JP2002190940A - Image-processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an easy-to-use image-processor while compensating for the incompleteness in image recognition processing. SOLUTION: When image data of a manuscript are read (#1), top and bottom are recognized for each page (#2), and the image data and the recognition result for each page are stored (#3). Then, based on the result of top and bottom recognition, the image data for each page are subjected to rotation processing as needed (#4), and one copy is outputted to the image data after the rotation processing (#5). Then, the result of the top and bottom recognition of each page is displayed on an LCD panel, and is subjected to an operator's individual correction (#7). The stored recognition result is corrected by operator's input (#8).

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to an image processing apparatus for reading an image of a document and recognizing their attributes (top and bottom, color, etc.) and performing image processing. More specifically, the present invention relates to an image processing device that corrects an image according to a recognized attribute.

[0002]

2. Description of the Related Art In recent years, in an image forming apparatus such as a copying machine, the spread of a plurality of originals can be automatically copied in a short time by the spread of an automatic original feeding function.
In addition, even if a bundle of originals to be copied includes an original in which the directions of the images are not aligned, the direction of the image is recognized using the character direction recognition technology, that is, the so-called top and bottom recognition technology, and the direction of the output image is changed. 2. Description of the Related Art An image forming apparatus including an image recognition processing device for aligning images has been proposed. For example, Japanese Patent Application Laid-Open No. HEI 10-117281 has been proposed as an application technique of the top and bottom recognition.
In the technique disclosed in this publication, when the reliability of the top and bottom recognition of the read image is lower than a predetermined level, the process is interrupted, and the operator is made to determine the direction of the image. In this way, recognition is incomplete,
It is designed to cope with erroneous recognition that occurs due to the difficulty in recognizing the image of the document or the restriction on the processing time.

[0003]

However, in the above-mentioned technique, when the reliability of recognition is low, the processing is temporarily interrupted even if the answer is correct, so that the technique is inconvenient. Further, even if the operator intentionally attempts to turn a specific page upside down, the page is automatically corrected. In this case, if the reliability of the image recognition of the page is high, a copy of the mode desired by the operator cannot be obtained unless the top / bottom recognition function itself is canceled and redone. However, when this was done, there was the expectation that the top and bottom recognition processing would not be performed on other pages. Further, since the recognition result is not stored, the operator has to make a new decision each time the same copy is output by memory recall.

The present invention has been made to solve the above-mentioned problems of the conventional image processing apparatus. That is, an object of the present invention is to provide an easy-to-use image processing apparatus while supplementing incomplete image recognition. More specifically, it is an object of the present invention to provide an image processing apparatus capable of correcting a recognition result of image data for each page.

[0005]

SUMMARY OF THE INVENTION An image processing apparatus according to the present invention for solving this problem includes an image reading means for reading an original to generate image data, and attribute information for each page included in the image data. Attribute recognizing means for recognizing image data, image data storing means for storing image data and attribute information for each page, correcting means for correcting image data for each page using attribute information for each page, and correcting means for correcting the image data for each page. Forming means for outputting the image data corrected by the correcting means, and displaying the image data corrected by the correcting means by the image forming means. The corrected image data information is displayed on the display unit, and the attribute information of each page of the image data is corrected by the operator. It is intended to accept.

According to this image processing apparatus, first, the image reading means reads an original and generates image data.
Then, the attribute recognizing means recognizes the attribute information of each page included in the generated image data. The image data storage means stores the image data and the attribute information for each page. On the other hand, the correction unit corrects the image data of each page based on the attribute information as needed. The image forming unit outputs image data of all pages including the image data corrected by the correcting unit. By once outputting the image data of all pages in this manner, the operator can confirm whether the attribute recognition of each page is correct. Thereafter, the information of the image data corrected by the correction means is displayed on the display means. Thereby, the correction of the attribute information of each page of the image data by the operator is accepted. When corrected by the operator, the attribute information relating to the correction is stored in place of the attribute information before correction stored in the image data storage means.

