KR102066688B1 - How to protect personal information in video using independent password for each object - Google Patents

Abstract

The present invention relates to a method for protecting personal information in an image using an independent password for each object, and more specifically, for each object performing masking and unmasking processing based on a security key for each object by tracking and classifying objects in the image. The present invention relates to a method for protecting personal information in an image using an independent password.

Description

How to protect personal information in video using independent password for each object}

The present invention relates to a method for protecting personal information in an image using an independent password for each object, and more specifically, for each object performing masking and unmasking processing based on a security key for each object by tracking and classifying objects in the image. The present invention relates to a method for protecting personal information in an image using an independent password.

When distributing CCTV images, they provide masking (blocking them with mosaic or other colors so that they cannot be recognized) to protect personal information in the images. After detecting and tracking the object, it grasps the object information on the screen, encrypts the area of the object by using the key in the encoding process, and transfers the object information into the image. On the decoder side, images are displayed that contain the decoded, encrypted, and masked objects. In this case, the objects encrypted for masking cannot be distinguished as what objects are mixed with video blocks. Using the information of the surrounding images except the masked object information, the image section that needs to be identified is identified. Then, after receiving the key and decoding it, the object that was indistinguishable from the mosaic is seen in the form of the original image. In this case, however, since all objects are decoded and restored to the original, all personal information is inevitably exposed.

As described above, the technology for protecting personal information on the surveillance video has been used to mask / unmask objects using a cryptographic key, but in this case, the problem of masking / unmasking all objects simultaneously occurs. In other words, there is a problem in that the personal information of another object needs to be exposed unnecessarily only by unmasking the required object.

In addition, when masking each object with a separate encryption key, it is inconvenient to require the keys of all objects when unmasking the whole. If you have the ability to unmask the whole group with one key, and individual objects can be unmasked with one key, you can unmask all with one key that unmasks the whole. By reducing the risk of over-exposing the information, it is possible to selectively check only the personal information that is needed, which may be the most desirable personal information protection video system. However, even with the introduction of such a convenient system, it should not be able to infer or calculate the group master key using the keys of individual objects.

An object of the present invention is to provide a method for protecting personal information in an image using an independent password for each object that performs masking and unmasking processing based on a security key for each object through tracking and classifying objects in the image.

In order to solve the above problems, the present invention provides an image masking processing method, comprising: an object detecting step of detecting an object in an image; An object tracking step of confirming whether the detected object is the same object as the existing detection object; An object classification step of classifying the type of the detected object; A masking processing step of performing a masking process on the detected object based on an object-specific security key; And a masking information insertion step of inserting information of the masked object. It provides, the masking processing method of the image.

In the present invention, the masking processing step, the region of interest derivation step of deriving a region of interest to be masked for each object; Generating a security key for the ROI; And encrypting the ROI using a security key. It may include.

In the present invention, the first security key generation step, the master key generation step of generating a master key based on the reference password and the predetermined unique information; And an individual key generation step of generating an individual key based on the master key and the individual information. Includes, the security key may include the master key and the individual key.

In the present invention, the individual key generation step, the group key generation step of generating a group key based on the master key and the group attribute information; And an object key generation step of generating an object key based on the group key and object property information. The private key may include the group key and the object key.

In the present disclosure, the encrypting may be performed by changing or scrambled macroblocks in the ROI according to a predetermined rule based on the object key.

In the present invention, the master key is generated by converting the reference password and the predetermined unique information through a hash function, the group key is generated by converting the master key and the group attribute information through a hash function, the object The key may be generated by converting the group key and the object attribute information through a hash function.

In the present invention, the information of the masked object, the total number of objects included in the image; The total number of groups contained in the image; The number of objects contained in each of the groups; Masking information including object identification information and object region information of each of the objects; It may include.

In the present invention, the object region information, the X-Y coordinate of the region of interest for the object; A width of the region of interest; A height of the region of interest; And a radius of the region of interest; It may include any one or more of.

In the present invention, the masking processing method of the image further includes an unmasking processing step, wherein the unmasking processing step comprises: a masking information extraction step of extracting information of a masked object; An unmasking information input step of receiving information on at least one object to be unmasked and a security key; An ROI extraction step of extracting an ROI from information of an object to be unmasked; And a decryption step of performing unmasking on the ROI using a security key.

In the present invention, the unmasking processing of the image may include: a second security key generation step of generating an object key from the master key or the group key when a master key or a group key is input in the unmasking information input step; The decoding step may further include unmasking using the object key.

In the present invention, when the information of the plurality of unmasked objects and the plurality of security keys are input in the unmasking information input step, the ROI extracting step may include a plurality of ROIs from the information of the plurality of unmasked objects. In the extracting and decrypting operation, unmasking may be performed using the plurality of security keys respectively corresponding to the plurality of ROIs.

In order to solve the above problems, the present invention provides an image masking processing method, comprising: an object detecting step of detecting an object in an image; An object tracking step of confirming whether the detected object is the same object as the existing detection object; An object classification step of classifying the type of the detected object; A masking processing step of performing a masking process on the detected object based on a master key; And a masking information insertion step of inserting information of the masked object. It provides, the masking processing method of the image.

In the present invention, the masking processing step, the region of interest derivation step of deriving a region of interest to be masked for each object; Generating a security key for the ROI; And encrypting the ROI using a master key. It may include.

In the present invention, the first security key generation step, the master key generation step of generating a master key based on the reference password and the predetermined unique information; A group key generation step of generating a group key based on the master key and group attribute information; And an object key generation step of generating an object key based on the group key and object property information. Includes, the security key may include the master key, the group key and the object key.

In the present invention, the masking processing method of the image further includes an unmasking processing step, wherein the unmasking processing step comprises: a masking information extraction step of extracting information of a masked object; An unmasking information input step of receiving information on at least one object to be unmasked and a security key; A security key verification step of confirming an unmasking authority of the received security key; An object specifying step of specifying an object to be unmasked; An ROI extraction step of extracting an ROI from information of an object to be unmasked; And a decrypting step of performing unmasking on the ROI using a master key.

