KR20210071358A - System and method for encrypting super-pixels - Google Patents

Abstract

The present invention relates to a super pixel image encryption apparatus and method. According to an embodiment of the present invention, an apparatus and method for encrypting a super pixel image increases security and processing speed by encrypting an image in units of super pixels through high-speed parallel operation. The super pixel image encryption apparatus includes an extraction part, an encryption part, a decryption part, an authentication key management part, and a transmission part.

Description

Apparatus and method for super pixel image encryption {SYSTEM AND METHOD FOR ENCRYPTING SUPER-PIXELS}

The present invention relates to a super-pixel image encryption apparatus and method, and more particularly, to a super-pixel image encryption apparatus and a super-pixel image encryption apparatus for encrypting and decrypting a large-capacity image in parallel at high speed using a plurality of computing devices in units of super pixels; it's about how

In general, when processing an image, an object is searched for by using one image as a pixel grid and matching it with the fixed size (50 X 50) of the generally stored image. In this case, the pixel grid is an artificial fixed expression, not a fixed expression. Therefore, a more natural and more efficient pixel grid is required for high-speed processing of current large-capacity images. In addition, the computational amount is too large even with current computer technology to process data in normal pixel units. Since a super pixel is a set of pixels having similar characteristics, the result of dividing the input image into meaningful units is obtained. These clustered pixels can improve processing speed without compromising performance. Because of the various advantages of super pixels, various methods for generating super pixels are being studied, and although the regular truncation method is used a lot recently, the normal truncation method needs to improve a lot of computation.

On the other hand, video or video service providers that require security apply many encryption and decryption processes so that the video is leaked or unauthorized users cannot check the video even after acquiring the video. However, there is a problem that a lot of time and resources are consumed to encrypt and decrypt the entire vast amount of multimedia data.

Background art of the present invention is disclosed in Korean Patent Registration No. 10-0848642.

The present invention provides an apparatus and method for encrypting a super pixel image having security that cannot reproduce an image even if the image is leaked by encrypting it in a super pixel unit.

The present invention provides a super-pixel image encryption apparatus and method without a time delay in encryption and decryption because high-speed parallel processing is performed using a plurality of computing devices.

According to one aspect of the present invention, there is provided an apparatus for encrypting a super pixel image.

In the apparatus for encrypting a super pixel image according to an embodiment of the present invention, an extracting unit for extracting a super pixel from an image, an encryption unit for encrypting the image, a decryption unit for decrypting the encrypted image, and an authorized user terminal are determined It may include an authentication key management unit and a transmission unit for transmitting the decrypted image.

According to another aspect of the present invention, there is provided a computer-readable recording medium in which a superpixel image encryption method and a computer program executing the same are recorded.

The super pixel image encryption method according to an embodiment of the present invention and a recording medium storing a computer program executing the same are the super pixel image encryption methods, comprising the steps of extracting a super pixel from an image, encrypting the image, and It may include determining a user terminal, decoding the image, and transmitting the image.

According to an embodiment of the present invention, the apparatus and method for encrypting a super-pixel image can enhance security by encrypting an image in units of super-pixels so that the image cannot be checked even if the image is played back due to leakage.

According to an embodiment of the present invention, since the superpixel image encryption apparatus and method use a plurality of computing units to perform high-speed parallel processing, encryption and decryption can be processed without a time delay.

1 to 10 are diagrams for explaining a super-pixel image encryption apparatus according to an embodiment of the present invention.
11 is a view for explaining a super-pixel image encryption method according to an embodiment of the present invention.
12 is an exemplary screen encrypted by the super-pixel image encryption apparatus according to an embodiment of the present invention.