[0007] As described above, the image data corrected based on the top-bottom recognition is output once for each page so that the operator can confirm it. For this reason, if there is an erroneous recognition, only the page can be corrected by a correction instruction from the operator. Further, when the attribute information is corrected, the image data storage means can store the attribute information after correction instead of the attribute information before correction. For this reason, the image data desired by the operator can be correctly output without the operator having to confirm each time the memory recall is performed. When once outputting image data of all pages, a reduced image,
An image displaying only the recognition result may be replaced with the image data itself.

[0008] A specific example of the attribute information recognized by the attribute recognizing means is a page direction. In this case, it is desirable that the correction unit performs correction for rotating the image.

Another specific example of the attribute information recognized by the attribute recognizing means is a color included in the image of the page. The color as the attribute information of the page is a distinction such as full color, two colors of black and red, and a single color of black.
In this case, it is desirable that the correction unit performs correction to reduce the number of colors of the image. Further, when the operator makes a correction for increasing the number of colors of the image, it is desirable to prompt the operator to re-read the document of the page related to the correction.

It is desirable that after the image data corrected by the correcting means is output by the image forming means, the information of the image data corrected by the correcting means is displayed on the display means by an operation of an operator. In this way, the operator checks the output image once,
Only when the correction is necessary, the information of the image data can be displayed on the display means. Then, a correction instruction can be given to a page that requires correction, which is more convenient.

[0011]

Embodiments of the present invention will be described below in detail with reference to the accompanying drawings.
This embodiment embodies the present invention as a digital copier that reads an image of a document, stores it as image data, forms an image based on the stored image data on a sheet, and outputs it as a copy. It is.

(First Embodiment) FIG. 1 shows the overall block configuration of a digital copier according to this embodiment. The digital copier includes an image data input unit 101 for capturing and inputting image data for each page, an attribute recognizing unit 102 for recognizing attribute information of the input image data for each page, and a plurality of images. A job management table 103 for managing a series of images to be processed as a job,
An image information management table 104 for managing each image data input from the image input unit 101, an image memory unit 105 for storing the recognition result of the image data and its attribute information, and an operation unit 1 for receiving an input from an operator
06 and a display unit 1 for displaying a message to the operator
07, an image processing unit 108 for performing image correction processing on image data based on attribute information, and an image data output unit 109 for forming and outputting image data on paper.

The image data input unit 101 is a scanner that reads an image on a document and generates image data. The attribute information in the present embodiment is the orientation of the page. Therefore, the image correction processing in the present embodiment refers to the 180
゜ rotation. The operation unit 106 and the display unit 107
, As shown in FIG. 2, constitutes an operation / display panel having a numeric keypad, a start key, and other various keys, and an LCD panel. The screen of the LCD panel in FIG. 2 is a screen displayed when a “Utility” key at the upper left is pressed. Further, when the "top and bottom recognition" button on the LCD screen is pressed, the screen shifts to the screen of FIG.

In the digital copying machine having the above configuration, the image data input by the image data input unit 101 is stored in the image memory unit 105 for each page. In parallel with this, the attribute recognition unit 102 recognizes the page direction (top and bottom) of the image data for each page. The result of this recognition is sent to the image memory unit 105 via the image information management table 104, and is stored in association with the image data of the page. When the image data is output from the image data output unit 109, image processing is performed in the image processing unit 108 based on the recognition data of the orientation of the page. The image data output unit 109 outputs image data that has undergone image processing. At this time, the image data subjected to the image processing and the recognition data included in the image information management table 104 can be confirmed on the display unit 107, and can be corrected by the operation unit 106 as necessary.

