KR101378295B1 - Method and Apparatus of privacy masking on Image - Google Patents

Abstract

Provided are a method and an apparatus for privacy masking of an image. A privacy masking method of an image according to an exemplary embodiment of the present invention may include detecting a privacy region requiring privacy protection in an image and performing masking on the image. have. Therefore, it is possible to effectively and easily encrypt / scramble and decrypt / descramble the important area (human face area, vehicle license plate area, house window area, etc.) detected from the image input through the CCTV camera. . In addition, even if a separate metadata file is not used, location consistency of important areas detected before and after scramble can be maintained. In addition, it is possible to apply a scrambling algorithm of the same key value regardless of the size of the important region. As a result, the CCTV-based video security system can effectively contribute to increase the efficiency of the system through the protection of personal information more effectively.

Description

Method and device for privacy masking of images {Method and Apparatus of privacy masking on Image}

The present invention relates to a privacy masking method and apparatus for an image, and more particularly, to a privacy masking method and apparatus for privacy masking while maintaining the characteristics of the image.

The present invention is derived from a study conducted as part of the IT growth engine technology development project of the Ministry of Knowledge Economy [Task management number: 2009-S-040-01, Title: Development of image security technology for personal security security].

Currently, various methods have been developed to protect personal information such as face in CCTV environment, but mainly masking or encrypting the detected object area by manually designating specific area or using existing object detection method Methods are being used.

These conventional methods have a problem that one-way encryption is possible, but the decryption of the encryption area is not performed or separate metadata is maintained for decryption.

Therefore, there is a need for the development of a technology that enables easier and effective redetection and decryption of important areas even after encryption or scrambling on important areas such as detected faces, license plates and windows.

An object of the present invention is to provide a privacy masking method of an image that can be privacy masked while maintaining the characteristics of the image.

Another object of the present invention is to provide a privacy masking apparatus for an image which can be privacy masked while maintaining the characteristics of the image.

Privacy masking method for an image to achieve the above object of the present invention comprises the steps of detecting a privacy area that requires privacy protection in the image (Image); Masking the image; Detecting a masked privacy area in the masked image; And performing unmasking on the masked image, wherein detecting the masked privacy region comprises: color distribution information of the masked privacy region and the color distribution information of the masked privacy region; The masked privacy area may be detected using the sameness between the color distribution information of the privacy area.

The detecting of the privacy area may include detecting the privacy area using at least one of edge information, shape information, color distribution information, and learning information of the privacy area.

The masking of the image may be performed by scrambling the privacy region of the image using a scrambling key.

delete

The detecting of the masked privacy region may further include detecting at least one of edge information of the masked image, shape information, and learning information of the masked privacy region.

delete

The detecting of the masked privacy area may be performed by further using an indication of the masked privacy area.

The detecting of the masked privacy region may be performed by further using a sub-image storing the masked privacy region.

The detecting of the masked privacy area may be performed by further using coordinate data of the masked privacy area.

Here, detecting by using the coordinate data of the masked privacy area may be detected by adding and preserving the coordinate data in an image in the form of a watermark.

The unmasking of the masked image may be performed by descrambling the masked privacy region by using the scrambling key in reverse.

Privacy masking apparatus for an image to achieve the above object of the present invention comprises a masking processing unit for detecting a privacy area that requires privacy protection in the image, and masking the image; A storage unit for storing the image and the masked image; An unmasking processor that detects a masked privacy area from the masked image and performs unmasking on the masked image; And a controller for controlling the masking processor, the storage unit, and the unmasking processor, wherein the unmasking processor includes color distribution information of the masked privacy area and color of the privacy area; The masked privacy region may be detected using the sameness between distribution information.

According to the privacy masking method and apparatus of the image as described above, encryption / scramble and decrypt / for the important area (human face area, license plate area of the vehicle, window area of the house, etc.) detected from the image input through the CCTV camera Describing can be done effectively and easily. In addition, even if a separate metadata file is not used, location consistency of important areas detected before and after scramble can be maintained. In addition, it is possible to apply a scrambling algorithm of the same key value regardless of the size of the important region. As a result, the CCTV-based video security system can effectively contribute to increase the efficiency of the system through the protection of personal information more effectively.

