance to the input pattern.

Description

[Detailed Description of the Invention] [Summary] When finding a kanji that should be detected by matching it with a standard pattern for a kanji that has been input by hand, a part of the kanji is input, and the remaining part is input without any special input, such as the total number of strokes. This is a search kanji input method that searches for a desired kanji by inputting matching data.

[Industrial Application Field] The present invention relates to a kanji input method in which kanji input by hand is matched with a standard pattern and input as a desired kanji, for example when using a kanji dictionary.

[Prior Art] Conventionally, an input kanji is recognized as a target kanji, and when used for the next processing, the kanji itself is input by hand, and character recognition technology is used to determine the character. FIG. 4 is a block diagram showing such a conventional technique.

In FIG. 4, 1 is a tablet with a writing input device, 2 is a keyboard, 3 is a cathode ray tube display device, 4 is a signal input unit, and 5 is a writing input device.
is a signal output section, 6 is a control section, 7 is a dictionary selection section, 81 to 8
n indicates a dictionary in which Kanji patterns are stored for each total number of strokes, 9 indicates a collation section, and 10 indicates a candidate selection section. By writing down kanji on the tablet 1, the data is input to the collation section 9 via the signal input section 4. On the other hand, the control section 6 judges the data, learns the predetermined total number of strokes, selects a predetermined dictionary from the dictionaries 81 to 8n in the dictionary selection section 7, and inputs the data to the collation section 9. The most consistent kanji candidates are displayed on the display device 3 via the candidate selection section 10 and the signal output section 5.

[The problem that the invention aims to solve] For the purpose of investigating kanji whose reading, meaning, usage, etc. are unknown,
When writing the kanji on the tablet 1 in FIG. 4, it may be difficult to write it accurately, such as when the target is an unknown kanji. Particularly for people who are not yet sufficiently familiar with kanji, such as children and foreigners, it may be virtually impossible to input them.

SUMMARY OF THE INVENTION An object of the present invention is to improve the above-mentioned drawbacks and to provide a method for inputting ``Ne'' kanji, which allows a desired kanji to be found even if only a portion of the kanji has been input.

[Means for Solving the Problems] FIG. 1 is a block diagram showing the configuration of the present invention. In FIG. 1, 1 is a tablet left writing input device, 2 is a keyboard, 3 is a cathode ray tube display device, 4 is a signal input section, 5 is a signal output section, 6 is a control section, 7 is a dictionary selection section, 81 to 8n 10 is a dictionary in which standard patterns of kanji for each total number of strokes are input, 1o is a candidate selection section, 11 is a data selection section, 12 is a contour information extraction section, 13 is a stroke data storage section, 14 is a normalization section, 15 16 indicates a partial matching section, and 16 indicates a search section.

[When a person draws a part of the kanji written from the action co-tablet 1, it is stored in the stroke data storage section 13 as a part of the pattern of the kanji, and the next input data (compared with the original data) is The normalization unit 14 operates so that the kanji can be specified by the kanji (meaning ``labor'', etc.). The user inputs a number regarding the total number of strokes of a kanji character and makes a selection regarding the dictionaries 81 to 8n. The selected dictionary and the output of the normalization section 14 are inputted to the partial matching section 15 for matching, and candidate kanji are obtained by the candidate selection section 10 based on the results. If the correct answer cannot be obtained, redo the partial matching.

Embodiment] In an embodiment of the present invention, keys of a keyboard 2 shown in FIG. 2 are additionally used in the apparatus having the configuration shown in FIG. First, the user presses "Kanji" on the keys shown in FIG. Next, write a portion of the kanji by hand on Tablet 1. For example, in the example shown on the far left in FIG. 3, it is assumed that only ``kokoro'' is written.

When finished, press the "Enter key". At that time, "Kanji"
Since the data is immediately after the key press and the "input" key has been pressed, the keyboard control section 6 switches the output terminal so that the data selection section 11 inputs the previous data into the stroke data storage section 13. Next, the user presses the "outline" key on the keyboard 2 to input other parts of the kanji. At this time, a symbol indicating that a stroke exists at this position (SI area), such as an outline or a circle arrow in FIG. 3, is written on the tablet 1.

