KR20190049169A - Apparatus and method for classifying images stored in a digital device - Google Patents

Abstract

According to one embodiment of the present invention, provided is an image classification apparatus for detecting personal information. The image classification apparatus classifies an image to be optically recognized in accordance with a predetermined classification standard and removes an image which is less likely to include personal information before performing optical recognition, thereby increasing optical recognition efficiency and reducing costs by reducing the number of optical images to be optically recognized for detecting personal information.

Description

[0001] APPARATUS AND METHOD FOR CLASSIFYING IMAGES STORED IN A DIGITAL DEVICE [0002]

More particularly, the present invention relates to an image classifying apparatus, and more particularly to an image classifying apparatus that classifies an image to be optically recognized according to a predetermined classification criterion, removes an image having low probability of existence of personal information before performing optical recognition, The present invention relates to an image classification apparatus and method for searching for a personal information image that can reduce the number of image recognition apparatuses and improve the optical recognition efficiency and cost.

In recent years, research on optical recognition technology for analyzing various information from images by applying optical recognition technology to images photographed from cameras has been actively conducted. Through such optical recognition technology, objects included in an image are identified It is possible to extract text information.

On the other hand, in recent years, the spread of smart phones has been rapidly increasing. Most of these smart phones are equipped with high-performance cameras, so that users can take pictures anytime and anywhere. Therefore, Is stored.

Currently, most users of smartphones use about 50% of their storage space for photo storage, with an average of more than 1000 photos. For these pictures, various information included in the image including the personal information can be extracted using the optical recognition technology.

However, it takes a lot of time and money to confirm personal information through optical recognition of images in many images.

As an application that performs image analysis by applying the above-described optical recognition technology, for example, Google Cloud Vision API and the like exist. When the optical recognition technology is used, as the number of images to be optically recognized increases, There is a problem that the time and cost required for optical recognition increases exponentially as shown.

Accordingly, in order to reduce the time and cost required for optical recognition, it is necessary to reduce the number of images as much as possible by classifying the images to be subjected to optical recognition in advance in correspondence with information to be analyzed before applying optical recognition.

(Patent Literature)

Korean Registered Patent No. 10-1735664 (registered on May 08, 2017)

Therefore, in an embodiment of the present invention, an image to be optically recognized is classified according to a predetermined classification criterion, and an image with low probability of existence of personal information is removed before performing optical recognition, And to reduce the cost of the image classification apparatus and method for searching personal information images.

The image classifying apparatus according to an embodiment of the present invention may include an image classifying unit that classifies all images in a first image group including a plurality of images according to a preset reference to classify the image to be removed, An image classifying unit for generating a second image group including an image, and an image producing unit for selecting a predetermined number of images among the images included in the second image group and synthesizing the selected images into one image.

The image classifying unit may further include a resolution-based classifying unit for classifying the image into the removal object image by checking the resolution of each image in the first group and if the resolution is lower than a preset reference resolution.

The image classifying unit may further include a metadata-based classifying unit for classifying the image into the removal object image if the metadata information for each image in the first group is checked and the information related to the copyright is included .

The image classifying unit may classify the images captured by the same camera for each image in the first group and generate the image based on the generation time information of each image captured by the same camera, And a consecutive photograph classifying unit for classifying the remaining images into the removal object image, excluding one image selected arbitrarily from a plurality of images included in the range.

The image classifying unit may further include a focus-based classifier for classifying each image in the first group as the removal object image, and classifying the image as the removal object image if the degree of focus mismatch is greater than or equal to a predetermined reference mismatch degree .

The image production unit may include a resolution check unit for checking the resolution of each image included in the second image group, and a controller for reducing each image included in the second image group to a predetermined ratio based on the checked resolution An image processing method comprising: an image reduction unit; an image processing unit that selects the predetermined number of images from the reduced image, and positions the reduced image in an area divided according to the predetermined number on the one image, And a synthesizing unit.

The image production unit may further include an image position storage unit for storing identification information for each of the reduced images included in the one image and area information on each of the reduced images, .

The apparatus may further include an image generation unit that generates an image in which the personal information exists in the plurality of images included in the one image based on the optical recognition result of the one image, And an image classifying unit for classifying the image.

The personal information may be text information included in the one image.