Next, the image output control in this embodiment will be described. In order to perform the top / bottom recognition processing as in the present embodiment, it is necessary to select the “Yes” button while the screen of FIG. 3 is displayed on the LCD panel of the operation / display panel. However, in the image forming apparatus of the present embodiment, the state in which the button is selected is the default. Therefore, unless an operation for canceling this is performed, the top and bottom recognition processing is performed. When the start key on the operation / display panel is pressed in this state, the image output processing including the top and bottom recognition processing is performed as shown in the flowchart of FIG.

In the flow of FIG. 4, first, an image of a document is scanned by the image data input unit 101 (# 1).
As a result, image data for each page of the document is obtained. Next, the image recognition is performed by the attribute recognition unit 102 (# 2). That is, for each page,
It is determined whether the image is upward or upside down. The recognition result is stored in the image information management table 104, and is also sent to the image memory unit 105 and the image processing unit 108.
Then, the image data and the result of the top / bottom recognition are stored in the image memory unit 105 in association with each page (#
3).

Next, a rotation process of the image data is performed in the image processing section 108 (# 4). This process is a process of rotating the image by 180 °. The target is
Only the page determined to be upside down as a result of the top and bottom recognition in # 2. Or, if the page determined to be upside down was rather a majority,
Conversely, it is also possible to target only the page determined to be upward. Then, the image data output unit 109
Is printed out (# 5). The output image is a rotated image of the page that has been subjected to the rotation processing in # 4. As a result, the operator can obtain a copy of one copy and determine whether or not the result of the top / bottom recognition is incorrect.

At this time, a screen shown in FIG. 5 is displayed on the LCD panel of the operation / display panel. This is because the subsequent processing is changed depending on whether or not the result of the top / bottom recognition has been incorrectly recognized (# 6). When the operator presses the “recognition correction” button on the screen of FIG. 5, the digital copier determines that the top and bottom recognition has been erroneously recognized (# 6: Yes). Then, the screen shown in FIG. 6 is displayed on the LCD panel of the operation / display panel. This screen is a screen for displaying the result of the top / bottom recognition for each page and accepting correction by the operator (# 7).

The screen shown in FIG. 7 may be displayed prior to the screen shown in FIG. The screen in FIG. 7 displays all the pages in the job on the screen in FIG. 6 to be corrected, or displays only the page with low recognition reliability in # 2 and is displayed and corrected. Is a screen for the operator to instruct whether or not to make the setting. That is, when the "all pages" button on the screen of FIG. 7 is pressed, all pages are displayed and corrected. On the other hand, when the button of “ambiguous page” is pressed, only the page whose recognition reliability is low is targeted.

Returning to the description of the screen shown in FIG. On this screen,
The page number to be corrected and the top / bottom recognition result are displayed as icons. When the operator touches the icon, the page number of the touched icon is designated as a page requiring correction of the top and bottom recognition result. Touch the same icon again to cancel the specification. The operator designates each page having the misrecognition by touching the icon, and then presses the “OK” button to exit the screen of FIG. Then, the recognition result is corrected for each of the designated pages (# 8).

Then, the process returns to step # 4. That is, a rotation process of the image data is performed on the page whose top-bottom recognition result has been corrected. Then, print output is performed in # 5.
At this time, the screen shown in FIG. 8 is displayed on the LCD panel of the operation / display panel. This is because the operator selects whether to output only the corrected pages one by one or to output the required number of copies for all pages. Then, the process proceeds to step # 6 again, and further correction is possible.

If there is no erroneous recognition in the top and bottom recognition in # 6 (# 6: No), the processing related to the top and bottom recognition ends. That is, when the operator presses the start key while the screen of FIG. 5 is displayed on the LCD panel,
# 6: No is determined. Thereafter, after setting the number of copies as appropriate, the same copy can be output by memory recall. Note that the determination of # 6 is also No when the "Data Discard" button is pressed on the screen of FIG. 5, but in this case, memory recall is of course not possible.