Since the present invention can have various changes and can have various embodiments, specific embodiments are illustrated in the drawings and will be described in detail through detailed description. However, this is not intended to limit the present invention to specific embodiments, and it should be understood to include all modifications, equivalents and substitutes included in the spirit and scope of the present invention. In describing the present invention, if it is determined that a detailed description of a related known technology may unnecessarily obscure the gist of the present invention, the detailed description thereof will be omitted. Also, as used herein and in the claims, the terms "a" and "a" and "a" are to be construed to mean "one or more" in general, unless stated otherwise.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings, and in the description with reference to the accompanying drawings, the same or corresponding components are given the same reference numerals, and the overlapping description thereof will be omitted. do it with

1 to 10 are diagrams for explaining a super-pixel image encryption apparatus according to an embodiment of the present invention.

Referring to FIG. 1 , the super pixel image encryption apparatus 10 includes an extractor 100 , an encryption unit 200 , an authentication key management unit 300 , a decryption unit 400 , and a transmission unit 500 .

The extractor 100 extracts a super pixel from the image.

Referring to FIG. 2 , a super pixel is a small image divided into uniform regions having similar features in order to process a large image at high speed. Fig. 2(a) is the original image, and Fig. 2(f), which has undergone the process of Figs. 2(b) to 2(e), is a super-pixel divided into regions having similar characteristics. A super pixel is a small uniform area made by grouping pixels with similar characteristics, and is different from a normal pixel having a grid structure. Super pixels are an efficient method for high-speed processing of large-capacity images.

A super pixel includes properties of minimizing compactness, region uniformity, boundary conformance with the original image, and under-segmentation regions.

The density is a measure of how simple the perimeter of a super pixel is, and can be expressed as in Equation 1 below.

C: density, P: perimeter of super pixel, S: area

Area uniformity is a measure of how uniform the characteristics of super pixels are.

Boundary coincidence with the original image is a measure of how much the superpixel and the boundary of the object in the original image match.

The minimization of the insufficient partition area can be refined by allocating the missing superpixels to adjacent superpixels through post-processing, since pixels that do not belong to any superpixels exist when superpixels are repeatedly implemented.

Superpixels are characterized by computational efficiency that simplifies complexity by subdividing them into uniform regions with similar characteristics. Superpixels have expressive efficiency by modeling long-distance behaviors between superpixels in an interactive pixel grid. Super pixel has conservation efficiency because it can preserve most of the image without loss due to over-segmentation.

Referring to FIG. 3 , when viewing an image 310 that is generally recognized by a human, an internal processing process 320 of the super-pixel image encryption apparatus 10 can be confirmed. To put it simply, the image of the part where a person is generally seen is updated after the 0th frame is initialized and then the block is updated and the pixel is updated. When moving to the first frame, there is no initialization process, and blocks and pixels are updated, and when blocks and pixels are updated while propagating to the second frame, the current frame is displayed. In this case, the super-pixel image encryption apparatus 10 processes an internal operation for updating blocks and pixels by applying a super-pixel algorithm. The super-pixel image encryption apparatus 10 enables natural and efficient image processing by updating pixels using a super-pixel algorithm and extracting a predetermined object.

Referring to FIG. 4 , the extractor 100 generates a sample from the same image or image sequence in order to extract a super pixel. The extractor 100 extracts an object from a plurality of samples generated from an image sequence. The extraction unit 100 efficiently extracts the super pixel by defining an objective function of the super pixel that prefers a uniform color. At this time, the objective function iteratively refines the partition by exchanging pixel blocks between superpixels to precisely extract the contour. When the value of the objective function approaches the optimal value, the extractor 100 divides and extracts a plurality of super pixels. In general, the outline (boundary) of the overlapping super pixels is not important, but the extraction unit 100 considers this and calculates it because it corresponds to the outline of the actually extracted object. The extractor 100 may measure a new and efficient object with natural image expansion through a tube composed of bounding boxes 410 spanning a time interval.

Referring to FIG. 5 , the super pixel algorithm over-segments an image in which pixels belonging to the same object are grouped, and the super pixel is increased. As the number of super pixels increases, the super pixels also need to be built with precision. Also, from the initial super-pixel partitioning, the boundary needs to be constantly modified. The extraction unit 100 defines a powerful and fast energy function by applying the color similarity between the boundary and the superpixel color histogram through the SEEDS (Superpixels Extracted via Energy-Driven Sampling) algorithm.