According to still another aspect of the present invention, there is provided an image classification method comprising the steps of: classifying an image to be removed into a first image group including a plurality of images, Generating a second image group including the remaining images, and selecting a predetermined number of images from the images included in the second image group to produce one image.

The classifying step may include the steps of: checking resolution of each image in the first group; and classifying the image into the removal object image if the resolution is lower than a predetermined reference resolution.

In addition, the classifying step may include: checking metadata information of each image in the first group; and classifying the metadata as the removal target image if the metadata information includes copyright-related information The method comprising the steps of:

The classifying may further include classifying images taken by the same camera for each image in the first group, identifying generated time information of each image taken by the same camera, And classifying the remaining images into the removal object image, excluding one image selected arbitrarily among a plurality of images whose generation time of the image is included in the predetermined time interval range based on the time information .

The classifying step may include the steps of examining a degree of focus mismatch for each image in the first group, and classifying the image as the mismatching image when the degree of focus mismatch is equal to or greater than a predetermined reference mismatching degree information The method comprising the steps of:

In addition, the producing may include: checking the resolution of each image included in the second image group; and reducing each image included in the second image group to a predetermined ratio based on the examined resolution Selecting the predetermined number of images from among the reduced images, and compositing the reduced image into the one image by locating the reduced image in an area divided according to the predetermined number on the one image .

The producing step may further include storing identification information of each of the plurality of reduced images included in the one image and area information of each of the reduced images located on the one image .

The method may further include classifying an image in which the personal information exists and an image in which the personal information does not exist based on the optical recognition result for the one image with respect to the plurality of images included in the one image And further comprising:

According to an embodiment of the present invention, there is provided an image classification apparatus for searching personal information, comprising: classifying an image to be optically recognized according to a predetermined classification criterion; removing an image having low probability of existence of personal information before performing optical recognition; It is possible to reduce the number of optical recognition target images for searching personal information, thereby increasing the optical recognition efficiency and reducing the cost.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a view showing the time required for optical recognition and the cost according to the number of conventional optical recognition target images.
2 is a functional block diagram of an image classification apparatus for searching personal information image according to an embodiment of the present invention;
3 is a diagram illustrating a composite image in which a plurality of images are synthesized according to an embodiment of the present invention.
4 is a functional block diagram of a first image classifying unit according to an embodiment of the present invention;
5 is a functional block diagram of an image producing unit according to an embodiment of the present invention;
6 is a flowchart of an operation control for sorting an image in an image classification apparatus according to an embodiment of the present invention.
FIG. 7 is a conceptual diagram illustrating an area in which personal information is recognized through optical recognition in one image in which a plurality of images are synthesized according to an embodiment of the present invention. FIG.
FIG. 8 is an illustration of optical recognition time and cost results after image classification according to an embodiment of the present invention; FIG.

Hereinafter, the operation principle of the present invention will be described in detail with reference to the accompanying drawings. In the following description of the present invention, detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear. The following terms are defined in consideration of the functions of the present invention, and these may be changed according to the intention of the user, the operator, or the like. Therefore, the definition should be based on the contents throughout this specification.

1 is a functional block diagram of an image classification apparatus for searching personal information images according to an embodiment of the present invention.

1, an image classification apparatus 200 for searching personal information according to a temporal example of the present invention includes a first image classification unit 202, an image production unit 204, a second image classification unit 206, .

First, the first image classifying unit 202 receives a plurality of images to be optically recognized for personal information extraction from the image storing unit 230. At this time, the plurality of images may mean a photograph taken through a smart phone, a digital camera, etc., and the personal information may be text information, for example, as personal information of the user included in the image . The text information may be, for example, a passport number included in the passport photo, a resident registration number included in the photograph of the resident ID card, transaction information included in the photograph of the receipt or various documents, evidence information, and the like , But is not limited thereto.

That is, when the above-described plurality of images are input, the first image classifying unit 202 generates a first image group including a plurality of images, and sets each image included in the first image group according to a preset reference And classifies the image to be removed with low probability of existence of personal information. Next, the first image classifying unit 202 generates a second image group including the remaining images except for the removal target image classified as above in the first image group.

In this case, the removal target image refers to an image that does not include the personal information exemplified above. For a specific method of classifying the image that does not include such personal information, the detailed block configuration of the first image classifying unit 202 Will be described later in detail with reference to FIG.