As described above, the top and bottom recognition is performed by the attribute recognition unit 102, and the image data for one job, including the image data corrected by the image processing unit 108, is output once, and the operator confirms it. I have to. Therefore, even if there is an erroneous recognition, it can be promptly corrected for each page. Of course, if there is no misrecognition, the operator has promptly obtained one correct copy. Also, the image data memory unit 105
It stores the read image data and the top and bottom recognition results,
Further, even when the recognition result is corrected, it is stored instead of the recognition result before correction. For this reason, the operator does not have to check each time the memory is recalled, and can output the image data desired by the operator accurately. Further, even when the operator intentionally turns a specific page upside down, it is possible to easily obtain a copy as intended by the operator while receiving the effect of automatic top and bottom recognition for other pages.

(Second Embodiment) This embodiment is a digital copier which performs color recognition as well as top and bottom recognition. The basic block configuration, the operation unit 106 and the display unit 107 are not particularly different from those shown in FIGS. However, the image data input unit 101 can acquire color image data. Attribute recognition unit 1
In 02, color recognition is performed in addition to top and bottom recognition. In the present embodiment, the colors of the image data recognized by the attribute recognition unit 102 are two types, “black monochrome” and “full color”. The image data output unit 109 can perform color printing. Furthermore, an operation / display panel (operation unit 106
And the screen displayed on the LCD panel when the “Utility” key is pressed on the display unit 107)
As shown in FIG. 9, a “color recognition” button is included. By pressing the "Color Recognition" button on this LCD screen,
Move to the screen of FIG.

In order to perform both the top and bottom recognition processing and the color recognition processing as in this embodiment, the "Yes" button is selected while the screen of FIG. 3 is displayed on the LCD panel of the operation / display panel. It is necessary to select not only the button but also the "Yes" button while the screen of FIG. 10 is displayed.
However, in the image forming apparatus of the present embodiment, the state in which the button is selected on the screen of FIG. 10 is the default as in the screen of FIG. Therefore, unless an operation for canceling them is performed, the top and bottom recognition processing and the color recognition processing are performed. When the start key on the operation / display panel is pressed in this state, image output processing including top-down recognition processing and color recognition processing is performed as shown in the flowchart of FIG.

In the flow of FIG. 11, first, an image of a document is scanned by the image data input unit 101 (# 1).
1). Thereby, color image data for each page of the document is obtained. Next, the attribute recognition unit 102 performs top and bottom recognition and color recognition of the image data (# 12).
That is, for each page, it is determined whether the image is directed upward or inverted, and whether the image is black or full color. The recognition result is stored in the image information management table 104,
Further, it is also sent to the image memory unit 105 and the image processing unit 108. Then, for the page recognized as a monochrome black color image, the color image data is reduced and converted into monochrome image data (# 13). This reduction processing reduces the amount of image memory used. Further, a later image data output unit 1
In step 09, the required processing time and toner consumption are also reduced. Then, the image data and the results of the top and bottom recognition and color recognition are stored in the image memory unit 105 in association with each page (# 14).

Next, a rotation process of the image data is performed in the image processing section 108 (# 15). Then, one copy is printed out from the image data output unit 109 (# 16). As a result, the operator can obtain a copy of one copy and determine whether the result of the top-bottom recognition and the color recognition has an erroneous recognition.

At this time, the screen shown in FIG. 5 is displayed on the LCD panel of the operation / display panel (# 17). When the operator presses the "recognition correction" button on the screen of FIG. 5, the digital copier determines that the top-bottom recognition or the color recognition has an erroneous recognition (# 17: Yes). The screen shown in FIG. 6, ie, the screen shown in FIG.
A screen for displaying the result of the top-bottom recognition for each page and accepting the correction is displayed (# 18). As in the case of the first embodiment, the screen of FIG. 7 may be displayed prior to the screen of FIG.