1 is a flowchart illustrating a privacy masking method of an image according to an embodiment of the present invention.
FIG. 2 is an exemplary diagram for describing detection and masking of a privacy region in an image in the privacy masking method of an image according to an embodiment of the present invention.
FIG. 3 is an exemplary diagram for describing detection, masking, detection of masked areas, and unmasking of privacy areas in an image in a privacy masking method of an image according to an embodiment of the present invention.
FIG. 4 is an exemplary diagram for describing the identity of a privacy area of a face of a person and color distribution information of a masked privacy area in an image privacy masking method according to an embodiment of the present invention.
5 is an exemplary view for explaining the identity of the privacy distribution for the license plate and the color distribution information for the masked privacy region in the image privacy masking method according to an embodiment of the present invention.
6 is a block diagram illustrating an apparatus for masking privacy of an image according to an exemplary embodiment of the present invention.

While the invention is susceptible to various modifications and alternative forms, specific embodiments thereof are shown by way of example in the drawings and will herein be described in detail.

It should be understood, however, that the invention is not intended to be limited to the particular embodiments, but includes all modifications, equivalents, and alternatives falling within the spirit and scope of the invention.

The terms first, second, etc. may be used to describe various components, but the components should not be limited by the terms. The terms are used only for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be referred to as a second component, and similarly, the second component may also be referred to as a first component. And / or < / RTI > includes any combination of a plurality of related listed items or any of a plurality of related listed items.

When a component is referred to as being "connected" or "connected" to another component, it may be directly connected to or connected to that other component, but it may be understood that other components may be present in between. Should be. On the other hand, when an element is referred to as being "directly connected" or "directly connected" to another element, it should be understood that there are no other elements in between.

The terminology used in this application is used only to describe a specific embodiment and is not intended to limit the invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. In this application, the terms "comprise" or "having" are intended to indicate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, and one or more other features. It is to be understood that the present invention does not exclude the possibility of the presence or the addition of numbers, steps, operations, components, components, or a combination thereof.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the relevant art and are to be interpreted in an ideal or overly formal sense unless explicitly defined in the present application Do not.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. In order to facilitate the understanding of the present invention, the same reference numerals are used for the same constituent elements in the drawings and redundant explanations for the same constituent elements are omitted.

1 is a flowchart illustrating a privacy masking method of an image according to an embodiment of the present invention.

Referring to FIG. 1, a method for masking privacy of an image according to an embodiment of the present invention may include detecting a privacy region requiring privacy in an image (step 110); And performing masking on the image (step 120).

In addition, the privacy masking method of the image according to an embodiment of the present invention comprises the steps of detecting a masked privacy area in the masked image (step 130); And performing unmasking on the masked image (step 140).

First, the detecting of the privacy area (step 110) may be to detect an area requiring privacy protection in the image.

That is, when the image includes a face plate for identifying a person and a license plate for identifying a car, the face and license plate areas may be detected for privacy masking.

The detecting of the privacy area may include detecting the privacy area using at least one of edge information, shape information, color distribution information, and learning information of the privacy area.

That is, the edge information may be obtained through analysis of the image, and the boundary between the face and the non-face portion may be detected using the edge information, and the shape of the analysis result and the face shape of the image, etc. The information may be compared to determine whether a face exists in the image. In addition, since the color distribution information may be compared with the color distribution information unique to the face, the face area may be detected more accurately. In addition, the face area may be detected using the learning information of the privacy area including the Adaboost learning technique.

Next, masking the image may be performed by performing scrambling on the privacy area of the image using a scrambling key.

(여기에서,

는 스크램블링 이전의 영상,

는 스크램블링 키,

는 스크램블링 이후의 영상)를 적용하는 것일 수 있다.
For example, the scrambling is

(From here,

Is the video before scrambling,

Scrambling key,

May be an image after scrambling).

FIG. 2 is an exemplary diagram for describing detection and masking of a privacy region in an image in the privacy masking method of an image according to an embodiment of the present invention.

Referring to FIG. 1 and FIG. 2, it can be seen that FIG. 2 (a) detects a face region, which is an area in which a person can be identified, from the image by using a general image such as a CCTV image.

It can be seen that the part represented by the rectangle in FIG. 2 (a) is detected as a face area in the image, and this means that at least one of edge information, shape information, color distribution information, and learning information of the privacy area of the image is detected. It will be appreciated that the detection has been made using.

In addition, FIG. 2 (b) shows masking of the portion represented by the rectangle in FIG. 2 (a). That is, masking is performed using a scrambling method on the privacy area, which is a human face area. Therefore, even if the image is used to mask the image so that it is not possible to identify who.