The image producing unit 204 selects a predetermined number of images for the images included in the second image group generated from the first image classifying unit 202 and combines them into one image. At this time, for example, the image production unit 204 may synthesize six images included in the second image group into one image, but the present invention is not limited thereto.

In addition, the image production unit 204 divides one image 300 into six regions and combines the six images selected in the second group into three regions, as shown in FIG. 3, in synthesizing the six images into one image. It is possible to combine the reduced image into a single image by arranging the reduced image in each area on one image 300 after reducing it to a certain ratio according to the size of the area. The above-described image combining operation will be described later in detail with reference to FIG. 5 showing the detailed block configuration of the image producing unit 204. FIG.

The secondary image classifying unit 206 classifies a plurality of images included in one image as a result of optical recognition of one image synthesized to include a plurality of images generated by the image producing unit 204, The image in which the personal information is detected and the image in which the personal information is not detected are classified. In this case, the secondary image classification unit 206 classifies the images for which no personal information is detected as a general image, and manages them as a general image group. The image in which the personal information is detected is classified as a personal information image, But it is not limited thereto.

In this case, optical recognition refers to a technique for extracting personal information from an image when personal information such as text is included in the image. Personal information in the image can be analyzed through optical recognition, The location information in which the corresponding personal information is detected in the information-based image can also be analyzed.

The optical recognition may also be applied to an application for performing optical recognition in a computing device such as a smartphone, a tablet, a notebook, and a desktop PC equipped with the image classification apparatus 200 according to an embodiment of the present invention The result of the optical recognition may be provided to the image classification apparatus 200 or may be provided to the image classification apparatus 200 after being executed in an external optical recognition server or the like connected through a wired or wireless communication network .

4 is a functional block diagram of a first image classifying unit according to an embodiment of the present invention.

Referring to FIG. 4, the first image classifier 202 according to an embodiment of the present invention includes a resolution-based classifier 400, a metadata-based classifier 402, a continuous picture classifier 404, A classification unit 406, and the like.

The resolution-based classification unit 400 examines the resolution of each image in the first group, and classifies the image as a removal target image when the resolution is lower than a preset reference resolution. At this time, the reference resolution can be set to a resolution that appears in an image taken by the user with a relatively low resolution front camera, for example, in the case of a smart phone, and accordingly, But the present invention is not limited thereto. That is, the reference resolution can be selectively changed depending on which image is to be classified as the image to be removed.

The metadata-based classification unit 402 may identify the metadata information about each image in the first group and classify the image as a removal target image having low possibility of existence of personal information when information related to copyright is included. At this time, the image including the copyright-related information may mean, for example, an image generated by a third party other than the user himself. Such an image may be, for example, an image produced by a corporation or the like, And the like. Such an image may include identification information of a vendor or a photographer who created the image in the metadata information included in the image data, and may include information such as the model name of the camera that photographed the image However, the copyright related information may mean such information, but is not limited thereto.

The continuous picture classifying unit 404 classifies the images generated from the same camera for all the images in the first group, and confirms the generated time information for each of the images generated from the same camera. Then, And the corresponding image within the predetermined time interval range is classified as the removal target image. At this time, the predetermined reference time interval range may be set to a time interval set in a continuous photographing mode, for example, a smart phone, a digital camera, etc., It can mean that it has been created. In addition, since the image generated by continuous photographing is likely to contain similar information, when there are a plurality of images generated by continuous photographing, one image can be excluded and the remaining images can be classified as a control target image The focus-based classifier 406 examines the degree of focus blur for each image in the first group, and when the degree of focus mismatch is equal to or greater than the predetermined reference mismatch degree, Classify. At this time, the predetermined reference focal point mismatch degree may indicate that the degree of the focal mismatch is expressed as a percentage when an image photographed through a smart phone, a digital camera, or the like is photographed out of focus and an image mismatch occurs in the image . In this case, the degree of the reference focus discrepancy can be determined as an experimental value capable of extracting the personal information from the image despite the focus discrepancy, for example, when 0 indicates that the focus is correctly focused and 100 indicates that the focus is completely discordant have. That is, the degree of focus discrepancy can be set to, for example, 0 to 20, and the image having the degree of focus discrepancy of 20 or more can be classified as an image to be removed, which is difficult to extract personal information.

5 is a functional block diagram of an image producing unit according to an embodiment of the present invention.