In the present invention, in the object specifying step, when the security key input in the unmasking information input step is an object key for an individual object, the object to be unmasked is specified as an individual object, and the decrypting step uses the master key. By unmasking the object to be unmasked.

In the present invention, in the object specifying step, when the security key input in the unmasking information input step is a group key for a group object, the object to be unmasked is identified as a plurality of objects in a corresponding group, and the decrypting step includes the The master key can be used to unmask the object to be unmasked.

In order to solve the above problems, the present invention provides an image masking processing method, comprising: an object detecting step of detecting an object in an image; An object tracking step of confirming whether the detected object is the same object as the existing detection object; An object classification step of classifying the type of the detected object; A masking processing step of performing masking processing on the entire image area based on the master key; And a masking information insertion step of inserting information of the masked object. It provides, the masking processing method of the image.

In the present invention, the masking processing step, the region of interest derivation step of deriving a region of interest for each object; Generating a security key for the ROI; And encrypting the entire image area using the master key. It may include.

In the present invention, the first security key generation step, the master key generation step of generating a master key based on the reference password and the predetermined unique information; A group key generation step of generating a group key based on the master key and group attribute information; And an object key generation step of generating an object key based on the group key and object property information. Includes, the security key may include the master key, the group key and the object key.

In the present invention, the masking processing method of the image further includes an unmasking processing step, wherein the unmasking processing step comprises: a masking information extraction step of extracting information of a masked object; An unmasking information input step of receiving information on at least one object to be unmasked and a security key; A second security key generation step of generating a group key and an object key from the master key; A security key verification step of confirming an unmasking authority of the received security key; An object specifying step of specifying an object to be unmasked; An ROI extraction step of extracting an ROI from information of an object to be unmasked; And a decrypting step of performing unmasking on the ROI using a master key. It may include.

In the present invention, in the object specifying step, when the security key input in the unmasking information input step is an object key for an individual object, the object to be unmasked is specified as an individual object, and the decrypting step uses the master key. To unmask the object to be unmasked.

In the present invention, in the object specifying step, when the security key input in the unmasking information input step is a group key for a group object, the object to be unmasked is identified as a plurality of objects in a corresponding group, and the decrypting step includes the The master key can be used to unmask the object to be unmasked.

In order to solve the above problems, the present invention provides a masking processing apparatus of the image, the object detecting unit for detecting an object in the image; An object tracking unit to check whether the detected object is the same object as the existing detection object; An object classification unit classifying the type of the detected object; A masking processor configured to perform a masking process on the detected object based on an object-specific security key; And a masking information insertion unit for inserting information of the masked object. It provides a masking processing apparatus of the image.

In order to solve the above problems, the present invention provides a masking processing apparatus of the image, the object detecting unit for detecting an object in the image; An object tracking unit to check whether the detected object is the same object as the existing detection object; An object classification unit classifying the type of the detected object; A masking processor configured to perform a masking process on the detected object based on a master key; And a masking information insertion step of inserting information of the masked object. It provides a masking processing apparatus of the image.

In order to solve the above problems, the present invention provides a masking processing apparatus of the image, the object detecting unit for detecting an object in the image; An object tracking unit to check whether the detected object is the same object as the existing detection object; An object classification unit classifying the type of the detected object; A masking processing unit which performs masking processing on the entire image area based on the master key; And a masking information insertion unit for inserting information of the masked object. It provides a masking processing apparatus of the image.

According to an embodiment of the present invention, the masked video may be encrypted using a security key, thereby maintaining the security of the original video.

According to an embodiment of the present invention, an unmasking process may be performed for each object in an unmasking process of the masked image, thereby protecting personal information.

According to an embodiment of the present invention, when performing unmasking for each object, each object is based on a security key (object key) for each object, a security key (group key) for each object group, or an entire security key (master key). This can have the effect of performing unmasking, group by group, or full unmasking.

According to an embodiment of the present invention, it is possible to generate a group key and an object key based on the master key, thereby achieving the effect of easily managing the security key.

1 is a view schematically showing an operation flow of a masking processing device and an unmasking processing device according to an embodiment of the present invention.
2 is a block diagram schematically showing an internal configuration of a masking apparatus according to an embodiment of the present invention.
3 is a diagram schematically showing masking information according to an embodiment of the present invention.
4 is a diagram schematically illustrating ROI information according to an embodiment of the present invention.
5 is a block diagram schematically illustrating an internal configuration of a masking processing unit according to an embodiment of the present invention.
6 is a flowchart schematically illustrating a method of generating a security key according to an embodiment of the present invention.
7 is a diagram illustrating a security key set according to an embodiment of the present invention.
8 is a flowchart schematically showing each step of the masking processing method according to an embodiment of the present invention.
9 is a diagram schematically showing an embodiment of a masking image according to an embodiment of the present invention.
10 is a block diagram schematically showing an internal configuration of an unmasking processing apparatus according to an embodiment of the present invention.
11 is a block diagram schematically showing an internal configuration of an unmasking processing unit according to an embodiment of the present invention.
12 is a flowchart schematically showing each step of the unmasking processing method according to an embodiment of the present invention.
13 is a flowchart schematically showing each step of the unmasking processing method according to an embodiment of the present invention.
14 is a block diagram schematically showing an internal configuration of an unmasking processing unit according to another embodiment of the present invention.
15 is a flowchart schematically showing each step of the unmasking processing method according to another embodiment of the present invention.

In the following, various embodiments and / or aspects are now disclosed with reference to the drawings. In the following description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of one or more aspects. However, it will also be appreciated by one of ordinary skill in the art that this aspect (s) may be practiced without these specific details. The following description and the annexed drawings set forth in detail certain illustrative aspects of the one or more aspects. However, these aspects are exemplary and some of the various methods in the principles of the various aspects may be used and the descriptions described are intended to include all such aspects and their equivalents.

