JP2017073673A - Image forming apparatus and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image forming apparatus and a program with which a binarized threshold can be detected capable of removing unnecessary objects, such as wrinkles and dust from an image obtained by reading a slip or a document and leaving a form and characters unchanged.SOLUTION: An image forming apparatus according to the present invention comprises: image reading means that reads image data; master image storage means that stores a master image; binarization means that binarizes the image data on the basis of a first threshold to create binarized image data; creation means that creates background image data in which character strings are removed from the binarized image data; and determination means that, when the degree of match between the background image data and mater image is equal to or larger than a second threshold, determines that the first threshold is a threshold to binarize the image data.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an image forming apparatus and a program.

  When thin or soft documents such as slips or forms are copied or scanned, they tend to become wrinkled, and the wrinkles often appear in the read image. A method of converting a read image into a binarized image is known in order to make the wrinkle inconspicuous. As an image binarization method, an optimum binarization threshold value is detected from a luminance histogram of a read image, and binarization is performed based on the binarization threshold value.

  However, with the conventional binarization method using the luminance histogram, the wrinkles and dust portions of the read image of the document are included on the same luminance distribution side as the form, and the form (characters) and wrinkles and dust are identified. The problem is that it is impossible to detect an optimal binarization threshold value that removes only wrinkles and dust from a read image and leaves a form and characters.

  Patent Document 1 identifies a blank page and a non-blank page when printing a plurality of originals, and further identifies a blank page that should be printed and a blank page that should not be printed even for a blank page, and a print job. The technology to control is released.

  In Patent Document 2, when an original is read with a copy or a scanner, the unnecessary character such as dust attached to the character string and the reading glass is identified to determine the unnecessary material such as dust as noise. A technique for notifying a user of where an unnecessary object such as dust is attached is disclosed.

  Patent Document 3 discloses a technique for extracting features of a sample image from a plurality of accumulated sample images and outputting an image of the same type as the sample image.

  The techniques disclosed in Patent Documents 1, 2, and 3 can detect noise such as dust and unnecessary objects from the read image, but remove unnecessary objects such as wrinkles and dust from the read image of the document at the same time. It may not be complete because it may remove forms and strings that should not be done.

Therefore, the present invention has been made in view of the above-mentioned problems, and the object is as follows.
This is to detect a binarization threshold that can remove unnecessary objects such as wrinkles and dust from an image obtained by reading a slip or a form and leave a form or a character.

  In order to achieve the object, the present invention has the following features.

  An image forming apparatus according to the present invention binarizes image data based on a first threshold value, an image reading unit that reads image data, a master image storage unit that stores a master image, and binarized image data Binarizing means for generating image data, generating means for generating background image data obtained by removing character strings from the binarized image data, and the degree of matching between the background image data and the master image is equal to or greater than the second threshold value. And determining means for determining that the first threshold value is a threshold value for binarizing the image data.

  A program for executing the image forming apparatus according to the present invention includes: a means for reading image data; a means for storing a master image; and binarizing image data based on a first threshold value. Means for generating data, means for generating background image data obtained by removing character strings from the binarized image data, and when the degree of matching between the background image data and the master image is equal to or greater than the second threshold value, Means for determining that the first threshold value is a threshold value for binarizing the image data.

  According to the present invention, when reading a slip, a form, or the like, a binarized threshold value that can generate image data that removes unnecessary items such as bags and dust and leaves a form and characters can be detected. it can.

It is a figure which shows the functional block in this embodiment. It is a figure which shows the hardware constitutions in this embodiment. 6 is a follow chart for extracting a master image having a maximum image matching rate with image data read into the image forming apparatus according to the present exemplary embodiment. 6 is a flowchart for extracting a threshold value for binarizing image data read into the image forming apparatus according to the present exemplary embodiment. It is an example of generating a background image in the separation of a background image of read image data and a character image in the present embodiment.

  DESCRIPTION OF EMBODIMENTS Embodiments for carrying out the present invention will be described with reference to the drawings. FIG. 1 shows functional blocks of the image forming apparatus of this embodiment. The image forming apparatus 1 used in this embodiment includes an image reading unit 10, a master image storage unit 11, a binarization unit 12, a generation unit 13, and a determination unit 14.

  The image reading unit 10 reads image data of an image using a copy, a scanner function, or the like. The color of the image data can be read in either monochrome or color, and any method can be used as long as the image can be read using a conventionally used device such as an image scanner or a CT scanner. But you can.