5, an image producing unit 204 according to an embodiment of the present invention includes a resolution checking unit 500, an image reducing unit 502, an image combining unit 504, an image position storing unit 506, and the like can do.

The resolution checking unit 500 checks the resolution of each image in the second image group, which is composed of images with high possibility of existence of personal information except for the primary classified image, (502).

At this time, the images included in the second image group include images having a predetermined reference resolution or higher through the primary classification, but the resolutions may be different from each other. Since the image included in the second image group must be reduced to a predetermined size based on the resolution of the original image through the image shrinking unit 502, it is necessary to check the resolution of each image.

The image reduction unit 502 reduces each image included in the second group to a predetermined ratio based on the resolution of each image. Each image included in the second group may be an image to be subjected to optical recognition in a state classified as an image in which personal information is likely to exist through the primary classification. When optical recognition is performed for each of these images, the cost for optical recognition may increase relatively, and data transmission efficiency may be lowered when the optical recognition server is transmitted to an external optical recognition server. Therefore, in order to increase the optical recognition efficiency, in one embodiment of the present invention, a plurality of reduced images are synthesized to be one image, thereby improving the optical recognition efficiency.

In other words, when the number of reduced images to be included in one image is determined, the image reducing unit 502 determines the original image according to the size of the corresponding area in consideration of the size of the area occupied by each image on one image according to the number of images .

The image synthesizing unit 504 selects a predetermined number of images to be synthesized into one image out of the reduced images at a predetermined ratio through the image reducing unit 502 and separates the selected images into a predetermined number of images And then synthesized into one image.

The image position storage unit 506 stores identification information of each of a plurality of reduced images included in one image synthesized by the image synthesis unit 504 and information on an area located on the corresponding image.

At this time, one image having been subjected to image synthesis as described above may be optically recognized through an optical recognition device provided in the image classification device 200 or an optical recognition server 250 connected through a wired / wireless communication network, The second image classifier 206 may be provided with an optical recognition result that includes the personal information detected in the first image classifier 206 and the corresponding image position information in which the individual information is detected.

When the second image classifier 206 receives the personal information detected in one image and the position information in which the personal information is detected in one image by performing the optical recognition as described above, Each image in which individual information is detected on one image is discriminated using stored information, and an image having personal information and an image having no personal information are classified.

In this case, the second image classifying unit 206 may classify an image for which no personal information is detected as a general image, for example, as a general image group, as described above, And can be managed as a personal information image group.

6 shows an operation control flow for classifying an image in an image classification apparatus according to an embodiment of the present invention. Hereinafter, an operation according to an embodiment of the present invention will be described in detail with reference to FIG. 2 to FIG.

First, the image classification apparatus 200 receives a plurality of images to be optically recognized for personal information extraction from the image storage unit 230 (S600). At this time, the image may be a photograph taken through a smart phone, a digital camera, or the like, and the personal information may be text information, for example, as personal information of the user included in the image.

When the plurality of images are input as described above, the image classification apparatus 200 generates a first image group including a plurality of images, and examines each image included in the first image group according to a preset reference A primary classification for classifying the image to be removed having a low probability of existence of personal information is performed (S602).

At this time, the image classifying apparatus 200 checks the resolution of each image in the first group in the first classification, and classifies the image as a removal target image if the resolution is lower than the preset reference resolution (S602-1). At this time, the reference resolution can be set to a resolution that appears in an image taken by the user with a relatively low resolution front camera, for example, in the case of a smart phone, and accordingly, It can be classified as a removal object image in which the possibility that information exists is low.

In addition, the image classification apparatus 200 checks the metadata information about each image in the first group and classifies the image as a removal target image if information related to copyright is included (S602-2). At this time, the image including the copyright-related information may mean, for example, an image generated by a third party other than the user himself. Such an image may be, for example, an image produced by a corporation or the like, And the like.

In addition, the image classifying apparatus 200 classifies the images photographed by the same camera for all the images in the first group, then confirms the generated time information of each image photographed by the same camera, (S602-3) except that one image is arbitrarily selected from a plurality of images included in the predetermined time interval range. At this time, the predetermined reference time interval range may be set to a time interval set in a continuous photographing mode, for example, a smart phone, a digital camera, etc., It can mean that it has been created.