Moreover, various aspects and features will be presented by a system that may include a number of devices, components, and / or modules, and the like. The various systems may include additional devices, components, and / or modules, etc., and / or may not include all of the devices, components, modules, etc. discussed in connection with the drawings. It must be understood and recognized.

As used herein, “an embodiment”, “an example”, “aspect”, “an example”, etc., may not be construed as having any aspect or design described being better or advantageous than other aspects or designs. . The terms '~ part', 'component', 'module', 'system', 'interface', etc. used generally mean a computer-related entity, for example, hardware, hardware And a combination of software and software.

In addition, the terms "comprises" and / or "comprising" mean that such features and / or components are present, but exclude the presence or addition of one or more other features, components, and / or groups thereof. It should be understood that it does not.

In addition, terms including ordinal numbers such as first and second may be used to describe various components, but the components are not 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 the second component, and similarly, the second component may also be referred to as the first component. The term and / or includes a combination of a plurality of related items or any item of a plurality of related items.

In addition, in the embodiments of the present invention, unless otherwise defined, all terms used herein including technical or scientific terms are generally understood by those skilled in the art to which the present invention belongs. Has the same meaning as Terms such as those defined in the commonly used dictionaries should be interpreted as having meanings consistent with the meanings in the context of the related art, and ideally or excessively formal meanings, unless explicitly defined in the embodiments of the present invention. Not interpreted as

1 is a view schematically showing the operation flow of the masking processing device 100 and the unmasking processing device 200 according to an embodiment of the present invention.

Referring to FIG. 1, the masking processing apparatus 100 according to an embodiment of the present invention receives an image photographed by CCTV and the like, classifies an area requiring masking, and inserts masking. Such masking is performed to protect personal information in the image. In an embodiment of the present invention, the original image may not be recognized by inserting a mosaic or inserting an image of a different color. Alternatively, a method of encrypting the macroblocks in the region by using a security key to modulate the contour without recognizing it. In the embodiment of FIG. 1, a masking process is performed to mosaic a device of an input image.

At this time, the masking processing apparatus 100 does not modulate the original image, but generates a masking image including the masking information encrypted in the original image, so that only the masked image can be viewed without a security key necessary for decryption. If you use the security key, you can see the original video without masking.

In an embodiment of the present invention, the masking processing apparatus 100 detects and tracks an object in an image, grasps object information on a screen, encrypts a corresponding object area using a security key in the encoding process, and stores information of the object in an image. To pass.

To this end, the unmasking processing apparatus 200 according to an embodiment of the present invention decodes a masked image generated by the masking processing apparatus 100 using a security key and removes the inserted masking to generate an original image. Do it.

As such, only the person with the security key can check the original image, while protecting the personal information included in the image, and the person having the authority can exert the effect of checking the original image.

For example, in the case of a CCTV image of a crime or accident, the video provided for the purpose of reporting, etc. is masked to prevent the leakage of personal information, but in order to process the information, such as investigative agencies that need to know the exact original image The security key can be used to check the original video without masking.

2 is a block diagram schematically showing an internal configuration of a masking apparatus according to an embodiment of the present invention.

Referring to FIG. 2, the masking processing apparatus 100 according to an exemplary embodiment of the present invention may include an object detector 110, an object tracker 120, an object classification unit 130, a masking processor 140, and a masking information insertion unit. And 150.

The object detector 110 detects an object in the image. The object detector 110 may detect objects in the image by separating them from the background through various image processing techniques.

The object tracking unit 120 checks whether the detected object is the same object as the existing detection object. As described above, the object tracking unit 120 checks whether the object detected by the object detection unit 110 is the same object as the object detected before or after the image. As such, by continuously tracking the object and checking the same object, information on the same object can be maintained and tracked even when the image is reproduced and the position of the detected object is changed.

The object classification unit 130 classifies the type of the detected object. Through this, a group and the objects belonging to the group can be determined. For example, if the object classification unit 130 is classified as a person, a vehicle number, a signboard, etc., the first group of people consists of the first person, the second person, and the like, and the second vehicle group number is the first vehicle. , The second vehicle, etc., and the third group will be composed of the first signboard, the second signboard, and so on. Each group and object information may be inserted into an image after data making and used for unmasking.

The masking processor 140 performs a masking process on the detected object based on an object-specific security key. In the present invention, a masking process may be performed on the detected objects using different security keys to protect the privacy of each object as much as possible. If masking processing is performed with one security key instead of a separate security key for each object, when unmasking with the security key, unmasking is impossible for each object individually and uniformly unmasking In addition, as well as the object that needs to be unmasked, as well as other unnecessary objects are unmasked, the privacy of other objects is not protected. Masking processor 140 according to an embodiment of the present invention can solve this problem by performing a masking process for each object based on the security key for each object.

The masking information insertion unit 150 inserts information of the masked object. In an embodiment of the present invention, when the masking processing unit 140 includes information on the masked object and the masked object in the image data, when the unmasking processing apparatus 200 unmasks the masked image, Get information about the object to help you perform unmasking.

3 is a diagram schematically showing masking information according to an embodiment of the present invention.

Referring to FIG. 3, masking information according to an embodiment of the present invention is first stored with the total number of objects 51 and the total number of groups 52 included in the image, and the index 53 of the group 1 is stored. Next, the number 54 of objects included in the group 1 and the masking information of the objects included in the group are stored from (55) to (56). After storing the group 1, the index 57 of the group 2 is stored, and the number 58 of objects included in the group 2 and the information of each object are stored in the same format as the group 1. At the end, the separator End Of Info (59) is inserted to indicate the end of the information.

As described above, the masking information according to an embodiment of the present invention may include the total number of objects included in the image, the total number of groups included in the image, the number of objects included in each group, and the masking information of each object. have. Such masking information can be used when performing unmasking later.

4 is a diagram schematically illustrating ROI information according to an embodiment of the present invention.