  The master image storage unit 11 stores a master image as a plurality of electronic data in the image forming apparatus. The electronic data to be stored are, for example, document formats such as various contracts used in commercial transactions, slips, etc., but any data can be stored as long as it can be stored as electronic data in the device. Electronic data can be stored by reading and storing it as image data using a scanner or copy function provided in the device, or by transferring data to the device using an external I / F or the like. Any storage method can be used as long as it can be stored.

  The binarization unit 12 performs a process of binarizing the image data read by the image reading unit 10 based on a predetermined binarization threshold value to generate binarized image data. Although it is assumed that the predetermined binarization threshold is set in advance, a setting screen may be output to a display device or the like so that it can be set for each read image data.

  The generation unit 13 binarizes the image data based on a predetermined threshold value, and generates background image data obtained by removing the character string from the image data. Here, the predetermined threshold is a condition for the number of pixels for judging black and white in binarization. As an example, the read image data is digitally converted and the pixel value at each position is detected, and “all pixels with 128 or less pixels are set to white and pixels with more than 128 pixels are set to black” or “pixel value There are various settings for the binarization threshold, such as black for all other than 0. The predetermined threshold value for binarization is determined in advance, but a configuration in which it is arbitrarily changed according to user settings is also possible. In addition, it is possible to separately generate character string image data including only background image data and character strings from which character strings have been removed.

  The degree of coincidence between the background image data generated by the generation unit 13 and the stored master image is compared. As a method of comparing the degree of coincidence, comparison by comparing the number of pixels at the position, Also, a comparison based on the difference in the number of pixels at the position or other methods may be used. The comparison of the degree of coincidence can be performed for all master images, but a master image having a high similarity with the read image data, that is, a master image having the highest image coincidence rate is extracted in advance. It is preferable to keep it. As an extraction method, for example, image data may be digitally converted and compared by comparing the number of pixels at each position, or may be compared by a difference in the number of pixels, or other methods may be used. Further, when reading the image data, it is preferable to have a function of rotating the paper to align the position, assuming that the paper is tilted and does not align with the stored master image. .

  When the degree of coincidence between the background image data generated by the generation unit 13 and the stored master image is equal to or greater than a predetermined threshold value, the determination unit 14 uses the binarization threshold used by the generation unit 13. It is determined that the value is a threshold value for binarizing the image data. Since all the character strings are removed from the background image data and the image data does not completely match as a comparison result, the threshold value used here is set somewhat lower. Further, the binarization threshold value is changed based on the determination of the degree of matching of the images, and the background image data generated based on the changed binarization threshold value and the master image are repeatedly compared. Configuration is required. In order to generate a binarized image from which wrinkles and dust that are captured when the image data is finally read is generated, an optimal binarization threshold value is extracted in advance. In many cases, it is necessary to change at least once.

  Next, a hardware configuration of the image forming apparatus according to the present exemplary embodiment will be described with reference to FIG. Note that the configuration of the image forming apparatus is not necessarily the same as that shown in FIG. 2, and it is sufficient if it has hardware capable of realizing the present embodiment.

  The image forming apparatus 1 executes a predetermined program to store a CPU 101 for realizing overall control of the image forming apparatus 1 and a program read by the CPU 101 when the image forming apparatus 1 is powered on. When the image forming apparatus 1 is powered off, the read-only memory (Read Only Memory (ROM)) 102, the random access memory (RAM) 103 used as the work memory by the CPU 101, and the image forming apparatus 1 are turned off. An HDD 104 capable of storing various data records, an input device 105 including a mouse and input keys, and a display device 106 including a display using a panel such as a liquid crystal display and an organic EL display. I have. Information such as a predetermined threshold value used in this embodiment is stored in the ROM 102, the RAM 103, and the like.

  In addition, the image forming apparatus 1 includes an image processing unit 107, a scanner unit 108, a copy unit 109, a communication I / F 110, and an image comparison unit 111 as other exclusive functions. These exclusive functions operate in conjunction with each other based on a command from the CPU 101.

  The image processing unit 107 is read by the scanner unit 108 or the like and performs various processes on the image data. Various processes include a series of processes related to image processing including image brightness and color correction, image rotation processing, and the like. Also, a process for removing the character string of the read image data is performed. Various processed image data is stored in the HDD 104 or the like as a recording medium.