When the operator presses the "OK" button after inputting the correction of the top and bottom recognition by touching the icon on the screen of FIG. 6, the screen shifts to the screen of FIG. This screen is a screen for displaying the result of color recognition for each page and accepting correction by the operator. On this screen, the page number to be corrected and its color recognition result are displayed in the form of icons. When the operator touches the icon, the page number of the touched icon is designated as a page that requires correction of the color recognition result. Touch the same icon again to cancel the specification. The operator designates each page having the misrecognition by touching the icon, and then presses the “OK” button to exit the screen shown in FIG. Then, the correction of the recognition result is executed for each page designated for the top and bottom recognition or the color recognition (# 19). This correction includes reduction conversion of color image data to monochrome image data for the page when correction of the color recognition result from full color to black monochrome is included.

Next, it is determined whether or not the correction of the input recognition result includes a correction of the color recognition result from black to full color (# 20). If the correction of the color recognition result from black to full color is included (# 2
0: Yes), and instructs the operator to rescan the page (# 21). That is, the operator is prompted to press the start key on the operation / display panel after placing the page of the document on the document tray of the image data input unit 101. If the operator does so, a rescan is performed (# 22). As a result, the color image data of the page is acquired again and replaced with the monochrome image data before correction (# 23).

Then, the flow returns to step # 15. If the correction of the color recognition result from black monochrome to full color is not included in # 20 (# 20: No), the process bypasses the processes of # 21 to # 23 and immediately returns to # 15. Returning to step # 15, image data rotation processing is performed on the page whose top-bottom recognition result has been corrected. Then, a print output is made in # 16. At this time, the LCD panel of the operation / display panel
The screen of FIG. 8 is displayed. Thus, the operator can select whether to output only the corrected pages one by one or to output a required number of copies for all pages. Then, the process proceeds to step # 17 again, and a state in which further correction is possible is made.

If there is no erroneous recognition in the top and bottom recognition and the color recognition in # 17 (# 17: No), the processing related to the top and bottom recognition and the color recognition ends. Thereafter, as long as the image data is not discarded, the operator can output the desired number of copies by memory recall.

As described in detail above, in the digital copying machine according to the present embodiment, the image data input unit 101
The attribute recognition unit 102 performs top-bottom recognition and color recognition for each page of the image data read in step (# 2, # 12). Based on the recognition result, the image data is appropriately rotated or reduced for each page (# 4, # 13, # 15), and one copy is output using the processed image data (# 4). #
5, # 16). By once outputting one copy, the operator can check whether the recognition result (top and bottom, color) of each page is correct.

After that, the results of the top and bottom recognition and color recognition are displayed on the LCD panel, and the results of the top and bottom recognition and color recognition for each page of the image data are accepted for correction by the operator (# 7, # 18). .
For this reason, even if there is an erroneous recognition in the top and bottom recognition or the color recognition, it can be corrected for each page as intended by the operator. Further, even if the operator intentionally turns a specific page upside down, the other pages can be output as intended by the operator while receiving the effect of automatic top and bottom recognition. The image memory unit 105 stores the read image data and the recognition result of each page, and further stores the corrected recognition result in place of the correction result before correction if the recognition result is corrected. I am trying to do it. For this reason, the image data desired by the operator can be output accurately without the operator having to make a new correction every time the memory is recalled.

The present embodiment is merely an example, and does not limit the present invention. Therefore, naturally, the present invention can be variously modified and modified without departing from the gist thereof. For example, in the present embodiment, only two types of page orientations, that is, an upward direction and an upside-down direction, are considered, and the rotation processing is limited to only 180 ° rotation. However, the present invention is not limited to this, and the page may be set to four directions including rightward roll and leftward roll, and the rotation processing may include 90 ° rotation. Of course, it is conceivable to use more types of page orientations. In addition, there are two types of color recognition of an image, such as two types of black and red, as well as two types of monochrome black and full color.
Other types may be included.