In addition, the image classification apparatus 200 checks the degree of focus blur for each image in the first group, and classifies the image as a removal target image when the degree of focus mismatch is equal to or greater than a predetermined reference mismatch degree (S602-4) . At this time, the predetermined reference focal point mismatch degree may indicate that the degree of the focal mismatch is expressed as a percentage when an image photographed through a smart phone, a digital camera, or the like is photographed out of focus and an image mismatch occurs in the image .

As described above, the image classifying device 200 eliminates the removal target image (photo A) in which the personal information in the first image group is unlikely to exist in the first image group (S604), collects the remaining images excluding the removal target image And generates a second image group (photo B) (S606).

Next, the image classification apparatus 200 selects a predetermined number of images for the images included in the second image group and synthesizes them into one image (S608). At this time, for example, the image classification apparatus 200 may synthesize six images included in the second image group into one image, but the present invention is not limited thereto. 3, the image classifying apparatus 200 divides one image 300 into six regions and combines the six images selected in the second group into six regions, One image can be synthesized by reducing the image to a size corresponding to the size of each area and then placing the reduced image on each area on one image 300.

One image synthesized so as to include a plurality of images classified in the above-described primary classification is optically recognized by an optical recognition apparatus mounted in the system or an external optical recognition server 250 connected through a wire / wireless communication network, An optical recognition in which personal information such as text is extracted is performed, and the optical recognition result may be input to the image classification apparatus 200 (S610).

If the result of the optical recognition is input as described above, the image classification apparatus 200 may extract an image in which the personal information is detected and an image in which the personal information is not detected, from a plurality of images included in one image based on the optical recognition result (S612). ≪ / RTI >

FIG. 7 shows an area where personal information is recognized through optical recognition in one image in which a plurality of images are synthesized.

7, the image classification apparatus 200 generates an image in which the personal information is detected in one image in which a plurality of images are synthesized based on the personal information input as a result of optical recognition and the position information in which the personal information is detected, And the second classification is performed.

Then, the image classification apparatus 200 classifies the images for which no personal information is detected as a general image and manages them as a general image group (S614). The image in which the personal information is detected is classified as a personal information image And may be managed as a personal information image group (S614).

Accordingly, it is possible to greatly reduce the number of images to be optically recognized for personal information search by removing images with low probability of existence of personal information by classifying the images to be optically recognized according to predetermined classification criteria, And the cost can be reduced.

8 is a diagram illustrating an example of optical recognition time and cost results after image classification according to an embodiment of the present invention.

As shown in FIG. 8, in the case of the image classification result according to an embodiment of the present invention, for example, in the case of the user A, 1000 optical recognition target images are reduced to about 100 through the first classification and image synthesis, It can be confirmed that the time required is greatly reduced, and that the same effect is also obtained in the case of users B, C, and D.

In this case, the calculation of the time and cost required for the above-described optical recognition is merely an example of using the Google Cloud Vision API as an optical recognition server for the convenience of explanation, It will be understood by those skilled in the art that the same result will be obtained even when the module is mounted.

Combinations of the steps of each flowchart attached to the present invention may be performed by computer program instructions. These computer program instructions may be loaded into a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus so that the instructions, which are executed via a processor of a computer or other programmable data processing apparatus, Lt; / RTI > These computer program instructions may also be stored in a computer usable or computer readable memory capable of directing a computer or other programmable data processing apparatus to implement the functionality in a particular manner so that the computer usable or computer readable memory It is also possible to produce manufacturing items that contain instruction means for performing the functions described in each step of the flowchart. Computer program instructions may also be stored on a computer or other programmable data processing equipment so that a series of operating steps may be performed on a computer or other programmable data processing equipment to create a computer- It is also possible for the instructions to perform the processing equipment to provide steps for executing the functions described in each step of the flowchart.

In addition, each step may represent a module, segment, or portion of code that includes one or more executable instructions for executing the specified logical function (s). It should also be noted that in some alternative embodiments, the functions mentioned in the steps may occur out of order. For example, the two steps shown in succession may in fact be performed substantially concurrently, or the steps may sometimes be performed in reverse order according to the corresponding function.

While the invention has been shown and described with reference to certain preferred embodiments thereof, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention. Accordingly, the scope of the invention should not be limited by the described embodiments but should be defined by the appended claims.

