JPH0250783A - Method for modifying recognition result in character recognizing device - Google Patents

Abstract

PURPOSE:To rapidly and easily execute modifying work by displaying the image information of a input character corresponding to the word of a recognition result based on a word coordinate from a recognizing part on a display scope, and making a position on the image information display of the word confirmable. CONSTITUTION:The word of a recognition result is stored at a word unit with cord information in a word storing area 18, and area coordinates (x1, y1) and (x2, y2) of the word segmented as one word are stored in a word coordinate storing area 19. The relation of the segmented word and the coordinate makes the coordinates of the front part upper edge and the rear part lower edge of a square area involving the segmented word as area coordinates. Based on word coordinates (x1, y1) and (x2, y2) in a word memory 14 of a recognizing part 1, since a storage result sentence and input image information are correspond-displayed and the both are indicated with making correspond, them to each other, the modification, etc. of the recognition result can be executed while watching the input image information.

Description

DETAILED DESCRIPTION OF THE INVENTION <Industrial Technical Field> The present invention relates to a recognition result correction method in a character recognition device.

Prior Art> A character recognition device that recognizes character information in a document by computer processing photoelectrically converts the character information to be recognized, such as alphanumeric characters, cuts out the photoelectrically converted electrical signals character by character, and performs recognition. The optical character reading system fl (OC
R) is known.

In this type of character recognition device, conventionally, when a recognized character has a low correct reading rate and is judged to be suspicious, so-called rejected, the rejected character is displayed on a display unit using a cathode ray tube (CRT), etc. The operator corrects the rejected characters using a correction means such as a keyboard while checking the display and comparing the rejected characters with the original.

However, if the rejected characters are e.g.
- If the problem is due to contact characters such as Ill, inverted characters such as "1liJ, [oO", etc., or an error in cutting out a character line or string, the operator may It is difficult to judge correctly, and it has to be checked against the manuscript one by one, which requires a lot of time and effort to make corrections, which has the disadvantage of poor work efficiency.

, <Problems to be solved by the invention> This invention solves the above-mentioned drawbacks and improves the a
An object of the present invention is to provide a recognition result correction method that allows the recognition result to be corrected very efficiently.

Means for Solving the Problems> In order to achieve the above object, the recognition result correction method of the present invention calculates image information of input characters corresponding to the word of the recognition result based on the word coordinates from the recognition unit. This is characterized in that the words are displayed on the same screen so that the position of the word on the image information display can be confirmed.

<Operation> According to the instruction information of the recognition result word from the input unit, the control unit displays the image information portion in the image memory corresponding to the word in a manner that distinguishes it from others. This makes the correspondence between the large recognition result display and the input image information even clearer.

<Example> The present invention will be described in detail below based on the drawings.

FIG. 1 shows a block diagram of a character recognition device to which the recognition result correction method according to the present invention can be applied, 1 is a recognition unit, 4
is the input section, 3 is the control section, and 2 is B? This is a display unit having a result display 2a and an image (image information) display 2b.

The recognition unit 1 includes an image scanner 10, as shown in FIG.
It consists of an image memory 11, a cutting section 12, a recognition section 13, and a word memory 14. The cutting unit 12 detects spaces between words and cuts out words, and the cutting information is sent to the word memory 14 via #112a and used as inter-word delimiter information. Regarding word cutting, an application filed by the same applicant (Japanese Patent Application No. 6l-310412)
Since it is disclosed in detail in , the explanation will be omitted.

FIG. 3(A) is a diagram showing an example of the structure of the word memory 14 in the recognition section 1 with seven storage formats. Word storage area 18
, a word coordinate storage area 19, and a flag storage area 20, and in this embodiment, the storage capacity is 50 bytes each, 16 bytes each, and 1 bit each, but the storage capacity is not limited to this.

In the word storage area 18, words resulting from recognition are stored in code information word by word, and in the word coordinate storage area 19, area coordinates (xlwy
l), (in Ly2) are stored.

The relationship between the cut out word and the coordinates is as shown in FIG. 3(B), and the coordinates of the front upper end and rear lower end of the rectangular area containing the cut out word are taken as area coordinates. Flag (
F) In the case of this embodiment, the memory area 20 is comprised of 1-bit memory area) F20a and X-pel check F20b.
, word length F20c and one symbol F20d.

Reject F 20g is written as Bn in the storage unit 13.
"1" for words containing character patterns that were not found.
In other cases, it is "0".

The spell check F20b is set to "1" for words that fail in the spell check process performed on a word-by-word basis, and to "0" for other words.

The word length F20c is set to "0" if the cut word length is greater than a predetermined number of characters, and is set to "1" if it is less. This is based on the empirical rule that when a word has a large number of characters, it is difficult to make mistakes. Symbol F20d
is set to "1" if the recognized word is a symbol, and "0" otherwise.

This is because spell checking is not effective in the case of symbols and they are prone to errors.

FIG. 5 shows the operational flow of the present invention. First, on the display screen 23 of the display unit 2 in FIG. The image information 23b of the input characters is displayed (
SI), FIG. 4 shows a case where both are displayed on the same screen, but the recognition result sentence and the image information may be displayed on separate screens.

Next, a word on the recognition result sentence 23a, for example, a word whose recognition result appears to be incorrect upon visual inspection, is specified using the cursor control key on the input unit 4 (Sl).

Then, the image information of the input character at the designated coordinate part of the word coordinate storage area 19 of the word memory 17 corresponding to the designated word is specified (S3), and it is, for example, marked n(b)k! as shown in figure m4. ? It is surrounded by and displayed to distinguish it from others (S4).

Although the above embodiments are aimed at English words, they can be implemented similarly for Japanese sentences. In this case, what is referred to as word units in the above embodiments refers to word units.

Effect> As is clear from the above explanation, according to the present invention, the recognition unit 1
The word locus [(xi-yl), (x2
Based on +y2), we have made it possible to display the memorized result sentences and input image information in correspondence, and to give instructions by associating both, so it is possible to modify the aa results without looking at the input image information, so the correction work is easier. It can be done quickly and easily.

[Brief explanation of the drawing]

! @Figure 1 is a block diagram of a character recognition device to which the method of the present invention can be applied, Figure 2 is a block diagram of the recognition unit, and Figure 3 (A)
Figure 3 CB is a diagram showing the storage format of word memory.
) is a diagram showing the relationship between words and word coordinates, FIG. 4 is a diagram showing a display example of the present invention, and FIG. tj & 5 is a diagram showing the operation flow of the present invention. 1: Recognition unit, 2: Display unit, 3: Control unit, 4: Input unit, 1
0: Image scanner, 11: Image memory, 12: Cutting section, 13: Recognition processing section, 14: Word memory, 12a
: Signal line, 15: Word memory information output line, 16: Image memory information output line, 17: Word memory 7-ormat, 18:
Word storage area, 19: Word coordinate storage area, 20: Flag storage head, 23: Display screen, 23a: Recognition result sentence, 2
3b Two-person character image information agent Patent attorney Sugiyama
Itaru Tsuyoshi (1 other person) (A) Figure Figure

Claims (1)

[Scope of Claims] In correcting the result recognized by the recognition unit, image information of the input character corresponding to the word of the recognition result is displayed on the display screen based on the word coordinates from the recognition unit, and the word is displayed on the display screen. A method for correcting recognition results in a character recognition device, characterized in that the position of a character on an image information display can be confirmed.