JPH0424885A - Reading processor - Google Patents

Abstract

PURPOSE:To improve reading environment by extracting a color character picture corresponding to a color characteristic as the object of reading from a medium on which a color character is recorded, and outputting it to a character recognizing part. CONSTITUTION:In the case of the medium designated for the reading, a picture input part 1 outputs its picture signal to a color characteristic storage part 3, and the color characteristic storage part measures and stores the color characteristic, and in the case of the document of the object of the reading, the picture input part 1 outputs its picture signal to a picture extracting part 4, and the picture extracting part 4 extracts the character picture corresponding to the color characteristic or the character picture in a color mark area as the object of the reading on the basis of the color characteristic stored in the color characteristic storage part 3, and outputs it to the character recognizing part 5. Then, the character recognizing part 5 executes the inclination correction, etc., of the extracted document picture, and segments the charac ter picture by every one character, and recognizes and converts it into a character code, and a voice synthesizing part 6 converts this character code into voice waveform, and a voice is outputted from a voice output part 7. Thus, the reading environment is improved.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a reading processing device that outputs a reading voice based on the recognition result of characters on a document or book.

[Prior art]

Conventionally, this type of reading processing device was published in IEICE Technical Report Vol. 1.87 No. 334 1.
Published on January 22, 1988, there was something disclosed on pages 79 to 86. FIG. 2 is a block diagram showing the synthesis of the above-mentioned conventional reading processing device of this type. As shown in the figure, the reading processing device includes an image input section 21, an image processing section 22, a character recognition section 23, a voice synthesis section 24, and a voice output section 25.

The image input unit 21 inputs documents and books page by page,
An image processing section 22 performs image processing such as tilt correction on this image, a character recognition section 23 cuts out and recognizes the output document image characters, and outputs audio as the final output. .

[Problem to be solved by the invention]

However, with the conventional reading processing device having the above-mentioned configuration, it has been impossible for healthy people to select the reading method described below, which they do on a daily basis when reading documents and books. In addition, it could not be said that it was a good reading environment for people with disabilities such as blind people or physically disabled people.

For example, if the document or book you are reading is an essay, (1) Read the bibliographic information such as the title and author, (2) Read the summary, (3) Read the title of chapter 2, etc., (4) References. (5) Knowing the above, decide whether to read the entire text or not.

The present invention has been made in view of the above-mentioned problems, and in order to eliminate the above-mentioned problems, it is possible to freely select and read the text in the color mark part of documents and books, so that sighted people can read it on a daily basis. To provide a reading processing device capable of reading in a manner similar to the browsing method of reading documents and books, which is useful for improving the reading environment for handicapped people.

[Means to solve the problem]

In order to solve the above problems, the present invention has configured a reading processing device as follows.

An image input section that scans the medium and outputs an image signal, a color characteristic storage section that measures and stores the color characteristics on the designated reading medium, and a color character corresponding to the color characteristics from the medium on which the color characters are written. The present invention is characterized in that it includes an image extraction unit that extracts an image as a reading target and outputs it to a character recognition unit.

Further, the medium is one on which a color mark is added or printed, and the image extraction unit detects the color mark based on the color characteristics and extracts the character image within the color mark area as a reading target. Features.

Further, the reading designation medium is characterized by being a card on which a letter indicating the type of reading object is written.

[Effect]

Since the reading processing device of the present invention is configured as described above, in the case of a medium designated for reading, the image input section outputs the image signal to the color characteristic storage section, and the color characteristic storage section measures and stores the color characteristics. In the case of a document to be read, the image input section outputs the image signal to the image extraction section, and the image extraction section extracts a color character image or a color character image corresponding to the color characteristics based on the color characteristics stored in the color characteristics storage section. Character images within the color mark area are extracted as objects to be read out and output to the character recognition unit. The character recognition unit performs processing such as tilt correction on the extracted document image, and
A character image is cut out, recognized, and converted into a character code, and a speech synthesis section converts this character code into an audio waveform, and audio is output from an audio output section.