200: image classifying apparatus 202: first image classifying unit
204: image production unit 206: second image classification unit
230: Image storage unit 250: Optical recognition server
400: resolution-based classification unit 402: metadata-based classification unit
404: Continuous picture classifying unit 406: Focus-based classifying unit
500: resolution checker 502:
504: image synthesis unit 506: image position storage unit

Claims (17)

An image classifying unit for classifying an image to be removed by checking all the images in the first image group including a plurality of images according to a preset reference and generating a second image group including the remaining images excluding the removal target image, ,
And an image producing unit for selecting a predetermined number of images among the images included in the second image group and synthesizing them into one image
Image classification device.
The method according to claim 1,
Wherein the image classifier comprises:
And a resolution-based classifying unit for classifying the image into the removal object image if the resolution is equal to or less than a preset reference resolution,
Image classification device.
The method according to claim 1,
Wherein the image classifier comprises:
And a metadata-based classifying unit for classifying the image into the removal object image by checking metadata information of each image in the first group and if information related to copyright is included
Image classification device.
The method according to claim 1,
Wherein the image classifier comprises:
Wherein the image processing unit classifies images taken by the same camera for each image in the first group and generates a plurality of images including a generation time point within a predetermined time interval range based on generation time information of each image taken by the same camera Further comprising a consecutive photo classifying unit for classifying the remaining images into the removal subject image except one image selected at random
Image classification device.
The method according to claim 1,
Wherein the image classifier comprises:
Further comprising a focus-based classifying unit for classifying each image in the first group as the removal object image, and if the degree of focus mismatch is greater than or equal to a predetermined reference mismatch degree,
Image classification device.
The method according to claim 1,
The image producing unit may include:
A resolution checker for checking the resolution of each image included in the second image group;
An image reduction unit for reducing each image included in the second image group to a predetermined ratio based on the examined resolution,
And an image combining unit for selecting the predetermined number of images from among the reduced images, and compositing the reduced image into the one image by positioning the reduced image in an area divided according to the predetermined number on the one image
Image classification device.
The method according to claim 6,
The image producing unit may include:
And an image position storage unit for storing identification information for each of the reduced images included in the one image and area information for each of the reduced images located on the one image,
Image classification device.
The method according to claim 1,
The apparatus comprises:
And an image reclassifier for classifying an image in which the personal information exists and an image in which the personal information does not exist, based on the optical recognition result of the one image, for the plurality of images included in the one image, More included
9. The method of claim 8,
The personal information includes:
The text information included in the one image
Image classification device.
Classifying each image in the first image group including a plurality of images according to a predetermined criterion to classify the image to be removed,
Generating a second image group including the remaining images excluding the removal target image,
Selecting a predetermined number of images from among the images included in the second image group to produce one image;
Image classification method.
11. The method of claim 10,
Wherein said classifying comprises:
Checking the resolution for each image in the first group,
And classifying the image as the removal object image if the resolution is less than or equal to a preset reference resolution
Image classification method.
11. The method of claim 10,
Wherein said classifying comprises:
Identifying metadata information for each image in the first group;
And classifying the metadata information into the removal object image if information related to copyright is included in the metadata information
Image classification method.
11. The method of claim 10,
Wherein said classifying comprises:
Classifying images taken by the same camera for each image in the first group;
Confirming the generated time information of each image photographed by the same camera;
And classifying the remaining images into the removal object image, excluding one image selected arbitrarily from a plurality of images in which the generation time of the image is included in the predetermined time interval range based on the time information
Image classification method.
11. The method of claim 10,
Wherein said classifying comprises:
Checking the degree of focus mismatch for each image in the first group,
And classifying the image as the removal object image if the degree of focus mismatch is greater than or equal to a preset reference mismatching degree information
Image classification method.
11. The method of claim 10,
Wherein the step of fabricating comprises:
Inspecting the resolution of each image included in the second image group;
Reducing each image included in the second image group to a predetermined ratio based on the checked resolution;
Selecting the predetermined number of images from among the reduced images, and compositing the reduced image into the one image by locating the reduced image in an area divided according to the predetermined number on the one image
Image classification method.
16. The method of claim 15,
Wherein the step of fabricating comprises:
Further comprising storing identification information for each of a plurality of reduced images included in the one image and region information on each of the reduced images,
Image classification method.
11. The method of claim 10,
The method comprises:
Further comprising classifying an image in which the personal information exists and an image in which the personal information does not exist based on the optical recognition result for the one image with respect to the plurality of images included in the one image
Image classification method.

Images