JPH04162183A - character recognition device - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to a character recognition device that recognizes characters in a photographic document inputted from an image reading device such as a scanner.

In conventional character recognition devices, when it is known in advance that a photographic area exists on a paper to be recognized, a paper to be recognized 41 is read as multivalued image data 42 by an image reading device as shown in FIG.
It was read as , extracted the area, and performed character recognition.

(Problems to be Solved by the Invention) However, the conventional character recognition device described above has a problem in that when a photographic area exists on the paper to be recognized, the entire paper to be recognized is read as multi-valued image data, resulting in a large amount of memory usage. was there.

The present invention is intended to solve the above-mentioned conventional problems, and it is an object of the present invention to provide a character recognition device that uses a small amount of memory even when a photographic area exists on a paper to be recognized.

(Means for Solving the Problems) In order to achieve the above object, the present invention provides a character recognition device that reads the entire paper to be recognized as binary image data, cuts out areas of this binary image data, and processes photographic areas. The position and size of the image are known, and only the photographic area is reloaded as a multivalued image. That is, the character recognition device includes an image reading device startup section, an image reading device,
It is equipped with an image memory, an area cutting section, a character recognition section, a layout section, and a display section, and an image reading device reads the entire paper to be recognized as binary image data and stores it in the image memory. The output unit cuts out the photo area and text area, and the text area undergoes recognition processing in the character recognition unit. If a photo area exists, only the photo area of the paper to be recognized is processed as multivalued data by the image reading device. By reading the data and storing it in the image memory, the multi-valued image data of the photographic area can be stored with a minimum amount of memory, and the photographic area can be reproduced as multi-valued image data in the recognition result.

(Function) Therefore, according to the present invention, by rereading only the photo area as multi-valued image data, it is possible to reduce the amount of memory used.

(Embodiment) FIG. 1 shows the concept of character recognition in an embodiment of the present invention. In FIG. 1, the character recognition device reads the entire paper 11 to be recognized as binary image data as shown at 12, cuts out areas from this binary image data, learns the position and size of the photo area, and identifies only the photo area. 13
It is reread as multivalued image data as shown in .

FIG. 2 shows the configuration of a character recognition device in an embodiment of the present invention. In FIG. 2, 21 is an image reading device startup unit, 22 is an image reading device capable of reading binary image data and multivalued image data, 23 is an image memory that stores read image data, and 24 is a read image. an area cutting unit that cuts out data into a text area and a photo area;
25 is a character recognition unit that performs character recognition; 26 is a layout unit that lays out the character recognition results and photo data according to the area cutting results; and 27 is a display unit that displays the layout results. FIG. 3 shows a flow chart of image editing of the character recognition device shown in FIG.

Next, the operation of the above embodiment will be explained with reference to FIGS. 2 and 3. First, the image reading device starting section 21 starts the image reading device 22 to read the entire paper to be recognized as binary image data (Sl).The image reading device 22 reads the entire paper to be recognized as binary image data and images it as binary image data. It is stored in the memory 23 (S,). Area cutting section 24
Then, based on the image data in the image memory 23, the binary image data is cut out into a text area and a photo area, and the positions and sizes of the text area and photo area are retained (S8).
. The character recognition unit 25 performs character recognition on the character area in the binary image data and holds the result (S,). The image reading device starting section 21 causes the image reading device 22 to read only the photographic region on the paper to be recognized as multivalued image data based on the position and size of the photographic region held in the area cutting section 24. Start (S,).

The image reading device 22 reads only the photo area on the paper to be recognized as multivalued image data and stores it in the image memory 23 (S,). At this time, the binary image data previously stored in the image memory 23 is no longer needed, so
Memory can be saved by storing the newly read and stored multivalued image data in a form that overwrites the binary image data. The layout unit 26 lays out recognition results from the area extraction results from the area extraction unit 24, the character recognition results from the character recognition unit 25, and the multivalued image data in the image memory 23, and sends the layout results to the display unit 27. Do (S,). The display section 27 displays the layout result (S,).

(Effects of the Invention) As is clear from the above embodiments, the present invention has the effect of reducing memory usage by reading and storing only the photo area on the paper to be recognized as multivalued image data. .

[Brief explanation of the drawing]

FIG. 1 is a conceptual diagram of character recognition in an embodiment of the present invention, FIG. 2 is a block diagram of a character recognition device in an embodiment of the present invention, and FIG. 3 is a flowchart of image editing in an embodiment of the present invention. FIG. 4 is a conceptual diagram of conventional character recognition. 11.41... Paper to be recognized, 12... Binary image data, 13.42... Multivalued image data, 21...
Image reading device starting unit, 22... Image reading device, 23.
. . . Image memory, 24 . . . Area cutting section, 25 . . . Character recognition section, 26 . Patent applicant Matsushita Electric Industrial Co., Ltd.-1)

Claims (1)

[Claims] In a character recognition device comprising an image reading device startup section, an image reading device, an image memory, an area cutting section, a character recognition section, a layout section, and a display section, the image reading device reads the entire paper to be recognized. is read as binary image data and stored in the image memory, the area cutting unit cuts out the photo area and the text area, the text area is recognized by the character recognition unit, and the photo area exists. In this case, by reading only the photo area of the recognition target paper as multi-value data from the image reading device and storing it in the image memory, the multi-value image data of the photo area can be stored with a minimum amount of memory usage, and A character recognition device characterized in that a photographic area can be reproduced as multivalued image data in recognition results.