That is, the reading processing device of the present invention adds or prints a color mark according to the importance to a desired part of a document book in the process of browsing the document book by a healthy person, and then freely edits the text in the color mark portion. Since the text is selected and read aloud, it is possible to read in a manner as close as possible to the browsing of documents and books that sighted people do on a daily basis.

By using a card with Braille characters indicating the type of reading target as the designated reading medium, it is also useful for improving the reading environment for people with disabilities such as the blind.

〔Example〕

Hereinafter, one embodiment of the present invention will be described based on the drawings.

FIG. 1 is a block diagram showing the configuration of a reading processing device that is an embodiment of the present invention. As shown in the figure, the reading processing device includes an image input section 1, a reading mode switching section 2, a color characteristic storage section 3, an image extraction section 4, a character recognition section 5, a speech synthesis section 6, and a speech output section 7. It is the composition.

The image input unit 1 scans a sheet of paper as shown in FIGS. 4 and 5 with main scanning in the horizontal direction (XX) and sub-scanning in the vertical direction (YY), and records marks and/or marks written on the paper. Or an image signal corresponding to one or both of the shading and color of the document 9 drawing, such as a density multilevel signal or R (red), G (green)
, B (blue) multilevel signals are output to the reading mode switching section 2.

The reading mode switching section 2 switches the output light of the image signal from the image input section 1 to the color characteristic storage section 3 or the image extraction section 4 using a switch or the like (not shown). In the case of a reading card as shown in FIG. 4, the reading mode switching unit 2 outputs the image signal from the image input unit 1 to the color characteristic storage unit 3,
In the case of a document to be read as shown in the figure, it is output to the image extraction section 4.

The color characteristic storage section 3 measures and stores the color characteristics of a predetermined area (ms) of the reading card as shown in FIG.

The image extraction unit 4 extracts a document image within a color mark area written on a document or book as shown in FIG. 5 as a reading target based on the color characteristics.

The character recognition unit 5 performs processing such as tilt correction on the extracted document image, cuts out the character image for each character, recognizes it, and converts it into a character code.

The character recognition unit 5 converts the character code into a voice waveform,
The signal is output to the audio output section 7, and the audio is output from the audio output section 7.

FIG. 3 is an explanatory diagram of color characteristics showing the principle for classifying reading cards and document marks to be read by color; FIG. Figure (b
) is an explanatory diagram of the density range of the mark.

When using the principle shown in FIG. 3(a), the image input section 1 requires a color scanner capable of outputting, for example, an RGB density multilevel signal.

When using the principle shown in FIG. 3(b), the image input section 1 requires a monochrome scanner or the above-mentioned color scanner capable of outputting a multilevel density signal.

The chromaticity diagram in FIG. 3(a) is a schematic diagram created by a known method, in which the RGB density signal is converted to the XYZ system using the coordinate conversion formula <1), X = 2.7689 R+ 1.7517 G+
1.1302 BY=R+4.5907G+
0.0601BZ=0.0565G+5.
5943B (1) Based on the obtained x, y, and z, calculate the chromaticity coordinates x r y using equation (2), and x = X / (X + Y + Z) y = y / (x + y - +-z) (2)
This uses orthogonal coordinates based on the obtained color coordinates X T 3'. The x and y represent hue and saturation, and Y represents lightness.

Conversion from the above RGB system to the XYZ system [using formula (1)] Calculation to obtain X, 7 representing hue and saturation [formula (2)]
], input R, G, B, x, y, z
using a ROM that outputs and inputs x, y, z,
By converting using a ROM that outputs x and y in integer format, calculation time is not required and processing speed can be increased.

In the case of Fig. 3(b), the document or drawing to be extracted is white;
A color that is expressed in black, the density level of the mark on the document is a density level that does not overlap with the density of the white and black, and the color of the mark has a density level that does not overlap with each other for each color. Color classification is possible by selection. Moreover, the above Y may be expressed similarly.