As in the example of FIG. 3, the masking information inserted into the image includes masking information of each object. In this case, each object may have a region of interest (ROI) set for each object and may be masked by inserting masking on the region of interest. Accordingly, the masking information of each object may represent information about the masking by including information of the ROI. Accordingly, the masking information according to an embodiment of the present invention includes at least one of an XY coordinate of the ROI with respect to the object, a width of the ROI, a height of the ROI, and a radius of the ROI. It may represent masking information about an object.

Referring to the example of FIG. 4, the information about the object 1 of the group 1 includes the X coordinate, the Y coordinate, the width and the height of the ROI. By such information, it is possible to know that the ROI has a rectangular shape, and through this, the ROI can be derived.

The information about the object 2 of the group 1 includes the X coordinate, the Y coordinate and the radius of the ROI. As such, when the ROI is represented by a radius, the ROI may be circular, and the ROI may be derived.

The information about the object 3 of the group 1 includes the X coordinate and the Y coordinate of the ROI. As such, when only the X-Y coordinate of the ROI is included, the ROI may be defined in a predetermined size and shape in order to derive the ROI. Alternatively, in another embodiment, the object detecting unit 110, the object tracking unit 120, and the object classification unit 130 of the masking processing apparatus 100 may include only the coordinates of the ROI and receive and process the image. By detecting the shape and size of the object, including a module such as), it is possible to detect the object to derive a region of interest for processing the masking on the object.

5 is a block diagram schematically illustrating an internal configuration of a masking processing unit according to an embodiment of the present invention.

Referring to FIG. 5, the masking processor 140 according to an embodiment of the present invention includes a region of interest derivation unit 141, a first security key generation unit 142, and an encryption unit 143.

The ROI-deriving unit 141 derives an ROI to perform masking. The ROI may be derived based on the information of the objects detected and classified by the object detector 110, the object tracker 120, and the object classifier 130. In one embodiment of the present invention, the ROI may have a rectangular or circular shape.

The first security key generation unit 142 generates a security key for the ROI for each object. According to an embodiment of the present invention, the security key includes a master key, a group key, and an object key, and generates and manages a security key hierarchically.

In the case of maintaining security using a separate security key for all objects included in the image as in the present invention, all the security keys as many as the number of the object must be transmitted and decoded using the security key. The process also increases in complexity linearly as the number of objects increases. However, as in the exemplary embodiment of the present invention, the key is generated hierarchically to facilitate the management of the security key, and at the same time, the security level of the individual objects can be maintained at the same level as using the respective security key.

The master key is a key for releasing masking for all objects included in the image, the group key is a key for releasing masking for all objects in the group, and the object key is a corresponding key. The key to unmask the object.

In an embodiment of the present invention, an administrator or the like may input a reference password to generate all of the master key, the group key, and the object key based on the reference password. The reference password is a seed for generating the security key as a string or a string of characters input by the administrator.

According to an embodiment of the present invention, a security key may be generated based on unique information derived from an image except for the reference password inputted by an administrator. According to an embodiment of the present invention, a master key may be generated based on a reference password and predetermined unique information, and an individual key may be generated based on the master key and the individual object information. The individual key may be a group key or an object key. In more detail, a group key may be generated based on the master key and the group attribute information, and an object key may be generated based on the group key and the object attribute information. By generating a hierarchical security key in this way, it is possible to facilitate security key management.

The encryption unit 143 performs masking on the ROI using a security key for each object. The encryption unit 143 encrypts masking inserted into each object by using the object key generated by the first security key generation unit 142. As described above, by encrypting each object key, the masking can be released without the respective object key, so that the original image cannot be identified, and only the masked image can be identified, thereby preventing privacy leakage.

In one embodiment of the present invention, the encryption unit 143 processes the masking on the ROI and encrypts the same, thereby releasing the masking without a security key so that the original image cannot be confirmed.

In one embodiment of the present invention, the encryption unit 143 performs an XOR or scramble process on a field having a large influence on the screen display of the macroblock in the ROI by using a predetermined rule based on the object key. The outline is also made into an unknown image and masking is performed. By masking the object using the object key as described above, the original image can be checked even if the original image is not recognized through the object key, so that the original image can be checked and the masking can be processed without loss of image information. The object key is required for unmasking, so that the original image cannot be checked without the object key, thereby maintaining security.

In another embodiment of the present invention, the encryption unit 143 may mask the region of interest based on the master key instead of the object key as in the above embodiment. In this way, by masking each region of interest based on the master key instead of the respective object key in each region of interest, it is possible to expect fast processing speed, while maintaining the security of the master key to maintain the privacy of the object. Will be able to protect.

Alternatively, when the encryption unit 143 performs masking by the master key, it does not mask the region of interest as in the above embodiment, but performs masking processing on the entire region of the image based on the master key. It may be. In this embodiment, the region of interest is generated, but when the masking process is performed, the region of interest is operated as the region of interest for unmasking to perform the unmasking on the corresponding region, and the masking process is performed on the entire region of the image. do.

Even in such an embodiment, the ROI and the object key for the ROI are generated to serve as a security key for unmasking the ROI. The administrator of the image delivers a security key to a user requesting the viewing of the image, and the like, while the user, etc., unmasks and checks only a necessary part of the image through the security key, while maintaining privacy for other objects. Unmask the object of interest.

In this case, during the unmasking process, the security key is generated by the master key in the same manner to check whether the security key matches with the security key input by the user, etc., and thus the authority of the unmasking for the corresponding area can be checked. have.

6 is a flowchart schematically illustrating a method of generating a security key according to an embodiment of the present invention.

According to an embodiment of the present invention, the group key and the object key of the security key are generated based on the master key, but do not infer or calculate the master key from the group key or the object key to the master key. Need to be secure. To this end, in an embodiment of the present invention, by generating a security key through a hash function, it is possible to prevent backtracking for the master key.