5, the same SEEDS segment of the same frame and different samples are combined, and an objectness score is determined from a superpixel (SEEDS) extracted through randomly extracted energy-based sampling. The objectivity score is a score that determines whether the existing boundary is damaged by calculating the sum of the distances to the general super-pixel boundary. The extraction unit 100 may determine whether to preserve or damage the boundary and the pixels therein with the objectivity score.

Equation 2 below is an equation for calculating the object point score.

O(i): represents the intersection of multiple superpixels, the value is 1 if all samples have a superpixel boundary at pixel i, otherwise 0

χ _R,C(p) : A pixel in the same column or Row as pixel p and within the bounding box

A: Normalize the score using the area of the bounding box

Per(χ) : set of pixels around the bounding box

d(p,i) : Distance to pixel i

The imaging approach to superpixels propagates superpixels over multiple frames to build 3D spatial and temporal structures. Over time new image superpixels may appear and end up with other image superpixels. The start and end of superpixels can be controlled by limiting the number of superpixel tubes (see Figure 4) in the sequence. Therefore, the extractor 100 uses the substitution constraint defined through two parameters. The constraint is the speed of superpixels and superpixels per frame. Superpixels per frame represents the number of superpixels every single frame is divided into, and superpixel rate is the rate at which superpixels are created/finished over time. The end of a super pixel that satisfies both constraints is the start of a new super pixel in the same frame, and successive super pixel generation is possible. The extraction unit 100 may extract continuous superpixels having the same color and no interruption through an energy function together with two constraints. The energy function of SEEDS, which increases color homogeneity within 3D superpixels, uses a color histogram of each superpixel. The maximum value having only one non-zero histogram section for each image superpixel is expressed by the following Equation (3).

: 영상의 유일한 파티션 집합(슈퍼 픽셀이 모든 프레임에 인접한 곳이며, 올바른 프레임 당 슈퍼 픽셀과 슈퍼 픽셀 속도 동작을 나타내는 파티션)

: the only set of partitions in the image (where superpixels are adjacent to every frame, partitions that exhibit correct superpixel and superpixel rate behavior)

: t 번째 프레임에서 슈퍼 픽셀 k가 속하는 픽셀들의 집합.

: A set of pixels to which super pixel k belongs in the t-th frame.

: t 번째 프레임까지 슈퍼 픽셀의 모든 픽셀들의 집합.

: the set of all pixels of the superpixel up to the t-th frame.

: SEEDS 에서 적용되는 에너지 함수와 유사하게 색상 히스토그램에서

를 나타냄.

: Similar to the energy function applied in SEEDS, in the color histogram

indicates.

: 히스토그램 구간(bin)에서 색상을 결정하는 색상 공간의 하위 집합.

: A subset of the color space that determines colors in a histogram bin.

Referring to FIG. 6 , the extraction unit 100 processes super pixels in hierarchical block units. Pixel exchanges between superpixels and their temporal propagation are regulated through blocks of pixels. The SEDDS algorithm started by dividing the image into a normalized grid of blocks, and unlike other algorithms, it considers a hierarchy of blocks of different sizes. Starting from the pixels of layer 1 610 as the most detailed scale, 2X2 or 3X3 pixel blocks are formed on layer 2 620 . Layer 3 630 combines the 2X2 blocks of the previous layer 2 620 . The pixel block size (2X2 or 3X3) and the number of layers in layer 2 620 are chosen such that a prescribed number of super pixels per frame are generated in the best layer 640 which is a super pixel layer. That is, a layer such as layer 3 630 may be generated until a prescribed number of super pixels are generated per frame. In other words, the extraction unit 100 starts by pixel units in the first layer per frame, groups in blocks in the next layer, and finds increasingly detailed and sophisticated super pixels to generate them.