  The scanner unit 108 reads image data from the image data using an optical device (not shown). As the scanner method, a conventional technique such as an image scanner or a CT scanner is used, and any method can be used as long as image data can be read.

  The copy unit 109 copies image data from a document to another printing paper. As a copying method, any method can be used as long as it can transfer image data, such as a transfer type using a photoconductor.

  The communication I / F 110 is an interface for inputting / outputting various information to / from an external device or an electronic device such as a computer. Input / output image data and the like via the communication I / F 110. The communication I / F 110 includes, for example, a USB port, a LAN port, a wireless LAN port, and the like. Any communication I / F 110 may be used as long as data can be transmitted / received to / from an external device.

  The image comparison unit 111 reads out the read image data and the stored master image, and compares both images. In the present embodiment, binarization processing is performed. The binarization processing can also be performed by the image comparison unit 111. Of course, a configuration in which the image processing unit 107 performs the binarization processing is also possible. is there.

Example: (Extraction of master image with maximum image matching rate)
Next, the procedure of this embodiment will be described with reference to a flowchart. FIG. 3 is a flowchart for extracting the master image having the maximum image matching rate with the image data read into the image forming apparatus according to this embodiment. Here, it is an example of a method for extracting in advance a master image having a high similarity with the read image data, and is not limited to this method, and other extraction methods may be used so that a master image having a high similarity is not extracted in advance. A simple configuration is also possible.

  First, image data of an image is read using a scanner or the like (step 1). As a scanner method, an image may be read by a conventional technique such as an image scanner or a CT scanner. Next, a master image stored in a storage medium or the like is extracted (step 2).

  In order to align the read image data, the inclination of the image data is corrected (step 3). In this embodiment, a fixed rotation angle is set for the image data, and the image is rotated little by little based on the rotation angle, and correction is performed so as to match the vertical and horizontal directions and inclination of the master image. The correction method for aligning the position of the image data is not limited to this.

  When the correction of the image data alignment is completed (step 4), the image matching rate between the image data and the master image is calculated (step 5). Here, as a method for calculating the image coincidence rate, image data may be digitally converted and compared by comparing the number of pixels at the position, or may be compared by the difference in the number of pixels at the position. It doesn't matter. There is also a method in which image data is not digitally converted but compared based on color, image brightness, or the like.

Example: (Extraction of optimum binarization threshold of image data)
Next, a procedure for extracting an optimum binarization threshold value of image data based on the image data and the extracted master image will be described with reference to a flowchart. FIG. 4 is a flowchart for extracting the optimum binarization threshold value of the image data read into the image forming apparatus according to the present embodiment.

  First, the read image data is binarized based on a predetermined threshold value (step 8). Here, the predetermined threshold value is a condition of the number of pixels for determining black and white at the time of binarization. As an example, the read image data is digitally converted and the pixel value at each position is detected, and “all pixels with 128 or less pixels are set to white and pixels with more than 128 pixels are set to black” or “pixel value There are various threshold values for binarization, such as “all other than 0 are black”. Of course, other conditions can be set.

  Next, a background image is generated by removing the character string from the binarized image data (step 9). As a method of removing a character string, there is a method of identifying a character string and a graphic as shown in FIG. 5 and separating them into an image of only a character string and an image of only a graphic. As another method, there is a method of identifying a character string from image data and whitening a pixel corresponding to the position of the identified character string, but in any case, if a background image from which the character string is removed can be generated Any method is acceptable.

  Next, the degree of matching between the generated background image and the master image is calculated (step 10). As a method of comparing the degree of coincidence, a comparison by comparing the number of pixels at a position, a comparison by a difference in the number of pixels at a position, or other methods may be used. The master image, which is the comparison target in this embodiment, calculates the image matching rate by a specific method, and the master having the highest image matching rate is extracted in advance, but the degree of matching with all the master images is calculated. May be. In addition, since all the character strings are removed from the background image data, the image data does not completely match as a comparison result.

  Next, it is determined whether the calculated degree of matching is larger or smaller than a predetermined threshold (step 11). The threshold used here is set somewhat lower. When the calculated degree of coincidence is larger than a predetermined threshold value, it is determined that the binarization threshold value at that time is a threshold value for binarizing this image data (step 12).

  When the calculated degree of coincidence is equal to or less than a predetermined threshold value, a predetermined value is added to the binarization threshold value of the image data at this time to change the binarization threshold value (step 13). Here, when changing the binarization threshold value, the threshold value may be changed not by increasing the threshold value for decreasing the black area but by decreasing the threshold value for increasing the black area. It can be changed by addition or subtraction.