Referring to FIG. 6, the first security key generation unit 142 according to an embodiment of the present invention first converts an image time, which is one of the reference password and unique information of the image, through a hash function to generate a master key ( S10). Thereafter, the first security key generation unit 142 checks how many groups of objects in the image are present (S20). At this time, if there is more than one group, a group key is generated by converting a group number among the master key and group attribute information generated in step S10 through a hash function (S30). After that, the first security key generation unit 142 checks how many objects exist in each group (S40). In this case, when there is more than one object, an object key is generated by converting an object number among the group key and the object attribute information generated in step S30 through a hash function (S50). Through this process, each group key and each object key can be generated using only one master key generated based on the reference password. This makes use of the characteristic that the result of the hash function is completely changed even if the message is changed a little due to the difficult traceback characteristic of the hash function and the avalanche effect.

7 is a diagram illustrating a security key set according to an embodiment of the present invention.

Referring to FIG. 7, when the reference password is 'pw', the first security key generation unit 142 according to an embodiment of the present invention adds the video time information '10124036' to the reference password 'pw'. Is inputted into hash function to obtain hash value. At this time, the hash value is 'CF1EB7325A671F80', which becomes a master key. In the embodiment of Figure 7, the hash function outputs a hash value of 16 digits, but in another embodiment of the present invention, SHA-1, SHA-224, SHA-256, SHA-384, SHA-512, HAS- Various hash functions such as 160 can be used.

Thereafter, in order to generate the group key, the first security key generation unit 142 inputs the hash value 'DAC819D2F23C85D4' to the master function 'CF1EB7325A671F80' by adding 'CF1EB7325A671F801', which adds the group number 1 of group 1 to the hash function. Is obtained, which becomes the group key # 1.

In order to generate the object key in the same manner, the first security key generation unit 142 hashes 'DAC819D2F23C85D41' and 'DAC819D2F23C85D42', which have added object number 1 and object number 2 to the group key # 1 'DAC819D2F23C85D4', respectively. By inputting to the function, '2C3F60B6A6B42113' and '45D6F1780B8A3CC8' are obtained, which becomes object key # 1-1 and object key # 1-2, respectively.

Similarly for the group 2, the first security key generation unit 142 obtains the hash value '1AF7A0CBE3D17F4F' by inputting 'CF1EB7325A671F802', which adds the group number 2 of group 2 to the master key 'CF1EB7325A671F80', into a hash function. This becomes group key # 2. In addition, '1AF7A0CBE3D17F4F1' and '1AF7A0CBE3D17F4F2', which have added object number object number 1 and object number 2 to the group key # 2 '1AF7A0CBE3D17F4F', respectively, are entered into the hash function to obtain 'C39F233110DECFFA' and 'FC45F3EB' This becomes the object key # 1-1 and the object key # 1-2, respectively.

As a result of adding the group numbers 1 and 2 to the master key and inputting it into the hash function, the input value is only one number difference, but the result value shows a big difference, and tracking the original input value from the hash value It becomes very difficult. This is similar to the case of adding an object number to a group key and inputting it into a hash function. Even though each object key is generated based on one reference password and video time information, another object key, group key, or master is generated from each object key. It is difficult to infer keys, which can be effective in maintaining security.

That is, although the lower object key can be generated from the master key or the group key, the upper master key cannot be traced back from the object key or the group key, so that the security level can be easily maintained and the security level can be maintained. It can be effective.

In an embodiment of the present invention, the method of converting the master key and the group number through a hash function may be converted by adding a group number after the string of the master key as shown in the example of FIG. 7, or the hexadecimal number of the master key. The master key may be converted into a form of adding, subtracting, or multiplying the code to a code, and then converting the master key through a hash function. For example, in order to obtain the group key # 1, the group key # 1 may be obtained by inputting the master key 'CF1EB7325A671F80' without subtracting the group number 1 of the group 1 into the hash function.

The same applies to the method of converting the group key and the object number through a hash function.

8 is a flowchart schematically showing each step of the masking processing method according to an embodiment of the present invention.

The following steps are the same as or similar to the operations of the respective parts of the masking processing apparatus 100 described above, and the description of the same parts will be omitted.

Referring to FIG. 8, the masking processing method according to an embodiment of the present invention first performs an object detection step S110 of detecting an object in an image.

Thereafter, the object tracking step (S120) is performed to determine whether the detected object is the same object as the existing detection object.

Thereafter, an object classification step S130 of classifying the detected types of objects is performed.

Thereafter, a region of interest derivation step S140 of deriving a region of interest for masking is performed.

Thereafter, a first security key generation step S150 of generating a security key for the ROI is performed.

Thereafter, an encryption step (S160) of performing masking on the region of interest using a security key for each object is performed.

Thereafter, a masking information insertion step S170 of inserting information of the masked object is performed.

Through such a series of steps, an image masking process may be performed using an independent password for each object.

9 is a diagram schematically showing an embodiment of a masking image according to an embodiment of the present invention.

9 (a) is an original image, and the object detector 110, the object tracker 120, and the object classifier 130 are the people 11, 12, 13, 14, and blank of the original image. (15) (16) is detected and grouped into a person and a ball, and the person is identified as 4 objects (11) (12) (13) (14) and a ball as 2 objects (15) (16), respectively.

9B illustrates an example of a masked image, in which 21 to 24 are masked with 4 objects in a human group (group 1), and 25 and 26 are masked with 2 objects in a ball group (group 2). In the example of FIG. 9, mosaics are inserted by masking. The ROI of each object has a rectangular shape, and the ROI information may store the X-Y coordinate, the width of the ROI and the height of the ROI.

9C illustrates an example of a screen for confirming a masked image. In the example of FIG. 9C, when the masked image is played through the dedicated viewer, when the mouse cursor 32 is moved to the masked object, the ID information 31 of the object where the cursor is located is displayed on the screen. do. In the embodiment of FIG. 9, the object indicated by the cursor is a person 14, and ID 1-4 is displayed on the screen as the fourth object of the person group (group 1). The user who confirms the ID informs the image provider of the ID of the object, receives the object key of the object of the ID, and unmasks only the object of the ID. If you want to check all of the objects, you can obtain the group key of Group 1 after requesting it, and you can check the entire group of people by unmasking. You can do this by unmasking the entire ball group.