Referring to FIG. 7 , the encryption unit 200 encrypts the super pixel generated by the extraction unit 100 . For example, the encryption unit 200 encrypts the super pixel from which the boundary area of the super pixel is calculated in units of frames. The encryption unit 200 hex-converts the super pixel. The encryption unit 200 encrypts the converted super pixel using a SEED encryption method. The encryption unit 200 parallelly encrypts the frame from which the super pixel is calculated by using a plurality of computing devices. The encryption unit 200 encrypts all packets of video/image information in units of frames in which super pixels are calculated based on a SEED encryption algorithm, which is a block encryption algorithm. The super-pixel image encryption device 10 has the security that the image/video photographed through the camera is pre-encrypted before transmission and cannot reproduce the image from the outside even if the image information is leaked. The encryption unit 200 has the frame data encrypted through the SEED encryption algorithm in a matrix structure.

Referring to FIG. 8 , since the encryption unit 200 performs encryption in parallel using the operation cores of at least four CPUs, the processing speed is high, so that image information photographed and output in real time can be pre-encrypted and then transmitted or stored. have.

Referring to FIG. 9 , the authentication key management unit 300 manages the authentication key to provide images/videos only to authorized user terminals. The authentication key management unit 300 generates an authentication key using one or more of customer number, USIM information, TIME STAMP information, GPS coordinates, and MAC Address, and registers permitted user terminal information.

The authentication key management unit 300 generates and manages a reproduction request for image/video information from one or more user terminals and an authentication key for decoding the image/video information. It acquires information in the OTP method using one or more of customer number, USIM information, TIME STAMP information, GPS coordinates, and MAC Address. If the obtained information matches the pre-stored authentication key, the permitted user terminal is determined.

Referring to FIG. 10 , the decryption unit 400 decrypts an encrypted image/video when a reproduction request occurs from the user terminal for which the authentication key is confirmed. The decryption unit 400 decrypts the packet information of the image information through the SEED encryption algorithm in the encryption unit 200 in a manner corresponding to the encryption.

The transmitter 500 transmits the decoded image/video so that the decoded image/video is reproduced through a decryption player executed on the user terminal. Decryption Player is a super pixel dedicated player that decrypts the encrypted frame and converts the superpixel in the decrypted frame according to the standard codec.

11 is a diagram for explaining a super-pixel image encryption method according to an embodiment of the present invention.

Referring to FIG. 11 , the super pixel image encryption apparatus 10 extracts a super pixel from an original image in step S1110.

In step S1120, the super-pixel image encryption apparatus 10 encrypts the extracted super-pixel units. The super pixel image encryption apparatus 10 encrypts the super pixel using a SEED encryption algorithm. Since the superpixel image encryption apparatus 10 performs encryption in parallel using at least four CPU operation cores, the processing speed is high, so that image information photographed and output in real time can be pre-encrypted and then transmitted or stored. The super-pixel image encryption device 10 has security that the image/video photographed through the camera is pre-encrypted before transmission and cannot reproduce the image from the outside even if the image information is leaked.

In step S1130, the super-pixel image encryption apparatus 10 determines whether one or more user terminals are permitted. When the superpixel image encryption apparatus 10 receives the reproduction request or image transmission request information from the user terminal, it is determined whether the user terminal is permitted. The super pixel image encryption device 10 obtains one or more of customer number, USIM information, TIME STAMP information, GPS coordinates, and MAC address from the user terminal in an OTP method. The super-pixel image encryption apparatus 10 determines whether the acquired information is an authorized user terminal by checking whether the obtained information matches the previously stored user terminal information. The super pixel image encryption apparatus 10 performs step S1140 only in the case of an authorized user terminal.

In step S1140, the super pixel image encryption apparatus 10 decrypts the encrypted super pixel according to the encrypted method, and transmits it to the authorized user terminal to be reproduced through the decryption player.

12 and 13 are exemplary screens of a super-pixel image encryption apparatus according to an embodiment of the present invention.