Next, the detecting of the masked privacy region in the masked image (step 130) may include using at least one of edge information, shape information, color distribution information, and learning information of the masked privacy region of the masked image. May be to detect.

As described above, since the masked privacy region of the masked image is distinguished from other regions, the distinguished region can be detected using edge information of the masked image, and the shape information and color can be detected. The distribution information may be used to detect a face of a person and a license plate of a vehicle. In addition, the face area may be detected using the learning information of the masked privacy area including the Adaboost learning technique.

In addition, the detecting of the masked privacy region from the masked image (step 130) may further include using at least one of edge information, shape information, color distribution information, and learning information of the masked privacy region. The following method may be used.

The detecting of the masked privacy area may be performed by using an indication of the masked privacy area. That is, the masked privacy area may be detected by using a method of displaying the masked privacy area in a shape of a specific color.

The detecting of the masked privacy area may be performed by using a sub-image storing the masked privacy area.

That is, the privacy zone may be detected by using a special image format (for example, an indexed image format) to store the masked privacy region and storing the masked privacy region in a separate sub-image.

The detecting of the masked privacy area may be performed by using coordinate data of the masked privacy area. In particular, detecting using the coordinate data of the masked privacy area may be detected by adding and preserving the coordinate data in an image in the form of a watermark.

That is, some coordinates indicating the masked privacy area may be found, and the detected coordinate data may be additionally inserted into the image in the form of a watermark and detected.

Next, the step of performing unmasking on the masked image (step 140) may be descrambling on the masked privacy area by using the scrambling key in reverse.

FIG. 3 is an exemplary diagram for describing detection, masking, detection of masked areas, and unmasking of privacy areas in an image in a privacy masking method of an image according to an embodiment of the present invention.

Referring to FIGS. 1 and 3 in parallel, FIG. 3 (a) illustrates detecting a face region, which is a region in which a person can be identified from the image. FIG. 3 (b) shows masking of the part represented by the rectangle in FIG. 3 (a).

In addition, FIG. 3 (c) illustrates detecting the masked privacy area in the masked image. It can be seen that the portion represented by the rectangle in FIG. 3 (c) is detected as the face region in the image.

This is because the masked privacy region of the masked image is distinguished from other regions, so that the masked privacy region is detected using edge information of the masked image, and the shape information, color distribution information, or the masked image is detected. It may be determined that the face is a human face using the learning information of the privacy area.

In particular, in order to detect the masked privacy region in the masked image, the sameity between the color distribution information of the masked privacy region and the color distribution information of the original privacy region before the masking is maintained. Can be used. The identity to both color distribution information is described in detail with reference to Figs.

In addition, FIG. 3D illustrates a result of unmasking the masked privacy region in the masked image. This may be obtained by applying the same scrambling key as when performing scrambling on the masked privacy region and applying the scrambling operation in reverse.

FIG. 4 is an exemplary diagram for describing the identity of a privacy area of a face of a person and color distribution information of a masked privacy area in an image privacy masking method according to an embodiment of the present invention.

Referring to FIG. 4, FIG. 4A illustrates a face area detected as a privacy area, and FIG. 4B illustrates a result of masking a face area detected as the privacy area.

4 (c) and 4 (d) show masking of the original face area detected as the privacy area as shown in FIG. 4 (a) and the face area detected as the privacy area as shown in FIG. 4 (b). For each color density is represented.

4 (e) and 4 (f) respectively mask the original face area detected as the privacy area as shown in FIG. 4 (a) and the face area detected as the privacy area as shown in FIG. 4 (b). For each color distribution.

As a result, it can be seen that color distribution information including color density and color distribution of the masked privacy area of the masked image is the same as the color distribution information of the privacy area.

That is, it can be seen that the color distribution information of the face region of the original, which is a privacy region, among the images, has the same characteristics as the color distribution information of the masked face region among the masked images.

5 is an exemplary view for explaining the identity of the privacy distribution for the license plate and the color distribution information for the masked privacy region in the image privacy masking method according to an embodiment of the present invention.

Referring to FIG. 5, FIG. 5 (a) shows a license plate area detected as a privacy area, and FIG. 5 (b) shows a result of masking a license plate area detected as the privacy area.

5 (c) and 5 (d) respectively mask an original license plate area detected as a privacy area as shown in FIG. 5 (a) and a license plate area detected as a privacy area as shown in FIG. 5 (b). Each result represents a color density.