In this way, the user who checks the image acquires only the object key of the required object and performs unmasking, thereby preventing unmasking while exposing privacy to other objects.

In addition, for each object, the group key of the corresponding group is acquired to perform unmasking, or the master key is obtained to perform unmasking. You can also unmask the entire object.

10 is a block diagram schematically showing an internal configuration of an unmasking processing apparatus according to an embodiment of the present invention.

Referring to FIG. 10, the unmasking processing apparatus 200 according to an embodiment of the present invention includes a masking information extracting unit 210, an unmasking information input unit 220, and an unmasking processing unit 230.

The masking information extracting unit 210 extracts information of the masked object. In the masking image according to an embodiment of the present invention, information of an object masked by the masking information insertion unit 150 is inserted. The masking information extracting unit 210 extracts this from the image and loads it.

The unmasking information input unit 220 receives the information and the security key of the object to be unmasked. According to an embodiment of the present invention, the unmasking information input unit 220 may receive an object to be unmasked and designate it as an unmasking target, or may receive a group to be unmasked, and collectively unmask the objects in the group. Can also be specified. In addition, when the security key for the unmasked object is input, the object key for the unmasked object may be input, or a group key or a master key may be received.

That is, when an object to be unmasked is input, a corresponding object key, a group key of a group including the object, or a master key may be input. When an object to be unmasked is input, a group key or master of the group The key can be input, and if the entire object is input for unmasking, the master key can be input.

According to an embodiment of the present invention, the unmasking information input unit 220 may receive information on a plurality of unmasked objects and a plurality of security keys for the plurality of unmasked objects. The unmasking information input unit 220 may receive information on a plurality of objects and a plurality of security keys as described above, and simultaneously process unmasking.

In another embodiment of the present invention, the unmasking information input unit 220 may receive only a security key for the object without information of the object to be unmasked. In such an embodiment, the unmasking processor 230 generates a group key and an object key from the master key in the same manner as the first security key generator 142, and the unmasking information input unit 220 receives the input. By comparing the security key with the group key and the object key, by identifying the object to be unmasked to generate information about the object to be unmasked, the unmasking processor 230 can perform the unmasking.

The unmasking processor 230 performs an unmasking process based on the received object information and the security key. According to an embodiment of the present invention, the unmasking processing unit 230 may include masking information extracted by the masking information extracting unit 210 and information and security keys of an object to be unmasked received from the unmasking information input unit 220. On the basis of this, the masking information for the object to be unmasked among the masking information is decoded based on the security key, so that the unmasking process can be performed.

11 is a block diagram schematically showing an internal configuration of an unmasking processing unit according to an embodiment of the present invention.

Referring to FIG. 11, the unmasking processor 230 according to an embodiment of the present invention includes an ROI extractor 231, a second security key generator 232, and a decryptor 233.

The ROI extractor 231 extracts an ROI from information of an object to be unmasked. The ROI extractor 231 may extract information about the ROI from the masked object information extracted by the masking information extractor 210. In this case, only the region of interest for the unmasked target object received from the unmasking information input unit 220 may be extracted to prepare for unmasking.

The second security key generation unit 232 generates a security key for each object from the received security key. When the security key received from the unmasking information input unit 220 is a master key or a group key, an object key is generated to perform unmasking. The method of generating the object key from the master key or the group key is the same as the first security key generation unit 142. When the second security key generation unit 232 receives a master key or a group key from the unmasking information input unit 220, the second security key generation unit 232 may generate all object keys that can be generated from the master key or the group key. Alternatively, the unmasking information input unit 220 may receive information on an object to be unmasked and may generate and generate only an object key corresponding to the object to be unmasked.

The decryption unit 233 performs unmasking on the ROI using the security key for each object. The decoder 233 decrypts the ROI extracted from the ROI extractor 231 using the object key and removes masking. This corresponds to the encryption unit 143 of the masking processing device 100 to decrypt the encryption performed by the encryption unit 143 to access and remove the masking of the object.

According to an embodiment of the present disclosure, the encryption unit 143 may perform masking by XORing a macroblock in the ROI using an object key or scrambled by a predetermined rule based on the object key. At this time, the decryption unit 233 may perform unmasking by XORing the macroblock in the ROI using the object key again based on the object key or scrambles the reverse of the preset rule. When masking and unmasking are performed as described above, the image is changed so that the original image cannot be recognized while performing masking without additional information on the masking and without loss of the information of the original image, and only through the object key. The effect can be restored.

As described above, the unmasking target object is specified for the image encrypted by the object key for each object by the operation of each unit of the unmasking processor 230, and the unmasking object is unmasked to recognize the original image. You can convert it into a video.

12 and 13 are flowcharts schematically showing steps of an unmasking processing method according to an embodiment of the present invention.

The following steps are the same as or similar to the operation of each part of the unmasking processing apparatus 200 described above, and the description of the same parts will be omitted.

Referring to FIG. 12, the unmasking processing method according to an embodiment of the present invention first performs a masking information extraction step (S210) of extracting information of a masked object.

Thereafter, an unmasking information input step (S220) of receiving information of at least one unmasked object and a security key is performed.

Thereafter, a region of interest extraction step S240 of extracting a region of interest from information of an object to be unmasked is performed.

Thereafter, a decryption step S250 of performing unmasking on the ROI using the security key is performed.

Through such a series of steps, the unmasked image may be unmasked using an independent password for each object.

FIG. 13 illustrates a step in a case where the security key received in the unmasking information input step S220 is a master key or a group key instead of an object key.

In the embodiment of FIG. 13, steps S210 and S220 are performed in the same manner as in the embodiment of FIG. 12.