Referring to FIG. 12 , it is difficult to check the corresponding image even if the image is leaked because the super-pixel image encryption apparatus 10 extracts and encrypts the super-pixel from the original image.

Referring to FIG. 13 , the super pixel image encryption apparatus 10 extracts super pixels in units of frames. The super-pixel image encryption apparatus 10 extracts the original image per frame, copies the area to be calculated, and extracts the super-pixel by changing the color. Superpixels have expressive efficiency by modeling long-distance behaviors between superpixels in an interactive pixel grid. The super-pixel image encryption apparatus 10 can preserve most of the image without loss by over-segmenting (over-segmenting) when extracting the super pixel, thereby having a conservation efficiency.

The above-described super-pixel image encryption method may be implemented as a computer-readable code on a computer-readable medium. The computer-readable recording medium may be, for example, a removable recording medium (CD, DVD, Blu-ray disk, USB storage device, removable hard disk) or a fixed recording medium (ROM, RAM, computer-equipped hard disk). can The computer program recorded in the computer-readable recording medium may be transmitted to another computing device through a network, such as the Internet, and installed in the other computing device, thereby being used in the other computing device.

In the above, even though it has been described that all components constituting the embodiment of the present invention are combined or operated as one, the present invention is not necessarily limited to this embodiment. That is, within the scope of the object of the present invention, all the components may operate by selectively combining one or more.

Although acts are shown in a specific order in the drawings, it should not be understood that the acts must be performed in the specific order or sequential order shown, or that all shown acts must be performed to obtain a desired result. In certain circumstances, multitasking and parallel processing may be advantageous. Moreover, the separation of the various components in the embodiments described above should not be construed as necessarily requiring such separation, and the program components and systems described may generally be integrated together into a single software product or packaged into multiple software products. It should be understood that there is

So far, the present invention has been looked at focusing on the embodiments thereof. Those of ordinary skill in the art to which the present invention pertains will understand that the present invention can be implemented in a modified form without departing from the essential characteristics of the present invention. Therefore, the disclosed embodiments should be considered in an illustrative rather than a restrictive sense. The scope of the present invention is indicated in the claims rather than the foregoing description, and all differences within the scope equivalent thereto should be construed as being included in the present invention.

10: Super pixel video encryption device
100: extraction unit
200: encryption unit
300: authentication key management unit
400: decryption unit
500: transmission unit

Claims (9)

In the super pixel image encryption device,
an extractor for extracting super pixels from an image;
an encryption unit for encrypting the image;
a decryption unit for decrypting the encrypted image;
an authentication key management unit for determining an authorized user terminal; and
and a transmission unit for transmitting the decrypted image.
According to claim 1,
the encryption unit
Super pixel image encryption device that always encrypts in units of super pixels.
According to claim 1,
The authentication key management unit
A super pixel image encryption device that acquires one or more of customer number, USIM information, TIME STAMP information, GPS coordinates, and MAC address from a user terminal as a one-time password (OTP).
4. The method of claim 3,
The authentication key management unit
A super pixel image encryption apparatus for determining whether the user terminal is an authorized terminal by checking whether the one-time encryption and stored authentication key information match.
According to claim 1,
A super-pixel image encryption device that performs encryption and decryption operations in parallel through a computing device with four or more computational cores.
In the super pixel image encryption method,
extracting super pixels from the image;
encrypting the image;
determining an authorized user terminal;
decoding the image; and
Super pixel image encryption method comprising the step of transmitting the image.
7. The method of claim 6,
The step of encrypting the video is
A super pixel image encryption method for encrypting the super pixel unit.
7. The method of claim 6,
The step of determining the permitted user terminal is
If one or more of customer number, USIM information, TIME STAMP information, GPS coordinates, and MAC address in the user terminal matches the stored authentication key information obtained as a one-time password (OTP), Super pixel image encryption method that is determined as an authorized user terminal.
[Claim 9] A computer program recorded in a computer-readable recording medium executing any one of the super pixel image encryption methods according to claim 6 to 8.

Images