The color characteristics stored in the color characteristics storage section 3 in FIG. It is assumed that the chromaticity diagram is stored using the chromaticity diagram shown in FIG. 3(a).

FIG. 4 is an example of a reading card, where Ml is a visible character string indicating the type of book, M2 is a visible character string indicating the name given to the area to be read out, and TI and T2 are each a character string Ml. , the Braille string corresponding to M2, ms is the character string M
This is a color mark that has the same color characteristics as a color mark that is added or printed in the reading target area of a document or book, corresponding to M2.

The reading card is structured as shown in Figure 4, so
It can be easily operated by both healthy people and visually impaired people, and can be created arbitrarily by the user of the device.

In addition, the book type can be changed to other patents, paperbacks, magazines, etc.
In-house documents, newspapers, etc. can be read aloud, and by preparing a reading card with an optimal name assigned to each book type, the range of reading objects can be expanded.

In the following explanation, reading cards are for papers, and the area names are "Abstract", "Bibliographical matters", "Title of chapter 1, section 1", "References", and the color marks are green and red, respectively. , blue, and yellow.

Next, the operation of the color characteristic storage section 3 will be explained. The color characteristic storage unit 3 stores chromaticity coordinates x obtained from RGB image signals.
, convert y into an integer, for example, multiply it by 100 and convert it into an integer, X%, y
It has a two-dimensional memory (color map) whose address is %, and when reading of a reading designation card as shown in Fig. 4 starts, the color map and full text reading flag are set to 10.
．． Initialize to . Next, scanning is performed on a predetermined area (m s )
When it reaches , the RGB image signal that can be inputted is converted into integers of X% and y%, and 11'' is written in the color map with this X%13'% as the address. Note that if the predetermined area (m s ) is black, that is, the image signals R, G, B
If all are below a certain level, set the full text reading flag.

Next, the mark identification process and the image extraction process in the mark area by the image extraction section 4 will be explained using FIGS. 5 and 6.

First, mark identification processing is performed as follows. The image input unit 1 scans a sheet of paper as shown in FIG. 5 with main scanning in the horizontal direction (XX) and sub-scanning in the vertical direction (YY), and outputs an RGB multi-level signal using the above formula. (1)
, (2) to convert to chromaticity coordinates x, y, and the x,
X%, which is y converted to an integer.

Refer to the color map in the color characteristic storage unit 3 using y%,
If the color map is "1", it is identified that the scanning point is on the mark.

The image extraction process within the mark area will be explained using FIG. 6. In FIG. 6, no mark image is detected in the sub-scanning row YYi, but in the YY2 row, the mark image with the color characteristic j that has been stored in the color map is identified by the mark identification process described above, and furthermore, in the YYa row, the mark image is connected to the mark image. and the sandwiched image (XX31 w-XX32 b
), the image extraction section 4 outputs the interposed image to the character recognition section 5 as an image to be read. In the above-described procedure, the image sandwiched between the mark images is output as an image to be read, but when the sub-scanning line YYn is reached, there is no longer a connected mark image, so the extraction of the image to be read is completed.

It should be noted that the determination as to whether or not the mark images are connected is made using the following equation (3) using a known technique, for example, as detailed in Japanese Patent Publication No. 60-55868.

XX(i-1)kb≦XXikw XX(i-1)kw≦XX1kb (3) However, XX(i-1)kb and XX(i-1)kw are respectively the previous scanning row YYi- of the scanning row YYi row. 1, the change point from color characteristic “OJ” to color characteristic “jJ” and color characteristic “j” to color characteristic “0”
XX1kb, XX indicating the X coordinate of the pair of change points to ``
i kw is the change point from the color characteristic r□ of the scanning row Yi to the same color characteristic 1j and the color characteristic ``j'' of the scanning row Yi, respectively.
, indicates the X coordinate of the change point bear from the color characteristic "OJ". However, if the full text reading flag is set a in the color characteristic storage section 3, the above processing is not performed and the entire image is output from the image input section 1. The image is output from the image extraction unit 4 as an image to be read out.