5 (e) and 5 (f) respectively mask an original license plate area detected as a privacy area as shown in FIG. 5 (a) and a license plate area detected as a privacy area as shown in FIG. 5 (b). Each result represents a color distribution.

As in the above-described results, it can be seen that color distribution information including the color density and the color distribution of the masked privacy area of the masked image is the same as the color distribution information of the privacy area.

That is, it can be seen that the color distribution information of the original license plate area, which is a privacy area, among the images, has the same characteristics as the color distribution information of the scrambled license plate area of the masked image.

These results indicate that the use of color distribution information and edge information is more effective in obtaining the same area detection result for the privacy area.

6 is a block diagram illustrating an apparatus for masking privacy of an image according to an exemplary embodiment of the present invention.

Referring to FIG. 6, the privacy masking apparatus 600 for an image according to an embodiment of the present invention may include a masking processor 610 for masking an image; A storage unit 620 for storing the image and the masked image; An unmasking processor 630 which performs unmasking on the masked image; And a controller 640 for controlling the masking processor, the storage unit, and the unmasking processor.

First, the masking processor 610 may perform masking on an image, and may require privacy protection using at least one of edge information, shape information, color distribution information, and learning information of the privacy area. The area may be detected, and masking may be performed on the detected privacy area.

In addition, the masking processing unit 610 may perform the masking on the image by performing scrambling on the privacy area using a scrambling key.

Next, the storage unit 620 may serve as a space for storing the image and the masked image.

Next, the unmasking processor 630 may perform unmasking on the masked image. This may be to detect the masked privacy area from the masked image and perform unmasking on the masked image.

The masking privacy area may be detected by the unmasking processor 630 using at least one of edge information, shape information, color distribution information, and learning information of the masked privacy area of the masked image. have.

Here, the color distribution information of the masked privacy area in the masked image may be the same as the color distribution information of the privacy area. That is, the color distribution information of the original image of the privacy area will have the same characteristics as the color distribution information of the scrambled privacy area after scrambling.

Additionally, the unmasking of the masked image may be performed by the unmasking processor 630 to descramble the masked privacy region by applying a scrambling key to the masked privacy region.

Next, the controller 640 may control the masking processor, the storage unit, and the unmasking processor.

It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined in the appended claims. It will be possible.

600: image privacy masking device
610: masking processing unit
620:
630: unmasking processing unit
640: control unit

Claims (12)

Detecting a privacy area requiring privacy protection in an image;
Masking the image;
Detecting a masked privacy area in the masked image; And
And performing unmasking on the masked image.
The detecting of the masked privacy area may include detecting the masked privacy area by using the sameness between the color distribution information of the masked privacy area and the color distribution information of the privacy area among the masked images. The privacy masking method of an image. The method of claim 1,
The detecting of the privacy area may include detecting the privacy area using at least one of edge information, shape information, color distribution information, and learning information of the privacy area. The method of claim 1,
The masking of the image may include: scrambling the privacy area of the image using a scrambling key. delete The method of claim 1,
The detecting of the masked privacy region may further include detecting at least one of edge information of the masked image, shape information, and learning information of the masked privacy region. delete The method of claim 1,
The detecting of the masked privacy area may further include detecting the masked privacy area by using an indication of the masked privacy area. The method of claim 1,
The detecting of the masked privacy area may further include detecting the masked privacy area by further using a sub-image storing the masked privacy area. The method of claim 1,
The detecting of the masked privacy area may further include detecting coordinate data of the masked privacy area by using coordinate data of the masked privacy area. 10. The method of claim 9,
Detecting using the coordinate data of the masked privacy area is the privacy masking method of the image, characterized in that for detecting and adding the coordinate data in the form of a watermark in the image. The method of claim 3,
The unmasking of the masked image may include descrambling of the masked privacy region by using the scrambling key in reverse. A masking processor configured to detect a privacy area requiring privacy protection from the image and to mask the image;
A storage unit for storing the image and the masked image;
An unmasking processor that detects a masked privacy area from the masked image and performs unmasking on the masked image; And
And a controller for controlling the masking processor, the storage unit, and the unmasking processor.
The unmasking processor detects the masked privacy region by using the sameness between the color distribution information of the masked privacy region and the color distribution information of the privacy region among the masked images. Masking device.

Images