Thereafter, a second security key generation step S230 of generating an object key from a master key or a group key is performed. The second security key generation step S230 is performed in the same manner as the first security key generation step S150 of the masking processing method to generate an object key, thereby obtaining an object key necessary for unmasking. Therefore, when the master key is input in the unmasking information input step (S220), in the second security key generation step (S230), the object keys of all objects may be obtained, or the unmasking information input step (S220). In case of receiving a group key, the second security key generation step (S230) may obtain object keys of all objects in the group.

Thereafter, steps S240 and S250 are performed in the same manner as in the embodiment of FIG. 12. At this time, in the decryption step S250, the original image may be obtained by unmasking the ROI using the object key generated in the second security key generation step S230.

In this way, by receiving the group key or the master key as well as the object key in the unmasking step, unmasking of all objects in the input group or all objects in the image can be performed to simplify security key management. .

14 is a block diagram schematically showing an internal configuration of an unmasking processor 230 according to another embodiment of the present invention.

In the embodiment of FIG. 14, when the encryption unit 143 masks the region of interest based on the master key instead of the object key, or when the entire region of the image is masked, the unmasking processing unit may be identified. have.

In such an embodiment, the unmasking processor 230 includes a third security key generator 234, an authorization checker 235, an ROI extractor 231, and a decryptor 233.

The third security key generation unit 234 generates a group key and an object key from the master key. The third security key generation unit 234 generates a group key and an object key through the same algorithm as the first security key generation unit 142. The group key and the object key generated as described above are used to confirm the unmasking authority of the security key input from the unmasking information input unit 220.

The authorization checker 235 checks the unmasking authority of the security key received from the unmasking information input unit 220 and specifies an object to be unmasked. The authorization checker 235 compares the security key received from the unmasking information input unit 220 with the group key and the object key generated by the third security key generation unit 234. If it matches with, it determines that the user has permission to unmask and specifies the group or the object as an object to be unmasked. By specifying the object to be unmasked according to the security key input as described above, it is possible to unmask only the specified object without unmasking other objects, thereby maintaining the privacy of other objects. Can be.

The region of interest extractor 231 extracts the region of interest from the information of the object to be unmasked specified in the authority checker 235, as in the above-described embodiment. Information about the ROI may be extracted from masking information embedded in the image.

The decoder 233 performs unmasking on the ROI or the entire image area based on the master key. As described above, the decryption unit 233 performs fast processing by performing single key masking and unmasking using the master key. Masking directly to the macro block of an object adversely affects the motion estimation function of the video codec, which may cause noise in which the masking flows to the lower right corner.

In another embodiment of the present invention, in order to prevent this, masking is performed on the ROI or the entire image area based on the master key, and when the unmasking is performed, the decoder 233 unmasks the mask based on the master key. Do this.

According to an embodiment of the present invention, when performing unmasking on the ROI, the original image from which the masking of the ROI is removed is displayed and the masked image is displayed for other objects, thereby individually displaying each object. Unmasking can be done with

Alternatively, when unmasking the entire image area, the unchecked original image is displayed only in the ROI for the object specified as the object to be unmasked by the permission checker 235, and for the other objects By displaying the masked image, only the corresponding object may display the unmasked final displayed image to the user, thereby maintaining the privacy of other objects.

15 is a flowchart schematically showing each step of the unmasking processing method according to another embodiment of the present invention.

FIG. 15 illustrates an unmasking processing method in an embodiment in which masking is performed by a master key instead of each object key in the masking process.

Referring to FIG. 15, the unmasking processing method according to another embodiment of the present invention first performs a masking information extraction step (S210) of extracting information of a masked object.

Thereafter, an unmasking information input step S220 of receiving information of an object to be unmasked and a security key is performed.

Thereafter, a third security key generation step (S231) of generating a group key and an object key from the master key is performed.

Thereafter, a security key verification step (S232) of checking the unmasking authority of the received security key is performed. In the security key checking step (S232), the security key is frozen by comparing the security key received in the unmasking information input step (S220) with a group key or an object key generated in the third security key generation step (S231). Check whether you have permission to perform masking. In this case, when the received security key is a group key or a master key, the group key or master key may have a right for unmasking all objects in the group or all objects in the image.

Thereafter, an object specifying step S234 of specifying an object to be unmasked is performed. In the object specifying step (S234), only the object whose unmasking authority of the security key is confirmed through the security key verification step (S232) is identified as an object to be unmasked, and then unmasking can be performed.

Thereafter, a region of interest extraction step S240 of extracting a region of interest from information of an object to be unmasked is performed.

Thereafter, a decryption step S250 of performing unmasking on the ROI using the security key is performed.

Through such a series of steps, the unmasked image may be unmasked using an independent password for each object.

According to an embodiment of the present invention, the masked video may be encrypted using a security key, thereby maintaining the security of the original video.

According to an embodiment of the present invention, an unmasking process may be performed for each object in an unmasking process of the masked image, thereby protecting personal information.

According to an embodiment of the present invention, when performing unmasking for each object, each object is based on a security key (object key) for each object, a security key (group key) for each object group, or an entire security key (master key). This can have the effect of performing unmasking, group by group, or full unmasking.

According to an embodiment of the present invention, it is possible to generate a group key and an object key based on the master key, thereby achieving the effect of easily managing the security key.

Methods according to an embodiment of the present invention may be implemented in the form of program instructions that may be executed by various computing devices and may be recorded in a computer readable medium. In particular, the program according to the present embodiment may be configured as a PC-based program or an application dedicated to a mobile terminal. An application to which the present invention is applied may be installed in a user terminal through a file provided by a file distribution system. For example, the file distribution system may include a file transmitter (not shown) for transmitting the file at the request of the user terminal.