Note that when the document or book to be read is achromatic, the image extraction section 4 is provided with an achromatic image extraction means, and the image output from the character recognition section 23 is only an achromatic image. It is possible to remove the influence of mark images.

The achromatic image extraction means uses a known method, for example, R, G
, H signals are all at nearly the same level.

Further, the achromatic color extraction means is connected with a binarization means,
Let the digitized image be the image to be read. Note that the threshold value for binarization may be determined using any known method.

As shown in Figure 5, specify ``Summary 1'' as the reading target, and
When the document shown in the figure is input, the image surrounded by the mark m2 is output as the image to be read, and after characters are extracted and recognized from the image, voice synthesis is performed and a reading voice is output. A similar process is performed when "1 bibliographic item" or "title of 1 chapter, 1 section" is specified as the object to be read out.

Note that the position, size, seven colors, number, and shape of marks to be added or printed on reading cards, documents, and books are not limited to those described above.

In the above explanation, it is assumed that the color mark part on the book is selected, but it is also possible to prepare a book printed with the text in a different color instead of the color mark, and configure it to select the color of the text. good. When characters of different colors are printed in each reading area, the image extraction unit 4 does not need the above-mentioned mark identification processing and mark area image extraction processing, and directly extracts the image with the set reading color characteristics from the reading target. Extract as an image.

Furthermore, by adding the following functions to the reading mode switching unit 2, automation of the switching operation becomes possible.

(1) A function that allows you to compare the size of a reading card with a text book and compare the size of the paper.

(2> A function that adds a discrimination mark to a predetermined position on a reading card and detects the presence or absence of a discrimination mark.

(3) A function to provide a notch at a predetermined position on a reading card and detect the notch.

(4) A function to provide an input location exclusively for reading cards (the image input location is the same) and detect whether the input location is passed through.

〔Effect of the invention〕

As explained in detail above, according to the present invention, a color image corresponding to the color characteristics of the document or a desired part of the document can be applied to a color characteristic according to the importance level in a procedure in which an able-bodied person browses through the document. The user can add or print out the color marks to be printed, and then freely select and extract the documents in the color marked portion of the book and read them aloud. The present invention has the excellent effect of being able to provide a reading processing device that is useful for improving the reading environment for people with disabilities such as blind people and people with physical disabilities, and can also provide a reading environment that is almost the same as a visual reading environment even for able-bodied people. is obtained.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of the reading processing device according to the embodiment of the present invention, FIG. 2 is an explanatory diagram of the reading area of the conventional embodiment, FIG. The figure shows an example of a reading card, and FIG. 5 shows an example of a document to be read.
FIG. 6 is an explanatory diagram of image extraction. In the figure, 1...image input unit, 2...reading mode switching unit, 3...color characteristic storage unit, 4...image extraction unit, 5...character recognition unit, 6...Speech synthesis unit, 7...Speech output unit. Patent applicant: Oki Electric Industry Co., Ltd. 4: Friendly patent attorney: Takashi Kumagai (1 other person) ! Explanation of Kasusha Figure 3 1ei*-)-sq''IiLyr-Km Figure 4 Figure 5 Meeting of Ika Ikamen ii±)-Sun/Monday 60 Figure 6

Claims (3)

[Claims] (1) In a reading processing device that includes a character recognition unit that recognizes characters on a medium and a voice synthesis unit that synthesizes voice based on the recognition result recognized by the character recognition unit, and outputs the recognition result as voice. , an image input section that scans the medium and outputs an image signal; a color characteristic storage section that measures and stores the color characteristics on the designated reading medium; A reading processing device comprising: an image extracting section that extracts a corresponding color character image as a reading target and outputs it to the character recognition section. (2) The medium has a color mark added or printed thereon, and the image extraction unit detects the color mark based on the color characteristics and extracts the character image within the color mark area as a reading target. The reading processing device according to claim 1, characterized in that: (3) The reading processing device according to claim 1 or 2, wherein the reading designation medium is a card on which Braille characters representing the type of reading object are written.