JP3072518B2 - Recognition result display device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Application Field] The present invention is used for automatic explanation machines, automatic announcement machines, etc.
The present invention relates to a recognition result display device for displaying a recognition result of a character or the like recognized by an optical character reading device (OCR).

[Conventional technology]

2. Description of the Related Art Conventionally, in an optical character reading apparatus, a character or the like on a read original is subjected to pattern recognition, and the recognition result is displayed on a display or output as voice by voice synthesis.

[Problems to be solved by the invention]

However, the above-described conventional recognition result display device merely displays the recognition result on a display or outputs sound, and combines image information such as figures and tables on the document with the recognition result to form the text on the document. While displaying images such as figures and tables on a document along a flow on a display, it was not possible to display a recognition result or output a sound.

For this reason, there are drawbacks that it is difficult to grasp the contents of the manuscript and it is not possible to efficiently communicate the contents to the other party.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a recognition result display device which makes it easy to grasp the contents of a document by combining a recognition result and image information, and which can efficiently transmit the contents of the document to a partner.

[Means for solving the problem]

In order to achieve the above object, the present invention is characterized in that an image is displayed corresponding to a keyword in the recognition result while outputting the recognition result.

[Action]

In the recognition result display device having the above configuration, while displaying the recognition result on the display or generating the recognition result by voice, an image is displayed on the display in accordance with the keyword such as a figure or a table in the recognition result. .

〔Example〕

 Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

FIG. 1 is a configuration diagram of an example of a recognition result display device of the present invention.

The apparatus shown in FIG. 1 stores a scanner 1 for reading an original, a central processing unit 2 for controlling the entire processing, a ROM 3 in which a program relating to optical character reading is stored, and a program relating to speech synthesis. ROM 4, an original image memory 5 for storing an original image read by the scanner 1, and a RAM 6 for storing coordinate values in the original image and additional information of a text area and an image area.
, A dictionary 7 for recognition, a keyword table 8, a voice generator 9, and a data storage RAM 10.

Next, the operation of the apparatus having such a configuration will be described with reference to the flowchart of FIG.

In step S1 of FIG. 2, an image is read from a document by the scanner 1 and stored in the original image memory 5 as an original image. Next, in step S2, a text area where characters are present and an image area where figures, tables, photographs, etc. are present are identified from the original image stored in the original image memory 5, and in step S3, each image, Save the information in RAM6.

There are two methods of identifying the area in step S2, that is, dividing the area, automatic and manual.

In the case of automatically identifying an area, a text area or an image area is identified by checking the state of an image, for example, the connection amount of black pixels. For the area determined to be an image area, a keyword KW as shown in the example of FIG. 3 is searched for in or around the area by referring to the keyword table 8, and a combination with the next character of the keyword KW is searched. It must be attached to the image as information (Figure 1, Table 1, etc.). It is also necessary to provide a function for inquiring a user when a keyword cannot be found.

On the other hand, when manually identifying the area,
Specify the text area or image area using the mouse or cursor. For example, the image is surrounded by specifying the upper left and lower right positions of the rectangular area, and the area is specified. Next, the user inputs information corresponding to the image (FIG. 1, Table 1, etc.).

FIG. 4 is a diagram showing a specific example of the above-described region identification processing, in which the original image OG is divided into a text region image TX and an image region image IM. Images IM ₁ and IM ₂ have “Figure 1” and “Figure
2 ”information I ₁ and I ₂ are added.

After performing the region identification processing in this manner, the step
The recognition process of the text area of S4, S5, S6 is performed using the dictionary 7. In the recognition processing shown in steps S4 to S6, a keyword is searched for while recognizing a certain unit. That is, in step S4, recognition is performed in a unit, for example, a word or a line, and a recognition result is displayed.
In step S5, it is determined whether or not there is a keyword registered in the recognition result as shown in FIG. 3, and if a keyword is detected, The combination pattern such as the number indicated next to the keyword is compared with the information added to the stored image, and a match is displayed on the display in step S6.
It is appropriate that the display time of the image is set to the next image.

In this manner, in the present embodiment, while displaying images such as figures and tables on the display along the flow of the text, the recognition result is displayed, and the recognition result is synthesized and output as voice. So, when applied to an automatic explanation machine,
The contents of the manuscript can be efficiently grasped and transmitted to the other party.

〔The invention's effect〕

As described below, according to the present invention, while outputting a recognition result, an image is displayed corresponding to a keyword in the recognition result. Tables and tables can be viewed on the display along the flow of the text, so that the contents of the document can be easily grasped and the contents of the document can be efficiently transmitted to the other party.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an example of a recognition result display device according to the present invention, FIG. 2 is a flowchart showing an example of processing of the device shown in FIG. 1, FIG. 3 is a diagram showing keywords, and FIG. It is a figure for explaining the example of processing. 1, scanner 2, central processing unit 3, 4, ROM 5, original image memory 6, RAM 7, dictionary 8, keyword table 9, sound generator 9, 10 Data RAM

Claims (4)

(57) [Claims] 1. A keyword registration means in which a plurality of types of area identification keywords are registered in advance, an area identification means for identifying a text area and an image area from an input image, and an area identified as an image area It detects whether there is any one of a plurality of types of area identification keywords registered in the keyword registration means in the area or at an arbitrary position around the area, and there is an area identification keyword. When it is detected that the detected area identification keyword is associated with the image information of the image area, a recognition processing means for performing a character recognition process on the text area identified by the area identification means, Display the character recognition result in the text area that has been subjected to the character recognition processing by the means and the image information in the image area Display means for displaying the character recognition result in the text area from the recognition and identification means, and when the character recognition result in the text area matches the keyword for area identification, A recognition result display device for displaying image information associated with a keyword. 2. A keyword registration means in which a plurality of types of area identification keywords are registered in advance, an area identification means for identifying a text area and an image area from an input image, and an area identified as an image area includes: It detects whether there is any one of a plurality of types of area identification keywords registered in the keyword registration means in the area or at an arbitrary position around the area, and there is an area identification keyword. When it is detected, an associating means for associating the detected area identification keyword with the image information of the image area; a recognition processing means for performing a character recognition process on the text area identified by the area identification means; Means for synthesizing and outputting a character recognition result in a text area subjected to character recognition processing by means When, and display means for displaying the image information of the image area, while the audio output of the character recognition result in the text area from the recognition processing means,
A recognition result display device, wherein when the character recognition result in the text region matches the region identification keyword, the association means displays image information associated with the region identification keyword. 3. A text region and an image region are discriminated from an input image by a region discriminating means, and an area determined as an image region by the region discriminating means is located within or around the image discriminated region. Of the plurality of types of area identification keywords registered in advance in the keyword registration means at the position of, and if it is detected that there is an area identification keyword, it is detected. The area identification keyword is associated with the image information of the image area, and a character recognition process is performed by the recognition processing unit on the area determined to be a text area by the area identification unit. While displaying the result of character recognition in the Recognition result display method characterized by displaying the image information associated with the keyword region identification when they match a word. 4. A text region and an image region are discriminated from an input image by a region discriminating means, and an area determined as an image region by the region discriminating means is located in or around the image discriminated region. It is detected whether or not any of the plurality of types of area identification keywords registered in advance in the keyword registration means is present at the position, and if it is detected that there is an area identification keyword, it is detected. The area identification keyword is associated with the image information of the image area, and a character recognition process is performed by the recognition processing unit on the area determined to be a text area by the area identification unit. The character recognition result in the text area is Recognition result display method characterized by displaying the image information associated with the keyword region identification when they match a frequency identification keywords.