The apparatus described above may be implemented as a hardware component, a software component, and / or a combination of hardware components and software components. For example, the devices and components described in the embodiments are, for example, processors, controllers, arithmetic logic units (ALUs), digital signal processors, microcomputers, field programmable gate arrays (FPGAs). May be implemented using one or more general purpose or special purpose computers, such as a programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions. The processing device may execute an operating system (OS) and one or more software applications running on the operating system. The processing device may also access, store, manipulate, process, and generate data in response to the execution of the software. For convenience of explanation, one processing device may be described as being used, but one of ordinary skill in the art will appreciate that the processing device includes a plurality of processing elements and / or a plurality of types of processing elements. It can be seen that it may include. For example, the processing device may include a plurality of processors or one processor and one controller. In addition, other processing configurations are possible, such as parallel processors.

The software may include a computer program, code, instructions, or a combination of one or more of these, and configure the processing device to operate as desired, or process it independently or in combination. You can command the device. Software and / or data may be any type of machine, component, physical device, virtual equipment, computer storage medium or device in order to be interpreted by or to provide instructions or data to the processing device. Or may be permanently or temporarily embodied in a signal wave to be transmitted. The software may be distributed over networked computing devices so that they are stored or executed in a distributed manner. Software and data may be stored on one or more computer readable recording media.

The method according to the embodiment may be embodied in the form of program instructions that can be executed by various computer means and recorded in a computer readable medium. The computer readable medium may include program instructions, data files, data structures, etc. alone or in combination. The program instructions recorded on the media may be those specially designed and constructed for the purposes of the embodiments, or they may be of the kind well-known and available to those having skill in the computer software arts. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tape, optical media such as CD-ROMs, DVDs, and magnetic disks, such as floppy disks. Magneto-optical media, and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine code generated by a compiler, but also high-level language code that can be executed by a computer using an interpreter or the like. The hardware device described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

Although the embodiments have been described by the limited embodiments and the drawings as described above, various modifications and variations are possible to those skilled in the art from the above description. For example, the described techniques may be performed in a different order than the described method, and / or components of the described systems, structures, devices, circuits, etc. may be combined or combined in a different form than the described method, or other components. Or even if replaced or substituted by equivalents, an appropriate result can be achieved. Therefore, other implementations, other embodiments, and equivalents to the claims are within the scope of the claims that follow.

Claims (26)

As a masking processing method of an image,
An object detection step of detecting an object in the image;
An object tracking step of confirming whether the detected object is the same object as the existing detection object;
An object classification step of classifying the detected objects into a specific group;
When masking the entire area of the image, the masking process is performed based on the master key, and when the masking process is performed on the detected object, the object key corresponding to the object or the master key is based on the master key. A masking processing step of performing masking processing;
A masking information insertion step of inserting information of the masked object; And
And an unmasking step;
The masking processing step,
Deriving a region of interest for deriving a region of interest for masking for each object;
Generating a security key for the ROI; And
And encrypting the ROI or the entire area of the image using the security key.
The first security key generation step,
A master key generation step of generating the master key based on a reference password and predetermined unique information;
A group key generation step of generating a group key based on the master key and group attribute information; And
And an object key generation step of generating the object key based on the group key and object property information.
The security key includes the master key, the group key and the object key,
The unmasking processing step,
A masking information extraction step of extracting information of the masked object;
An unmasking information input step of receiving information on at least one object to be unmasked and at least one security key;
When the master key is input in the unmasking information input step, the group key is generated based on the master key and the group attribute information, and the object key is generated based on the generated group key and the object attribute information. A second security key generation step of generating the object key based on the group key and the object property information when the group key is received;
A region of interest extraction step of extracting at least one region of interest from the information of the at least one object to be unmasked; And
And decrypting the at least one region of interest using the object key or the master key generated in the second security key generation step.
delete delete delete The method according to claim 1,
The encryption step,
And masking by changing or scrambled macroblocks in the ROI based on a predetermined rule based on the object key.
The method according to claim 1,
The master key is generated by converting the reference password and the predetermined unique information through a hash function,
The group key is generated by converting the master key and the group attribute information through a hash function,
And the object key is generated by converting the group key and the object attribute information through a hash function.
The method according to claim 6,
The information on the masked object is
The total number of objects contained in the image;
The total number of groups contained in the image;
The number of objects contained in each of the groups; And
Masking information including object identification information and object region information of each of the objects; Includes, the masking processing method of the image.
The method according to claim 7,
The object area information,
An XY coordinate of the region of interest for the object;
A width of the region of interest;
A height of the region of interest; And
A radius of the region of interest; Masking processing method of the image, including any one or more of.
delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete As a masking processing device for an image,
An object detector detecting an object in the image;
An object tracking unit to check whether the detected object is the same object as the existing detection object;
An object classification unit for classifying the detected objects into a specific group;
When masking the entire area of the image, the masking process is performed based on the master key, and when the masking process is performed on the detected object, the object key corresponding to the object or the master key is based on the master key. A masking processing unit which performs masking processing;
Masking information insertion unit for inserting the information of the masked object; And
It includes; unmasking processing unit,
The masking processing unit,
A region of interest derivation unit for deriving a region of interest for masking for each object;
A first security key generation unit generating a security key for the ROI; And
And an encryption unit for masking the ROI or the entire region of the image using the security key.
The first security key generation unit,
A master key generation unit generating the master key based on a reference password and predetermined unique information;
A group key generation unit for generating a group key based on the master key and group attribute information; And
And an object key generation unit generating the object key based on the group key and object property information.
The security key includes the master key, the group key and the object key,
The unmasking processing unit,
Masking information extraction unit for extracting the information of the masked object;
An unmasking information input unit configured to receive information on at least one object to be unmasked and at least one security key;
When the master key is input by the unmasking information input unit, the group key is generated based on the master key and the group attribute information, and the object key is generated based on the generated group key and the object attribute information. A second security key generation unit generating the object key based on the group key and the object property information when the group key is received;
A region of interest extraction unit which extracts one or more regions of interest from the information of the one or more objects to be unmasked; And
And a decryption unit which performs unmasking on the at least one region of interest using the object key or the master key generated by the second security key generation unit. delete delete

Images