  Based on the newly set binarization threshold value, the procedure from Step 8 to Step 10 is repeated, the degree of matching between the generated background image and the master image is calculated, and the calculated degree of matching is a predetermined level. It is determined whether it is larger or smaller than the threshold value (step 14). When the calculated degree of coincidence is larger than a predetermined threshold value, it is determined that the binarization threshold value at that time is optimum for this image data (step 12).

  When the calculated degree of coincidence is equal to or less than a predetermined threshold value, the processing from step 13 is repeated. If the degree of matching does not become larger than the predetermined threshold even after repeating a certain number of processes, the binarization threshold when the degree of matching is closest to the predetermined threshold of the degree of matching Is determined to be optimal for this image data (step 15). Although the number of times is not specified in the binarization threshold value changing process, the number of times the process is repeated may be specified. For example, if the number of times of matching is less than a predetermined threshold even if the processing is repeated up to five times and the processing is performed five times, it is the closest to the predetermined threshold among the five times of data. A binarization threshold value based on the degree of coincidence is detected as a threshold value to be binarized.

  The image data is binarized based on the binarization threshold value determined in step 12 and step 15 (step 16). Since the binarized image to be generated is a binarized image that has been compared with a master image in which no wrinkles, dust, etc. are reflected, and has been determined to have the highest degree of coincidence with the master image, It is image data from which unnecessary items such as dust are removed, and even when printing is performed, unnecessary items such as soot and dust are not attached.

  In the above, the Example which performs this embodiment with a suitable form was described. Although a specific specific example has been shown and described here, various modifications and form changes of the specific example can be made without departing from the spirit and scope of the claims.

  According to the present invention, when a slip or a form is stored as image data by copying or scanning, unnecessary data such as a bag or dust can be removed to generate image data that leaves a form or characters. I am confident that it will be possible to detect the threshold value.

  As long as the present invention can be realized, the contents of the apparatus used, the number of apparatuses, and the like are not limited to the present embodiment.

DESCRIPTION OF SYMBOLS 10 Image reading means 11 Master image storage means 12 Binarization means 13 Generation means 14 Judgment means

JP 08-335209 A JP 2011-130267 A Japanese Patent Laid-Open No. 04-348475

Claims (7)

Image reading means for reading image data;
Master image storage means for storing a master image;
Binarization means for binarizing the image data based on a first threshold value to generate binarized image data;
Generating means for generating background image data obtained by removing character strings from the binarized image data;
When the degree of coincidence between the background image data and the master image is equal to or greater than a second threshold value, the first threshold value is determined to be a threshold value for binarizing the image data. Judgment means,
An image forming apparatus comprising: A comparison means for comparing the image data with the master image;
A master image extracting means for extracting a master image having a maximum degree of coincidence with the image data by the comparing means;
The determination means determines the degree of coincidence between the background image data and the master image having the largest degree of coincidence with the image data.
The image forming apparatus according to claim 1. The comparison means rotates the image data based on a predetermined rotation angle, and compares the degree of coincidence by comparing pixel values at each position of the image data and the master image.
The image forming apparatus according to claim 2. When the determination unit determines that the degree of matching between the background image data and the master image is smaller than the second threshold value,
The binarization means binarizes the image data based on a third threshold value that is changed by adding or subtracting a predetermined value to or from the first threshold value.
The image forming apparatus according to claim 1, wherein the image forming apparatus is an image forming apparatus. The determination means determines whether the degree of coincidence between the background image generated by binarization based on the first threshold value and the third threshold value and the master image is the second threshold value. When it is smaller than the value, it is determined that the binarization threshold value having the maximum degree of coincidence is the optimum binarization threshold value for the image data.
The image forming apparatus according to claim 4. Binarizing the image data based on the binarization threshold value determined by the determination means;
The image forming apparatus according to claim 1, wherein the image forming apparatus is an image forming apparatus. Means for reading image data;
Means for storing a master image;
Means for binarizing the image data based on a first threshold value to generate binarized image data;
Means for generating background image data obtained by removing character strings from the binarized image data;
When the degree of coincidence between the background image data and the master image is equal to or greater than a second threshold value, the first threshold value is determined to be a threshold value for binarizing the image data. Means,
A program for running

Images