JPH0475185A - Input device - Google Patents

Abstract

PURPOSE:To speedily and easily correct a misread character by providing a 2nd recognition means which rerecognizes the misread character based on instruction information and a recognition result. CONSTITUTION:This device is provided with a 1st recognition means 3 which segments and recognizes character patterns in an image inputted from a document image input means 1 and outputs candidate characters for the respective character patterns. Further, the device is provided with a display means 5 which displays the recognition result, an instruction means 11 which instructs the misread character or correctly read character of the displayed recognition result, and the 2nd recognition means 3a which rerecognizes the misread character based on the instruction information and recognition result. Consequently, the operation for the correction is facilitated and the correct recognition result can be obtained speedily.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an input device that recognizes and manually inputs characters written in documents and drawings.

[Prior Art] FIG. 3 is a block diagram showing the configuration of a conventional input device disclosed in, for example, Japanese Unexamined Patent Publication No. 175878/1983.

In the figure, 1 is a document image input means for optically reading documents, drawings, etc. and inputting them as image information, 2 is a document image storage means for storing input image information, and 3 is for extracting character patterns from the image information. Recognition means for cutting out and recognizing, 4
5 is a recognition result storage means for storing the recognition result, 5 is a result display means for displaying the recognition result, and 6 is a recognition result correction means for modifying the recognition result. The solid line in the figure is the process flow, and the double line is the data. It shows the flow.

FIG. 4 is a diagram showing the results of each processing stage, and the same figure (a
7 in ) is the input cover pattern, 8 is the character pattern cutout position candidate, 9 in ) is the recognition result, and 10 in FIG.

Next, the operation will be explained.

The document image manually generated by the document image human input means 1 is stored in the document image storage means 2 and sent to the recognition means 3. The recognition means 3 obtains character pattern cutout position candidates 8 from the document image, outputs recognition results 9 based on the recognition results for each combination and knowledge processing results considering word combinations, etc., and performs recognition. The result is stored in the result storage means 4. This is displayed on the result display means 5 and checked visually, and then in the recognition result correction means 6.

Correct the recognition result 9, "Izoshin Shinobi" to "Izo no Kenji" using the keyboard.

[Problem to be solved by the invention] Since the conventional input device is configured as described above, there was a problem that all misread characters had to be corrected, which required a lot of time and effort for inputting. .

This invention was made to solve the above-mentioned problems, and its purpose is to provide a device that can automatically correct misread characters by simply indicating the misread characters, and allows quick and easy input. shall be.

[Means for Solving the Problems] An input device according to the present invention cuts out and recognizes a character pattern from an image input from a document image input means,
a first recognition means for outputting a candidate character for each character pattern; a display means for displaying a recognition result; and an instruction means for instructing a misread character or a correctly read character of the displayed recognition result;
The second recognition means re-recognizes misread characters based on the instruction information and the recognition result.

[Operation] The input device of the present invention recognizes with the first recognition means, and if there is an error in the result, indicates it with the instruction means, and uses information such as words based on the information to re-read the misread characters. Since recognition is performed using the recognition means described in Section 2, the operation for correction becomes simple and correct recognition results can be obtained quickly.

[Example] Hereinafter, an example of the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention, in which solid lines indicate the flow of processing, double lines indicate the flow of data, and the same reference numerals as in the conventional example are the same or equivalent. The explanation thereof will be omitted. 48 in the figure is
Among the recognition results of the first recognition means 3, the degree of similarity between the extraction position candidate 8 of the character pattern shown in FIG. 4(a), the recognition candidate character of each character pattern, and the similarity between each character pattern and each recognition candidate character. 11 is an instruction means for the user to visually check the recognition results displayed on the result display means 5 and indicate the misread characters found by the user using a mouse or the like. , 3a, based on the instruction information from the instruction means 11 and the recognition result stored in the first recognition result storage means 4a, recognizes the misread character again using the relationship such as word information with the correct characters before and after it. , is the second recognition means that outputs the optimal result.

FIG. 2 is a diagram showing the results of each processing step in this example for the input cover turn 7 of FIG. 4 ta),
9 in FIG. 2 fa) is the recognition result by the first recognition means 3;
10 in FIG. 2(b) is the recognition result by the second recognition means 3a.

Next, the operation will be explained.

In the first recognition means 3, as in the conventional case, as shown in FIG.
A pattern that is considered to be one character is cut out as a character pattern from the input cover pattern 7 as shown in FIG. However, as shown in character pattern extraction position candidate 8, if the extraction position is not clear, multiple combinations of characters are extracted. Next, recognition is performed for each combination of cuts,
The recognition candidate characters and the degree of similarity are output and stored in the first recognition result storage means 4a. Then, the result display means 5 displays the recognition result 9 that is cut out and recognized at the cutout position that is considered to be the most correct at this point (indicated by the O mark in the cutout position candidate 8).
to be displayed. In this recognition result, since "Izo Shinnin" is misread, these are indicated by the instruction means 11.

Next, in the second recognition means 3a, since it became clear in the first recognition result storage means 4a that the first characters "ga" and "sense" were correct and the characters between them were misread, " By finding the character following ``picture'' and the character preceding ``sensei'' using word information and information on recognition candidate characters, ``image'' and ``recognition'' can be obtained, respectively. Therefore,
The recognition result 10 can be automatically obtained by simply giving instructions without correcting misread characters. In this way, according to this device, by indicating a misread character, it is automatically corrected in relation to the correctly read characters before and after it, and
Enables input to be done quickly and easily.

In the above embodiment, an example was shown in which a misread character is specified, but the principle is the same even if a correctly read character is specified.

[Effects of the Invention As described above, according to the present invention, misread characters can be corrected simply by indicating the misread characters or correctly read characters, so that the desired result can be quickly and easily obtained. There is.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention.
FIG. 2 is a diagram showing the results of each processing step in FIG. 1, FIG. 3 is a block diagram showing the configuration of the conventional example, and FIG. 4 is a diagram showing the results of each processing step of the conventional example. 1 is a document image input means, 2 is a document image storage means, 3 is a first recognition means, 3a is a second recognition means, 4 is a recognition result storage means, 4a is a first recognition result storage means, and 5 is a result The display number and 11 are instruction means. In addition, in the figures, the same reference numerals indicate the same or equivalent parts. Agent Patent Attorney Jun Miyazono Figure 1 Figure 3 Figure 4 Figure 2 Vocabulary Words

Claims (1)

[Scope of Claims] An input device that recognizes and inputs characters written in documents, drawings, etc., comprising: a document image input means for optically reading a document, etc. and inputting it as image information; and this document image input means. A first recognition means that cuts out and recognizes character patterns from an image input from the image and outputs candidate characters for each character pattern, a display means that displays recognition results, and misreading of the displayed recognition results. An input device comprising an instruction means for indicating a character or a correctly read character, and a second recognition means for re-recognizing a misread character based on the instruction information and the recognition result.