When the "input" key is pressed, the keyboard control section 6 takes in the data after pressing the "contour" key into the contour information extraction section 12. Next, in the normalization section 14, the stroke data storage section 1
Looking at the data of 3 and the data of the contour information extraction unit 12,
Normalize Kanji stroke data.

Next, the user inputs the total number of strokes using the numeric keys and presses the "stroke" key. The keyboard control section 6 selects a dictionary with a predetermined number of strokes using the dictionary selection section 7, and performs partial verification on the output of the normalization section 14 in the partial verification section 15.

Generally, the number of strokes for an input pattern is often smaller than the number of strokes for a dictionary pattern. Since partial matching is performed on the assumption that the input cover pattern is a part of the dictionary pattern, matching is performed while ignoring strokes in the dictionary pattern that correspond to strokes omitted on the input side. In reality, it is not known which strokes have been omitted, so basically all combinations are checked when selecting the same number of strokes as the number of strokes in the input pattern from the stroke group of the dictionary pattern in the order of appearance. . However, stroke position information may be used to reduce the number of combinations to be matched, or measures may be taken to deal with variations in the number of strokes and variations in stroke order.

After all the kanji patterns in the dictionary have been compared, candidate kanji are arranged in the candidate selection section 10 in the order of kanji that are determined to be highly similar to the input pattern. Then, the top candidates, for example, 20 characters, are output and displayed on the cathode ray tube display device 3. If you find the desired kanji among them, enter its number using the numeric keys and press the "Number" key. The device ends the input processing operation for that Kanji character. If the desired character is not found among the kanji characters, by pressing the "next candidate" key on the keyboard 2, the control section 6 causes the candidate selection section 10 to output and display the next candidate, for example, 20 characters.

[Effects of the Invention] In this way, according to the present invention, when inputting kanji to use a kanji dictionary, a part of the kanji is written on an input device such as a tablet, and by adding some data, the desired kanji can be input. can be requested and entered. Therefore, even children and foreigners who do not know much about kanji or who do not know how to read kanji can easily input and process them.

[Brief explanation of drawings]

Fig. 1 is a block diagram showing the configuration of the present invention, Fig. 2 is a diagram showing an example of the key arrangement of the keyboard in Fig. 1 as an embodiment of the present invention, and Fig. 3 is a diagram showing writing on the tablet shown in Fig. 1. FIG. 4 is a block diagram showing a conventional device. 1-tablet 2-keyboard 3...-cathode ray tube display device 6-control section 7...dictionary selection section 81-8n...dictionary 9...verification section 10-candidate selection section 11... Data selection unit 12 - Contour information extraction unit 13 - Stroke data storage unit 14... Normalization unit 15... Partial matching unit Patent applicant Fujitsu Co., Ltd. Agent
Patent Attorney Eisuke Suzuki Roro Convict E Trouble Rollo Trouble Figure 2 Figure 3

Claims (1)

[Claims] 1. A device for inputting written kanji (1), and a kanji pattern dictionary (81) storing standard patterns of kanji...
8n) and a dictionary selection section (
7) and a display device (3) that compares the input kanji with the contents of the selected kanji pattern dictionary and displays the matched kanji. In the search kanji input method for , it is equipped with a partial matching section (15) for matching the input signal as a partial pattern of kanji, executes the search by the partial matching section, and displays the search results by the partial matching section. A search kanji input method characterized by redoing partial matching if there is an error later. 2. A means (11) for classifying the data input as kanji information into two types (front and back), a means (13) for storing the former as part of the data constituting the kanji pattern, and a means (13) for classifying the data input as kanji information into two types: A search kanji input method according to claim 1, characterized by: means (12) for processing data representing information; and obtaining input pattern data from both means (14) to a partial matching unit (15). . 3. Equipped with a plurality of dictionaries (81) to (8n) in which a group of kanji to be compared is divided into a plurality of kanji patterns and each kanji pattern is stored, and a dictionary to be compared is selected for the same input data. 3. The search kanji input method according to claim 1 or 2, characterized in that the search kanji input method comprises means (7) for changing or changing the name and means (6) for instructing re-execution of the collation operation.