When one copy is output once at # 5 or # 16, instead of printing the image data itself, a reduced image or an image displaying only the recognition result may be output. Further, it is conceivable to output them together with an identification code such as a page number in a thumbnail form on one sheet of paper. In the second embodiment, when correcting the recognition result of step # 18, first, correction of top and bottom recognition is received (FIG. 6), and then correction of color recognition is received (FIG. 12). May be reversed. Alternatively, both corrections may be accepted simultaneously on one screen.

[0037]

As is apparent from the above description, according to the present invention, there is provided an image processing apparatus which is easy to use while compensating for imperfect image recognition. More specifically, an image processing apparatus capable of correcting a recognition result of image data for each page is provided.

[Brief description of the drawings]

FIG. 1 is a schematic overall configuration diagram of an image processing apparatus.

FIG. 2 is a diagram showing an operation unit and a display unit.

FIG. 3 is a diagram illustrating a display example of a display unit.

FIG. 4 is a flowchart illustrating control when performing a top-bottom recognition process.

FIG. 5 is a diagram illustrating a display example of a display unit.

FIG. 6 is a diagram illustrating a display example of a display unit.

FIG. 7 is a diagram illustrating a display example of a display unit.

FIG. 8 is a diagram illustrating a display example of a display unit.

FIG. 9 is a diagram illustrating a display example of a display unit.

FIG. 10 is a diagram illustrating a display example of a display unit.

FIG. 11 is a flowchart illustrating control when performing top-bottom recognition and color recognition processing.

FIG. 12 is a diagram illustrating a display example of a display unit.

[Explanation of symbols]

 101 image data input unit 102 attribute recognition unit 105 image memory unit 106 operation unit 107 display unit 108 image processing unit 109 image data output unit

──────────────────────────────────────────────────の Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) H04N 1/46 H04N 1/46 Z (72) Inventor Kazuhiro Ueda 2-3-13 Azuchicho, Chuo-ku, Osaka-shi No. Osaka International Building Minolta Co., Ltd. (72) Inventor Kazuyuki Kawabata 3-13 Azuchicho, Chuo-ku, Osaka City Osaka International Building Minolta Co., Ltd. F-term (reference) 2H027 EK15 FA04 FA28 FA35 FB05 FB19 FD00 FD08 GA20 GA23 GA34 GA45 GA47 GB14 GB16 HA07 ZA07 2H030 AD07 AD08 AD09 BB63 5C062 AA05 AB20 AB22 AB23 AC04 AC24 5C076 AA24 AA26 BA06 5C079 HB03 LA02 LA36 LA37 LB15 MA11 MA17 NA17 PA02

Claims (5)

[Claims] An image reading means for reading a document to generate image data, an attribute recognizing means for recognizing attribute information for each page included in the image data, and an image storing image data and attribute information for each page Data storage means, correction means for correcting the image data for each page based on the attribute information for each page, image forming means for outputting the image data corrected by the correction means,
Display means for displaying information on the image data corrected by the correction means, wherein the image data corrected by the correction means is output by the image forming means, and then the information on the image data corrected by the correction means is output. An image processing apparatus for displaying on the display means and accepting an operator's correction of attribute information of each page of image data. 2. The image processing apparatus according to claim 1, wherein the attribute information recognized by the attribute recognition unit is a page direction, and the correction unit performs a correction for rotating the image. apparatus. 3. The image processing apparatus according to claim 1, wherein the attribute information recognized by the attribute recognizing means is a color included in the image of the page, and the correcting means performs correction for reducing the number of colors of the image. An image processing apparatus characterized by performing. 4. The image processing apparatus according to claim 3, wherein when the operator makes a correction for increasing the number of colors of the image, the operator is prompted to re-read the document of the page related to the correction. Image processing device. 5. The method according to claim 1, wherein:
In the image processing apparatus described above, after the image data corrected by the correction means is output by the image forming means, information of the image data corrected by the correction means is displayed on the display means by an operation of an operator. An image processing apparatus comprising: