KR20020018988A - Network camera apparatus, network camera server and digital video recorder for preventing forgery and alteration of an digital image, and apparatus for authenticating the digital image from said apparatus, and method thereof - Google Patents

Abstract

PURPOSE: A network camera system for preventing forgery of a digital image, a network camera server and a digital video storage, a system and a method for authenticating digital images output from the systems are provided to prevent illegal operations of images that are photographed for monitoring. CONSTITUTION: A network camera system includes a video input part(20) for receiving a video signal photographed in real time, a video data processor(30) for converting the video signal into a digital signal, and an information generator(90,92) for generating information to be inserted as a watermark. The system further has a watermark generator(94) for generating a watermark using the information, and a watermark insertion part(40) for inserting the watermark into the video signal output from the video data processor. The system also has a video compressor(50) for compressing the video signal into which the watermark was inserted, and a network connector(70) for transmitting the compressed video signal to a network.

Description

Network camera device, network camera server and digital video storage device for preventing forgery of digital video, device for authenticating digital video output from these devices and method thereof AND ALTERATION OF AN DIGITAL IMAGE, AND APPARATUS FOR AUTHENTICATING THE DIGITAL IMAGE FROM SAID APPARATUS, AND METHOD THEREOF}

The present invention inserts a strong watermark and a soft watermark into an image input through a camera in real time to insert a watermark into a network camera device capable of preventing forgery of a digital image, a network camera server, and a photographed and transmitted image. The present invention relates to a digital video storage device capable of authenticating, and a device and a method for authenticating original and forgery of an image output from these devices.

Such image capturing and transmitting cameras and related devices are commonly used in security systems. As such, a system that receives an image through a camera and transmits the image so that the displayed image can be viewed.

1. General closed circuit camera (CCTV) system

2. Digital CCTV System (DVR: Digital Video Recorder)

3. Network camera (web camera) system

4. System using a USB camera

Etc. are typical ones.

Such image transmission cameras transmit the photographed images to be viewed by the general public (or authorized persons) using a web browser and other monitors at a near or remote location. Among these, network cameras are often called "web cameras" or "internet cameras" because they include a web server function.

In another form of the network camera, the network camera server receives a video signal from a plurality of cameras each consisting of a lens and an image sensor that are separated in an external location, converts the video signal into a single integrated video signal, and transmits the same to the network. It performs a network server function on video signals captured from multiple cameras.

Such a network camera or a network camera server has its own unique IP and uses a standard web browser to capture the acquired image using a standard web browser, JPEG, M-JPEG, wavelet or MPEG compression. It has the function of transmitting to network at high speed of minimum 10 frames per second and maximum 30 frames per second.

Meanwhile, DVR (Digital Video Recorder), a digital CCTV system, is widely used as a next-generation surveillance system that replaces a surveillance system composed of conventional CCD cameras, VCRs, TAPEs, etc., and analog surveillance systems monitor the environment to be monitored. In addition, the DVR recorded and searched and stored the required video data in TAPE, but the DVR converts the captured video data into digital signals and stores them on a hard disk or DVD-RAM. . In addition, a single DVR can record and manage a plurality of cameras (for example, 16 cameras) and a plurality of images (16 divided images).

For reference, the network referred to in the present invention is an ISDN, a connection through a dedicated line, a connection through a LAN, a connection through a telephone network (PSTN, PSDN), a connection through a WAN, in particular through the Internet, TCP / IP (protocol) Includes all commonly used "network" connections, such as remote video retrieval.

However, the video signals photographed by the cameras by recognizing the object are tended to be photographed using a digital camera device due to the recent rapid development of digital imaging technology, such as a DVR. In the case of a DVR, the captured video is stored in digital form. Users also tend to use digital video for processing. In the case of analog video, it is digitized using an A / D converter.

However, in accordance with this trend of digitization, the user can manipulate the digital image captured by a general image editor as desired, so that the network cameras or DVRs are applied for surveillance requiring the same as the original image originally captured. The problem arises that there is a limit.

In other words, the forgery and editing are very easy due to the characteristics of the digital image. Therefore, the question of how to confirm that the image taken and transmitted is the effect of the evidence and the original image is a problem.

In addition, if the image is forged while authenticating the original of the digital image, the digital image is required to have a legal effect by accurately detecting the position of the forged image.

SUMMARY OF THE INVENTION The present invention has been made to solve the above problems, and an object of the present invention is to provide a network camera, a network camera server, and a digital video storage device (DVR) having a function of embedding a watermark, in particular, for surveillance. In addition to preventing illegal video manipulation that can be applied to the captured image, it is possible to detect when such surveillance video is manipulated.

Another object of the present invention is to insert a strong watermark and a soft watermark in the image to be photographed, and to verify whether the image is forgery and the position of the forgery, as well as to verify the original status of the image with information about the operator and the image of the image You can authenticate until.

1 is a block diagram illustrating a configuration of a network camera apparatus for inserting a strong watermark and a soft watermark into an input image and transmitting the same to an input image according to a first embodiment of the present invention.

2A is a block diagram illustrating a configuration of an apparatus for authenticating original and forgery of an image transmitted from a network camera apparatus of FIG. 1 through a network;

FIG. 2B is a flowchart illustrating a process of authenticating an original of an image by extracting a strong watermark in the apparatus of FIG. 2A;

FIG. 2C is a flowchart illustrating a process of authenticating an image forgery by extracting a soft watermark in the apparatus of FIG. 2A.

FIG. 3 is a block diagram illustrating a configuration of a network camera server for inserting a strong watermark and a soft watermark into a plurality of images output from a plurality of image input apparatuses and transmitting them to a network according to a second embodiment of the present invention.

FIG. 4 is a block diagram illustrating a configuration of an apparatus for authenticating original and forgery of an image transmitted from a network camera server of FIG. 3 through a network.

5A illustrates a watermark embedding apparatus for inserting a strong watermark and a soft watermark into a plurality of images inputted from a plurality of image input apparatuses, and recording and storing a watermark-embedded image connected thereto according to a third embodiment of the present invention. Is a block diagram showing the configuration of a digital video storage device,

5B illustrates a watermark embedding apparatus for inserting a strong watermark and a soft watermark into a plurality of images inputted from a plurality of image input devices according to a modification of the third embodiment of the present invention, and a watermark embedded image connected thereto; It is a block diagram showing the configuration of a digital video storage device for recording and storing.

FIG. 6 is a block diagram illustrating a configuration of an apparatus for authenticating an original and a forgery of an image output from the digital video storage device of FIGS. 5A and 5B.

FIG. 7A is a block diagram illustrating a configuration of a watermark embedding apparatus separately connected to each of a plurality of image input apparatuses and a digital video storage apparatus for recording and storing a watermark embedded image connected thereto for each of a plurality of image input apparatuses according to the fourth embodiment of the present invention. Degree,

FIG. 7B is a block diagram showing the detailed configuration of the watermark embedding apparatus of FIG. 7A.

FIG. 8 is a block diagram illustrating a configuration of an apparatus for authenticating original and forgery of an image output from the digital video storage device of FIG. 7A.

Explanation of symbols on the main parts of the drawings

10: network camera device 20: video input unit

30: image data processing unit 40: RW insertion unit

50: image compression unit 60: FW insertion unit

70: network connection unit 80: real time operation unit

90: fixed information generation unit 92: random information generation unit

94: watermark generation unit 100: authentication device

110: image input unit 120: image storage unit

130: compressed image expansion unit 140: image authentication unit

142: RW authentication unit 144: FW authentication unit

150: image authentication result output unit 400: network camera server

410: a plurality of image input apparatus 420: a plurality of image data processing unit

430: a plurality of RW insertion portion 450: a plurality of FW insertion portion

460: video signal integration unit 500: authentication device

520: Image signal separation unit 530: A plurality of image storage unit

540: video signal selection unit

610, 710: watermark inserting device

630, 730: Digital Video Storage (DVR)

In order to achieve the above object, the present invention provides an image input unit for receiving an image signal captured in real time; An image data processor converting an image signal output from the image input unit into a digital signal; An information generating unit for generating information to be inserted as a watermark; A watermark generator for generating a watermark using information of the information generator; A first watermark inserting unit inserting a first watermark generated by the watermark generating unit into an image signal output from the image data processing unit; An image compression unit to compress an image signal to which the first watermark inserted from the first watermark insertion unit is inserted; A second watermark inserting unit inserting a second watermark generated by the watermark generating unit into the compressed image signal output from the image compressing unit; And a network connection unit configured to transmit an image signal including the second watermark embedded from the second watermark insertion unit to a network.

In the network camera device, the information generating unit comprises a fixed information generating unit for generating unique information of the network camera device; And an arbitrary information generating unit for generating information transmitted from a remote location via a network, wherein the first watermark inserted into the video signal by the first watermark inserting unit is a robust watermark, and a second watermark embedding. Preferably, the second watermark embedded in the video signal is a soft watermark.

In addition, the network camera device preferably further comprises a real-time operating unit for controlling the insertion of the watermark, the compression of the video signal, and the generation of arbitrary information in real time.

According to another aspect of the present invention, there is provided an apparatus for authenticating a watermark-embedded image transmitted from a network camera device, the apparatus comprising: an image input unit for receiving a watermark-embedded video signal through a network; A compressed image extension unit for restoring an image signal output from the image input unit to an image signal before compression; An image authentication unit for determining authentication of an image by calculating a correlation between the watermark extracted from the image signal reconstructed by the compressed image extension unit and the watermark generated from the image authentication information; And an image authentication result output unit configured to output an authentication result of the image authentication unit.

Here, the image authentication unit includes a strong watermark authentication unit for authenticating the original of the image by detecting the strong watermark; And a weak watermark authentication unit for detecting whether the image is forged and the forged position by detecting the weak watermark.

In order to achieve another object of the present invention, the present invention provides a method for inserting a watermark on a video signal photographed through a network camera device and transmitting it to a network, the method comprising the steps of: converting a video signal input in real time into a digital signal; ; Inserting a strong watermark containing unique information of the network camera device into the converted video signal in real time; Compressing a video signal having a strong watermark embedded therein; Inserting a soft watermark containing random information transmitted through a network into a compressed video signal in real time; And transmitting the video signal having the second watermark embedded therein to the network.

In order to achieve another object of the present invention, the present invention includes a plurality of image data processing unit for converting each of the plurality of image signals input in real time from the plurality of cameras into a digital signal; An information generator for generating information to be inserted as a watermark corresponding to each of the plurality of video signals; A watermark generator for generating respective watermarks corresponding to each of the plurality of video signals using information of the information generator; A plurality of first watermark inserting units respectively inserting first watermarks generated by the watermark generating unit into each of the plurality of image signals output from the plurality of image data processing units; An image compressor which compresses a plurality of image signals into which watermarks output from the plurality of first watermark inserters are inserted; A plurality of second watermark inserting units respectively inserting second watermarks generated by the watermark generating unit into each of the plurality of image signals output from the image compressing unit; An image signal integrating unit for integrating a plurality of image signals embedded with watermarks output from the plurality of second watermark embedding units into one image signal; And it provides a network camera server including a network connection unit for transmitting the integrated video signal output from the video signal integration unit to the network.

In order to achieve another object of the present invention, the present invention is an apparatus for authenticating a watermark-embedded image transmitted from a network camera server, the image that accepts a video signal inserted in the compressed state and integrated in a compressed state over the network An input unit; An image signal separation unit for dividing the integrated image signal output from the image input unit into a video signal corresponding to each camera; A plurality of image storage units respectively storing the image signals separated by the image signal separator; An image signal selection unit for selecting an image signal requiring authentication from among the image signals stored in the image storage unit; A compressed image extending unit which restores the image signal selected by the image signal selecting unit to the image signal before compression; An image authentication unit for determining authentication of an image by calculating a correlation between the watermark extracted from the image signal reconstructed by the compressed image extension unit and the watermark generated from the image authentication information; And an image authentication result output unit configured to output an authentication result of the image authentication unit.

In order to achieve another object of the present invention, the present invention provides a method of inserting a watermark into a plurality of video signals input from a plurality of cameras and transmitting the watermark to a network, wherein the plurality of cameras are inputted in real time from the plurality of cameras. Converting an image signal into a digital signal; Inserting, in each of the plurality of converted video signals, a robust watermark containing unique information of a plurality of cameras or a network camera server in real time; Compressing a plurality of video signals each having a robust watermark embedded therein; Inserting in each of a plurality of compressed video signals a soft watermark containing arbitrary information transmitted through a network in real time; Incorporating a plurality of video signals including a soft watermark into one video signal; And transmitting the integrated video signal to a network.

In order to achieve another object of the present invention, the present invention is a digital video storage device (DVR) including a watermark embedding device for embedding a watermark in a plurality of video signals input from a plurality of cameras, watermark embedding The apparatus includes a plurality of image data processing units for converting each of the plurality of image signals input in real time from the plurality of cameras into digital signals; An information generator for generating information to be inserted as a watermark corresponding to each of the plurality of video signals; A watermark generator for generating respective watermarks corresponding to each of the plurality of video signals using information of the information generator; A plurality of first watermark inserting units respectively inserting first watermarks generated by the watermark generating unit into each of the plurality of image signals output from the plurality of image data processing units; And a video signal integrator for integrating a plurality of video signals embedded with watermarks output from the plurality of first watermark inserters into a single video signal, wherein the video signal output from the video signal integrator is compressed and recorded. It provides a digital video storage device.

Here, the watermark embedding apparatus may include a plurality of second watermark embedding units which respectively embed a second watermark generated by the watermark generator in each of a plurality of image signals output from the plurality of first watermark embedding units. It is preferable to further include.

In addition, the watermark embedding apparatus is preferably integrated into the digital video storage device as a hardware or software module.

In order to achieve another object of the present invention, the present invention is a device for authenticating a watermark embedded image from a digital video device, the transmission or file of the video signal embedded in the watermark and integrated into one signal over a network An image input unit for receiving in the form of; An image signal separation unit for dividing an image signal integrated into one signal output from the image input unit into an image signal corresponding to each camera; An image authentication unit for determining the authentication of the image by calculating a correlation between the watermark extracted from the image signal from the image signal separation unit and the watermark generated from the image authentication information; And an image authentication result output unit configured to output an authentication result of the image authentication unit.

In order to achieve another object of the present invention, the present invention provides a method of inserting a watermark into a plurality of video signals input from a plurality of cameras and recording and storing the watermark, wherein the plurality of cameras are inputted in real time from the plurality of cameras. Converting an image signal into a digital signal; Inserting, respectively, a robust watermark containing unique information of the plurality of cameras into each of the plurality of converted video signals in real time; Inserting, in real time, a weak watermark containing arbitrary information transmitted through a network into each of a plurality of video signals each having a strong watermark embedded therein; Incorporating a plurality of video signals including a soft watermark into one video signal; And compressing and recording and storing the integrated video signal.

In order to achieve another object of the present invention, the present invention provides a digital video storage device (DVR) for recording and storing a plurality of video signals input from a plurality of cameras, a plurality of watermark embedding apparatus is provided in each of the plurality of cameras Separately installed, each of the plurality of watermark embedding apparatus comprises an image data processing unit for converting the image signal input in real time from the corresponding camera into a digital signal; An information generating unit for generating information to be inserted as a watermark; A watermark generator for generating a watermark using information of the information generator; And a first watermark inserting unit inserting a first watermark generated by the watermark generating unit into an image signal output from the image data processing unit, wherein the video signal output from the first watermark inserting unit is compressed and recorded and stored. It provides a digital video storage device.

Here, each of the plurality of watermark embedding apparatuses may further include a second watermark embedding portion which embeds a second watermark generated by the watermark generating portion into an image signal output from the first watermark embedding portion. .

In order to achieve another object of the present invention, the present invention provides an apparatus for authenticating a watermark embedded image from a digital video apparatus, comprising: an image input unit for receiving an image signal in a form of a transmission or a file through a network; An image authentication unit which determines authentication of an image by calculating a correlation between a watermark extracted from an image signal output from the image input unit and a watermark generated from image authentication information; And an image authentication result output unit configured to output an authentication result of the image authentication unit.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

First, with reference to Figs. 1 and 2, an embodiment in which the technique of overlapping the watermark of the present invention is applied to a network camera apparatus will be described.

FIG. 1 is a block diagram showing the configuration of a network camera device 10 for inserting a strong watermark and a soft watermark into an input image and transmitting it to a network according to the first embodiment of the present invention.

Referring to FIG. 1, the network camera device 10 according to the first embodiment of the present invention includes an image input unit 20, an image data processing unit 30, a fixed information generation unit 90, and an arbitrary information generation unit 92. ), A watermark generator 94, a robust watermark inserter 40 (hereinafter referred to as "RW (Robust Watermark) inserter"), a soft watermark inserter 60 (hereinafter referred to as "FW (Fragile Watermark)) Insertion unit ", the image compression unit 50, the network connection unit 70 and the real-time operation (80), the above components are hardware or software in a single device called the network camera device (10) These are included.

The image input unit 20 means a lens, an image sensor, or the like, and receives (ie, photographs) an image signal in real time. The image data processing unit 30 converts the image signal into a digital signal after amplifying and correcting the image signal of the image input unit 20 so as to be suitable for processing.

The fixed information generation unit 90 generates information to be inserted, which is stored in advance in the network camera device 10 itself, and the arbitrary information generation unit 92 stores the information to be inserted, which is transmitted from a remote place via a network, in real time. Generated by the control signal formed at 80.

That is, the fixed information generation unit 90 generates information, which is fixed and stored in the network camera itself as watermark embedding information, such as a serial number or a unique image of the network camera. The random information generator 92 generates key generation information or unique image generation information transmitted from a remote location. When the operator inputs information to be inserted through a dedicated browser at a remote location, the real-time operation unit 80 forms a control signal. By supplying it to the arbitrary information generating unit 92, it is possible to perform the processing according to the operator's instructions.

The fixed information or arbitrary information may generate information in the form of key generation or image or in the form of simultaneously employing the key generation and the image.

In the embodiment of the present invention, although the fixed information generation unit 90 and the arbitrary information generation unit 92 are described in a form employing both, the fixed information generation unit 92 and the arbitrary information generation unit 92 are required. It is also possible to carry out either one independently.

The watermark generator 94 generates a watermark by using the information generated by the fixed information generator 90 and the arbitrary information generator 92. In the embodiment of the present invention, the watermark is generated by a key generation method. (Hereinafter referred to as " key watermark ") and generating an image watermark using discrete wavelet transform or discrete cosine transform, the watermarking method for a watermark to be inserted into an image is not limited thereto. Of course, all watermarking techniques can be used.

First, generation of a key watermark depends on a method of generating a watermark signal by a pattern, and the generated watermark signal is inserted and extracted as a form of a pattern, and the generation of such a pattern is a pseudo random number. (Pseudo Random Number) Generate function.

That is, a watermark can be said to have a certain key value and be inserted in the form of a pseudo random number generated from it. More specifically, if a random number generated from a key is Rc, It is in the form of some binary code. In general, Rc∈ {1, -1} and (1, 1, -1, 1, 1, -1, 1, -1 ???). The reason why the binary code generated here is not 1 and 0 but 1 and -1 is only 1 and 0. When Rc is added as a watermark to the image data, the value is 1 and 0 continuously. Because it results in an increase.

However, when the binary codes are composed of 1 and -1, it is possible to ensure that the energy value does not change even if they are added. Codes that consist of 1 and -1 are usually distributed so that they have a normal distribution with a mean of 0 and a variance of 1. Here, the "key" for generating a watermark is literally information that is a key necessary for inserting and extracting a watermark, which may be a key with a serial number assigned to each network camera ( Fixed information), and information (arbitrary information) transmitted through a network from a remote location may be used, or both may be used.

Next, the image watermark is generated by inserting the image information itself as a watermark, and it can be said that a digital image itself, such as a user's own signature or a company logo, is inserted into the video as a watermark. Similar to a key watermark, an image watermark has a unique image data stored in a network camera (fixed information), and there is a method of generating it as a watermark. Also, image data (arbitrary information) can be generated from a remote location via a network. There is a method of receiving a message and generating it as a watermark, and there is also a method of using both of the above.

As mentioned above, the watermarking technique used as the core of the present invention can apply all the watermarking techniques previously proposed or published, for example, by analyzing data such as images in terms of spatial aspects. A spatial domain method that scatters the information to be inserted in space so that it cannot be easily distinguished.In (ai, bi) where n pairs are randomly selected in the image, ai adds contrast values one by one and bi subtracts contrast values one by one. A patchwork method that allows digital watermarks to be inserted into the space by zooming, and to improve the disadvantage of the patchwork method degrading the visual quality of the image, the image is divided into two equally sized subsets. The amount computed as the variance of two subsets for pixels in one subset of them To compensate for the drawbacks of the method proposed by Pitas and Kaskalis by adding the integer k-factor, which is impossible to apply to color images with a large amount of data, pixels To reduce the amount of computation in the unit, divide the image into blocks of N and insert a bit stream into the luminance value of each image block. If the average value of the pixels in the block is greater than the threshold, the value is '1'. If it is encoded by using and is smaller than the threshold value, the Caronni method such as encoding by '0' may be applied.

In addition, as a spread spectrum method in the spatial domain, all pixels of an image are basically divided into A, B, and C sets according to the condition of | A | = | B |, which is a pseudo random number. It is made up of a pseudo-random number generator and a secret key value. In a black and white image, when matrix A increases gray level by k, matrix coefficients of B decrease. Therefore, the C matrix value does not change. That is, since the A and B matrices have basically the same distribution, the average value of serialization of the image is not changed, and the detection of the watermark is equal to the two image coefficient sets A and B for gray levels of the black and white image. . In addition, if it is possible to separate A '= A and B' = B by k, a method of detecting a watermark may be applied thereto.

In addition, a frequency domain method of converting multimedia data into analog signals of frequency components and converting and inserting watermarks to be inserted into analog signals in the same manner, by applying discrete cosine transform, fast Fourier transform, discrete wavelet transform, etc. How to create, etc. can be applied.

The watermark generated by the key generation method and / or the image method is inserted into the acquired digital image. The watermark embedding method is a robust watermark method according to the embedding strength of the watermark. RW-method "and soft watermark (hereinafter referred to as" FW-method ").

That is, both the key watermark and the image watermark are divided into a strong watermark method that survives an external attack and a soft watermark method that breaks even a very fine attack according to the strength of inserting the watermark. The key generation method and the image method can both be robust or fragmented.

In short, a robust watermark means that the image is inserted in such a way that it can be extracted without being damaged by any transformation of the image externally, that is, compression or filtering. It can be said to have a stamp that is thick and invisible so as not to be erased by any manipulation.

On the other hand, a soft watermark is a weak insertion of a watermark signal so that the inserted watermark can be broken even in the case of deformation or processing of a very fine image, and it is determined whether the watermark is damaged or not. Because it can be determined, it can be used as a method to detect and detect forgery of an image.

In summary, in the embodiment of the present invention, a key watermark or an image watermark is generated, and the generated watermark is inserted into the image step by step in the RW-FW method. The key watermark and the image watermark may be used one by one independently or in various combinations according to the purpose. In addition, the RW-method and the FW-method may be independently performed according to the purpose, and the two methods may be used together.

The embodiment to be mainly described in the present invention uses a FW-method capable of detecting a forgery of an image in inserting a watermark, and also uses the RW-method in combination with the watermark. The important data to be inserted is inserted in the RW-method, and the Fragile watermark is also inserted in order to prevent forgery of the image.

The RW inserter 40 inserts the first watermark of the watermark generator 94 as a robust watermark into the digital video signal output to the video data processor 30, and generates and inserts the watermark into the video. Later, the information must be strongly inserted in order to survive without any external attack. The RW scheme is described in detail in the same pending Korean Patent Application Nos. 10-2000-0053755, 10-1998-0037273 and 10-1998-0037274 to the applicant of the present invention.

First, a detailed description of embedding a key watermark into a digital video signal as a robust watermark is disclosed in pending Korean Patent Application No. 10-2000-0053755 of the same applicant as the applicant of the present invention. By using the technology of the patent application of the present invention, the key watermark is inserted in the RW method. Even if the watermark-embedded image undergoes image transformation such as dithering or halftone, the inserted watermark can be detected well. In addition, since the watermark can be detected without the original image, it is suitable for demanding a watermark resistant to deformation by any external attack for fast insertion and quick detection or preservation of the original, such as a network camera.

Next, a detailed method for inserting an image watermark, such as a specific logo of an operator, into a digital video signal as a robust watermark also uses an earlier patent.

This is a method using Discrete Wavelet Transform (DWT) and Discrete Cosine Transform (DCT) when inserting an image watermark into an image. In particular, this method can provide a watermarking method that is not damaged even in JPEG, MPEG, etc., which are currently used for video compression. Of course, both color and black and white images are possible. Detailed description thereof is described in the same pending Korean patent application No. 10-1998-0037273 and Korean Patent Application No. 10-1998-0037274 as applicants of the present invention.

The above algorithms for embedding a watermark may be implemented as one chip and mounted in the network camera apparatus 10.

The video compression unit 50 compresses the digital video signal having the watermark embedded therein in the RW insertion unit 40 in correspondence with each compression scheme such as JPEG, Wavelet, and MPEG so as to be transmitted to the network.

The FW inserter 60 inserts the second watermark generated by the watermark generator 94 as a soft watermark into the digital video signal compressed by the image compressor 50, and spatially inserts pixel values of the original image. By properly adjusting in the area and inserting a watermark in the adjusted image data, it is possible to accurately detect the position of the forged data in the original image. The detailed description thereof is in the same pending Korean as the applicant of the present invention. Patent Application No. 10-2000-0064767.

The insertion of the watermark can be arbitrarily set and determined according to the operator's selection and the security purpose or level of the image. As in the embodiment of the present invention, both the RW inserter 40 and the FW inserter 60 are employed. In addition, only one of the RW inserting portion 40 and the FW inserting portion 60 may be employed.

The network connection unit 70 transmits the digital video signal including the soft watermark inserted in the FW inserting unit 60 to the network, and also transmits information to be generated in the arbitrary information generating unit 132, which is transmitted from the remote place via the network. In addition, it functions to transmit to the real-time operating unit 80.

Since the real-time operating unit 80 inputs and processes an image from 10 frames to 30 frames per second, the image compression unit 50 may control a watermark processing for this in real time and transmit a digital image having a watermark embedded thereon in a network. ) Controls the video signal to be compressed according to each compression method such as JPEG, Wavelet, MPEG, etc., and controls the information to be inserted into the watermark received from the network through the network connection unit 70 to the arbitrary information generation unit 92. send.

The operation of the network camera device 10 of the present invention described above is as follows.

The image signal input through the image input unit 20 composed of a lens and an image sensor is amplified and corrected by the image data processing unit 30 so as to be suitable for processing, and then converted into a digital signal and sent to the RW inserting unit 40. Lose.

On the other hand, the fixed information generation unit 90 generates fixed information fixedly stored in the network camera itself, such as a camera serial number or a unique image to be inserted as a watermark, and the arbitrary information generation unit 92 tries to insert the fixed information. When the information is input through a dedicated browser at a remote location, the real-time operating unit 80 forms a control signal to generate random information.

After the information to be inserted into the watermark is generated by the fixed and arbitrary information generating units 90 and 92, the watermark generating unit 94 generates a watermark using these pieces of information.

The watermark generated by the watermark generator 94 is inserted in real time as a watermark that is robust to the digital image by the RW inserter 40. Since RW can survive compression, it is inserted before the video compression process.

The digital image embedded with the strong watermark is compressed by the image compression unit 50 to correspond to each compression scheme such as JPEG, Wavelet, MPEG, etc. to be transmitted over a network, and the digital image compressed by the soft watermark by the FW insertion unit 60. It is inserted into the signal in real time.

After the insertion of the soft watermark, the image is transmitted to the network through the network connection unit 70 to provide a digital video signal capable of authenticating the original and / or forgery of the provided image as well as providing the original surveillance image. to provide.

Next, an apparatus and an authentication process for authenticating a watermark-embedded image transmitted from the network camera apparatus 10 will be described. 2A is a block diagram illustrating a configuration of an apparatus 100 for authenticating original and forgery of an image transmitted from a network camera apparatus of FIG. 1 through a network.

Referring to FIG. 2A, the authentication apparatus 100 includes an image input unit 110, an image storage unit 120, a compressed image extension unit 130, an RW authentication unit 142, and an FW authentication unit 144. The authentication unit 140 and the image authentication result output unit 150, the authentication device according to an embodiment of the present invention can be implemented in a window environment of a personal computer for the user's convenience.

In the image input unit 110, a digital image (image to be authenticated) inserted with a watermark transmitted from the network camera device 10 of FIG. 1 via a network is input through a network in a compressed state.

The video storage unit 120 stores the video signal of the video input unit 110 in a compressed state. The compressed image decompressor 130 decompresses the image signal stored in the image storage 120 to restore the image signal to the image signal before compression.

Image authentication unit 140 is composed of a RW authentication unit 142 for detecting a strong watermark to authenticate the original image and a FW authentication unit 144 for detecting whether or not forgery of the image by detecting a weak watermark, compression The image authentication is determined by calculating a correlation between the watermark extracted from the reconstructed image signal of the image extension unit 130 and the watermark generated from the image authentication information.

The RW authentication unit 142 calculates a correlation between the extracted strong watermark and the strong watermark generated for image authentication, and if there is a deformation in the watermark, the result is a forged image. Is transmitted to the image authentication result output unit 150.

In addition, the FW authentication unit 144 calculates a correlation between the extracted soft watermark and the soft watermark generated for image authentication, and if the watermark is deformed, detects a forged and altered position of the image. The image authentication result output unit 150 is transmitted, and if there is no deformation, the image authentication result output unit 150 is transmitted.

The video authentication result output unit 150 outputs the results of the authentication decision of the RW authentication unit 142 and the FW authentication unit 144.

The video authentication result output unit 150 detects a watermark strong by the RW authentication unit 142 as described in one embodiment of the present invention and, according to the request, authenticates the original video and the FW authentication unit ( 144), it is possible to detect whether the image is forged or forged by detecting the weak watermark, and only the original authentication of the image transmitted through the network in the RW authentication unit 142, or the network in the FW authentication unit 144 It is also possible to detect whether the forgery of the image transmitted through and the forgery position.

The operation of the authentication device 100 according to an embodiment of the present invention described above is as follows.

The compressed video signal into which the watermark entered into the image input unit 110 through the network is inserted is stored in the image storage unit 120, and when the operator needs watermark authentication for the image, the image storage unit 120 The image is extracted from the compressed image decompression unit 130, and decompressed and restored to the original image signal.

The reconstructed image outputs the forged image and the authenticated image according to whether or not the watermark is deformed from the correlation between the watermarks strong by the RW authentication unit 142 of the image authentication unit 140, and the FW authentication unit 144 By detecting the altered position of the image according to whether or not the watermark is deformed from the correlation between the soft watermark and the transmission of the authentication image to the image authentication result output unit 150, and outputs the original position for the original monitoring if necessary In addition to providing the image, it authenticates whether the provided image is original and / or forged.

Next, a process of authenticating a strong watermark and a soft watermark in the image authenticator 140 will be described with reference to FIGS. 2B and 2C.

The method of extracting the watermark is an inverse process of the embedding process. When the key watermark is inserted, the watermark extraction process is as follows.

First, simply input the same unique key used to insert the watermark, and then calculate the correlation between the watermark generated from the key (i.e., the watermark for authentication) and the watermark extracted from the image. The image may be authenticated by determining the match, and the original may be authenticated through the extracted watermark. Details are described in pending Korean Patent Application No. 10-2000-0053755 of the same applicant as the applicant of the present invention.

In addition, the extraction process of the watermark when the image data is inserted into the watermark is similar, which is described in detail in Korean Patent Application No. 10-1998-0037273 and Korean Patent Application No. 10-1998-0037274.

In addition, the extraction process of RW is also described in detail in Korean Patent Application Nos. 10-2000-0053755, 10-1998-0037273 and 10-1998-0037274.

Details of the FW extraction may also use the contents of the corresponding patent application, which is described in Korean Patent Application No. 10-2000-0064767.

When the contents of the video authentication process by watermark extraction are summarized, it is understood that the watermark extraction process is the inverse process of the process of inserting the watermark.

FIG. 2B is a flowchart illustrating a process of authenticating an original image by extracting a strong watermark in the apparatus 100 of FIG. 2A.

First, an image to be authenticated is imported (S200), and a robust watermark is extracted from the imported image (S210). In addition, a watermark initially inserted among the key generation or the unique image is generated using the authentication information (S220).

In order to determine whether the extracted watermark is deformed, the correlation is calculated by comparing the extracted watermark with the watermark generated using the image authentication information (S230).

According to the correlation calculated as a result of the above, it is determined whether the watermark is deformed (S240). If it is determined that the watermark is deformed, a result is output as a forged image (S250). (S260).

In addition, through the information from the extracted watermark, it is possible to check the information on the authenticated image. For example, it means information that is strongly inserted to confirm the date of acquisition of the image, information about the operator, and the acquired place. RW is also the basis for authenticating video, but the extracted information is the basis for giving certainty to the authentication.

FIG. 2C is a flowchart illustrating a process of authenticating an image forgery by extracting a soft watermark in the apparatus of FIG. 2A.

The process of authenticating the weak watermark of FIG. 2C is not different from that of FIG. 2B. However, the purpose of the FW is to determine whether the image is forgery, so as a result of extracting the FW (S310), the process for identifying the forged part by identifying the forged position and visually marking the part (S350). Let go through.

As shown in FIGS. 2B and 2C, the image reconstructed for authentication calculates a correlation between the watermarks that are robust by the RW authentication unit 142 of the image authentication unit 140, and if there is a deformation in the watermark, the forgery is forged. If the image is not displayed, the image is authenticated. In addition, the FW authentication unit 144 calculates a correlation between the weak watermarks, detects and outputs a forged position of the image if there is a deformation in the watermark, and outputs an authentication image if there is no deformation.

That is, if the RW is not detected, it is determined that the image is primarily forged, and when the RW is detected, the image goes through the FW detection step. If the FW is properly detected, the image is considered fully certified. Even if the RW is detected, if the FW is not properly detected, the image is judged to be forged.

Next, with reference to FIGS. 3 and 4, an embodiment in which the technique of overlapping the watermark of the present invention is applied to a network camera server will be described.

3 illustrates a configuration of a network camera server 400 for inserting a strong watermark and a soft watermark into a plurality of images output from a plurality of image input devices 410 and transmitting them to a network according to a second embodiment of the present invention. The block diagram shown.

Referring to FIG. 3, the network camera server 400 includes a plurality of image data processing units 420, fixed information generating units 490, arbitrary information generating units 492, respectively, connected to the plurality of image input apparatuses 410. Watermark generation unit 494, a plurality of RW insertion units 430, a plurality of FW insertion units 450, an image compression unit 440, an image signal integration unit (MUX) 460, a network connection unit 470, And a real time operating unit 480.

The network camera server 400 according to the second embodiment of the present invention inserts a watermark for a plurality of image signals obtained from a plurality of image input apparatuses 410 (that is, external cameras) and transmits them to a network. As a function, it is almost the same in terms of the function of inserting a watermark to a video signal input through the network camera apparatus 10 according to the first embodiment of the present invention with reference to FIG. The detailed description is omitted.

However, since the plurality of video signals are input in real time through the plurality of external cameras 410, the fixed and random information generators 490 and 492 are to be inserted into watermarks corresponding to each of the plurality of video signals. Generate information.

In addition, the video signal integrating unit (MUX) 460 integrates a plurality of digital video signals in which a strong watermark and a soft watermark are overlapped into one digital video signal for efficient transmission to a network.

The operation process of the network camera server 400 according to the second embodiment of the present invention described above is as follows.

Image signals obtained from several external cameras 410 are input to respective image data processing units 420 using respective cables, and RW is inserted first in real time with respect to each image signal that is input and digitally converted. Since the processing must be performed, a plurality of RW inserting units 430 respectively corresponding to the plurality of image data processing units 420 insert watermarks that are robust to each image signal.

After generating the desired information in the fixed and random information generating units 490 and 492, respectively, to insert different information for each camera into the watermark, the watermark generating unit 494 then separates the cameras according to the generated information. A watermark is generated using the unique information of or a watermark is generated using arbitrary information transmitted.

In order to generate random information, the operator inputs information to be inserted into the video signal photographed by the camera through a dedicated browser at a remote location, and the real-time operating unit 480 forms a control signal and supplies it to the random information generation unit 492. Thus, the processing according to the operator's instruction can be performed.

The watermark generated by the watermark generator 494 is inserted in the RW-method into the digital image by the respective RW inserting units 430. At this time, since the input image is 10 frames to 30 frames per second, all watermark processing for this is to be performed in real time, and this operation is controlled by the real-time operating unit 480.

Each RW-inserted digital image is compressed in accordance with each compression scheme such as JPEG, Wavelet, MPEG, etc. by the image compression unit 440 so as to be transmitted in a network, and then each compressed image is FW insertion unit 450. A soft watermark is inserted at. For efficient transmission, the video signal integrator 460 integrates one video signal and then transmits it to the network through the network connection unit 470. In addition to providing surveillance video, it is possible to authenticate the original and / or forgery of the provided video.

Next, a device and an authentication process for authenticating a watermark-embedded image transmitted from the network camera server 400 will be described. 4 is a block diagram illustrating a configuration of an apparatus 500 for authenticating original and forgery of an image transmitted from a network camera server 400 of FIG. 3 through a network.

Referring to FIG. 4, the authentication apparatus 500 includes an image input unit 510 for inputting an image to be authenticated, an image signal separator (DEMUX) 520, a plurality of image storage units 530, and an image signal selector ( 540, an image authentication unit 560 including a compressed image extension unit 550, an RW authentication unit 562, and an FW authentication unit 564, and an image authentication result output unit 570. The authentication apparatus 500 according to an example may be implemented in a window environment of a personal computer for the convenience of a user.

In the image input unit 510, a plurality of digital images including a watermark embedded in the watermark transmitted from the network camera server 400 of FIG. 3 are compressed and combined into an image signal through a network.

The image signal separator 520 separates the image signal integrated into the one output from the image input unit 510 into the image signal of each compressed state. The image storage unit 530 stores each image signal separated by the image signal separation unit 520 in a compressed state.

The video signal selector 540 selects a specific video signal requiring authentication from among the plurality of compressed video signals stored in the video storage 530. The compressed image decompressor 550 decompresses an image signal stored in the image storage unit 530 selected by the image signal selector 540 and restores the image signal before compression.

The video authentication unit 560 includes a RW authentication unit 562 for detecting a strong watermark and authenticating the original image, and an FW authentication unit 564 for detecting whether the image is forged or not by detecting a weak watermark. The image authentication is determined by calculating a correlation between the watermark extracted from the reconstructed image signal of the image extension unit 550 and the watermark generated from the image authentication information.

The RW authentication unit 562 calculates a correlation between the extracted strong watermark and the strong watermark generated from the image authentication information, and if there is a deformation in the watermark, the result is a forged image. The result of the image is transmitted to the image authentication result output unit 570.

In addition, the FW authentication unit 564 calculates a correlation between the extracted soft watermark and the soft watermark generated from the image authentication information, and if the watermark is deformed, detects the forged position of the image. This is transmitted to the image authentication result output unit 570, and if there is no deformation, transmits the result of the authentication image to the image authentication result output unit 570.

The video authentication result output unit 570 outputs the results of the authentication decision of the RW authentication unit 562 and the FW authentication unit 564.

The image authenticator 570, upon request, detects a watermark that is robust by the RW authenticator 562 in the same manner as the image authenticator 140 according to an embodiment of the present invention with reference to FIG. The authentication of the image and the FW authentication unit 564 may detect a weak watermark to detect whether the image is forged or forged, and the RW authentication unit 562 may only authenticate the original image transmitted through the network. The FW authentication unit 564 may detect only the forgery and the forgery position of the image transmitted through the network.

The operation of the authentication apparatus 500 according to the second embodiment of the present invention described above is as follows.

The digital image signal transmitted from the plurality of external cameras 410 to the image input unit 510 through the network by inserting watermarks for the plurality of image signals, respectively, is integrated into the image signal separation unit (DEMUX) ( 520 is separated into an image signal photographed by an individual camera and stored in each image storage unit 530, and when an operator needs watermark authentication for the image, the image storage unit 530 displays the image. The video signal selection unit 540 extracts and decompresses the compressed image extension unit 550 to restore the original video signal.

The reconstructed image outputs that the image is a forged image and an authentication image according to whether or not the watermark is deformed from the correlation between the watermarks strong by the RW authentication unit 562 of the image authentication unit 560, and the FW authentication unit 562 ) Detects the forged position of the image and outputs the authenticated image to the image authentication result output unit 570 according to whether the watermark is deformed according to the correlation between the soft watermarks. While monitoring, if necessary, authentication of the original and / or forgery of the provided image is authenticated, as well as providing the original surveillance image.

Next, with reference to FIGS. 5 and 6, an embodiment in which the technique of overlapping the watermark of the present invention is applied to a digital video storage device (DVR) will be described.

5A illustrates a watermark inserting apparatus 610 for inserting a strong watermark and a soft watermark into a plurality of images input from a plurality of image input devices 600 and a watermark connected thereto according to a third embodiment of the present invention. A block diagram showing a configuration of a digital video storage device 630 for recording and storing an inserted image.

Since the functions and processes of the internal components of the watermark embedding apparatus 610 shown in FIG. 5A are similar to those of the network camera server 400 of FIG. 3, the description of similar parts will be simplified.

Such a watermark inserting device 610 may be integrated into the existing DVR in hardware or as a software function module.

First, an image signal (analog: NTSC, PAL) output from a plurality of video input devices (ie, a video camera, a digital camera, an analog camera, etc.) (analog: NTSC, PAL) may be integrated into a watermark inserting device (610: may be integrated into a DVR). Is entered.

The plurality of input video signals are time-divisionally connected by the video signal integrator (MUX) 612, and the video signal integrator 612 transmits each video signal to the video data processor 614 in time division.

The difference between the watermark embedding apparatus 610 of FIG. 5A and the watermark embedding apparatus 710 of FIG. 5B and the network camera server 400 of FIG. 3 is different from that of the position of the video signal integrating unit 612. As such, the video signal integrator 612 of FIG. 5A is at the front end of the device 610.

In the embodiment of FIG. 5A, the correlation between the signal passing through the image signal integrator 612 and the watermark embedding algorithm (ie, the insertion of the RW and FW) for each individual time-divided input under the control of the real-time operating unit 626. In order to implement such a system in real time, a watermark corresponding to an image must be inserted, and a fast watermarking insertion algorithm can be implemented. However, if the watermark insertion algorithm is small enough and the speed of a hardware processor is fast, the method of FIG. Embodiments are also useful.

Next, the video data processor 614 converts the analog video signal into a digital video signal (ie, A / D conversion).

The RW watermark and the FW watermark generated by the watermark generator 624 are inserted into the video signal output from the image data processor 614. The RW inserter 616 inserts a strong watermark. Insert data such as information, camera identification number and frame order to survive external attacks.

The FW inserter 618 inserts information for determining whether the video signal is forged / modulated. The Fragile Watermark Algorithm used here is strong in compression and has a feature that makes it possible to detect a forged / modulated part by disappearing the watermark signal when other forgery / modulation is performed.

More specifically, in the network camera apparatus and the network camera server with reference to FIGS. 1 and 3, since the FW insertion unit is located at the rear of the image compression unit, the FW does not need to be strong in compression. Since the video compression unit exists inside the DVR after the FW is inserted, the FW needs to be inserted with strong strength for compression and weak strength for other forgery and forgery attacks.

Unlike the network camera server 400 of FIG. 3, the image compression unit is not included in the watermark inserting apparatus 610 in the present embodiment, unlike the embodiment of FIG. Is a compression process.

The real-time operating unit 626 controls and controls the function of the watermark embedding apparatus 610 to insert watermark information. The video signal integrating unit 612 is controlled to make a plurality of input signals into one data string, and insert watermark information suitable for the corresponding video information.

The watermark-embedded video signal is input to the digital video storage device 630 for recording and storing. Since the signal output from the watermark inserting device 610 is one signal MUX, the video signal is input to one channel of the DVR 630. The watermark embedded image output from the watermark embedding apparatus 610 is stored in a storage medium such as a computer hard disk or a DAT by the DVR 630.

The DVR 630 compresses and stores an image. When the video signal is converted to digital, the amount of data increases, and a compression algorithm is performed to efficiently store the data. Currently used compression algorithms are MJPEG, H.263 and MPEG4. Therefore, FW that survives the compression algorithm and uses watermark disappears for other forgery / modulation (image substitution, image shape change, etc.).

5B illustrates a watermark embedding apparatus for inserting a strong watermark and a soft watermark into a plurality of images inputted from a plurality of image input devices according to a modification of the third embodiment of the present invention, and a watermark embedded image connected thereto; It is a block diagram showing the configuration of a digital video storage device for recording and storing.

5B is almost similar to the embodiment of FIG. 5A, except that only the position of the video signal integrator 718 is located at the rear end of the apparatus 710, and the watermark insertion portion RW and The difference is that it has a separate FW).

With this configuration, the embodiment of FIG. 5B is more advantageous for real time implementation of the watermarking algorithm than the embodiment of FIG. 5A. The watermark embedded video signal is transmitted to the digital video storage device 730 for recording. The detailed description is omitted since it is similar to the embodiment of FIG. 5A except that a watermark is separately inserted for each image signal.

Next, an apparatus and an authentication process for authenticating the watermark embedded image output from the DVR 630 of FIG. 5A and the DVR 730 of FIG. 5B will be described. FIG. 6 is a block diagram illustrating a configuration of an apparatus 800 for authenticating original and forgery of an image output to the digital video storage device of FIGS. 5A and 5B.

Referring to FIG. 6, the authentication apparatus 800 includes a video input unit 810 and a video signal separation unit (DEMUX) 820 in which a video signal inserted into a watermark and recorded and stored in the DVR is transmitted through a network or inputted in the form of a file. , An image authentication unit 830 consisting of an RW authentication unit 832 and an FW authentication unit 834, and an image authentication result output unit 840. Such an authentication device 800 has a structure or DVR integrated into a DVR. Both network and network-connected structures are possible.

Also, since the video output from the DVR is decompressed and output when the video stored in the DVR is output to the outside, it is not necessary to increase the compression in the authentication part as shown in FIGS. 2A and 4.

The image input unit 810 may receive the image in a form of transmission means or a file such as a series of networks for authentication of the digital image embedded with the watermark stored in the digital video storage device (DVR) 630 or 730.

The video signal separator (DEMUX) 820 performs a function of separating the watermark-embedded video signal into respective individual camera signals. That is, in FIG. 5A and FIG. 5B, when the watermark is inserted, a plurality of video signals are made into one stream information by using MUX. However, the authentication apparatus 800 attempts to authenticate from one stream information by using DEMUX. The RW and FW are authenticated by extracting the video signal from the camera.

The image authentication unit 830 functions to authenticate the RW and the FW from the separated image signal, and uses the same watermark detection algorithm as described with reference to FIGS. 2 and 4.

The image authentication result output unit 840 outputs the inserted information (copyright information, frame order of the image, camera number, etc.) as the authentication result of the RW, and if there is a case of forgery in the input image as the authentication result of the FW, Performs a function that displays the part that is displayed.

Next, with reference to FIGS. 7 and 8, an embodiment in which a watermark embedding apparatus is separately provided for each image input apparatus will be described.

FIG. 7A illustrates a watermark embedding apparatus 910 separately connected to each of a plurality of image input apparatuses 900 and a digital video storage apparatus for recording and storing a watermark embedded image connected thereto for each of the plurality of image input apparatuses 900 according to the fourth embodiment of the present invention. 930 is a block diagram showing a configuration.

The fourth embodiment with reference to FIG. 7A is different from the third embodiment with reference to FIGS. 5A and 5B. In the third embodiment, a watermark insertion apparatus is installed on the DVR side from a plurality of external cameras. On the other hand, in the fourth embodiment, a watermark embedding apparatus is provided for each external camera to insert a watermark into the video signal.

Referring to FIG. 7A, a watermark embedding apparatus 910 is separately attached to each image input apparatus 900 (ie, an external camera) to insert a watermark, and the inserted watermarked image signal is re-assigned if necessary. The image converter converts the analog image signal into an analog image signal or outputs a digital image as it is, which is input to a digital video storage device (DVR) 930 to be recorded and stored.

Image signals (digital or analog) signals are obtained from a plurality of video input devices (including a digital video camera, an analog video camera, a digital camera, an analog camera, etc.). The obtained video signal is input to each watermark embedding apparatus 910 to embed a watermark. The inserted video signal is recorded and stored by the DVR 930. Currently available DVRs can receive and process up to 16 channels. Therefore, when the watermarking system is implemented as shown in FIG. 7A, the existing DVR system may be used as it is. That is, the watermark inserting device 910 may be installed as a hardware or software module in an existing camera (image input device).

FIG. 7B is a block diagram showing the detailed configuration of the watermark embedding apparatus 910 of FIG. 7A.

The input signal of the watermark embedding apparatus 910 is an analog or digital video signal (NTSC, PAL, etc.), and when it is an analog video signal, is converted into a digital video signal by the image data processing unit (A / D) 912.

The watermark is inserted into the digital video signal through the RW insertion unit 914 and the FW insertion unit 916.

The video signal output from the watermark embedding apparatus 910 can be in both analog and digital forms, and in the case of analog output, A / D conversion is performed through the image data processing unit in the DVR 930. It can be used when the DVR can directly receive and process the digital video signal.

FIG. 8 is a block diagram illustrating a configuration of an apparatus 1000 for authenticating an original and a forgery of an image output from the digital video storage device 930 of FIG. 7A.

The authentication apparatus 1000 of FIG. 8 is very similar to the authentication apparatus 800 of FIG. 6, except that there is no video signal separation unit DEMUX. This is because the video signal with the watermark inserted according to the fourth embodiment of the present invention of FIG. 7 is stored in the DVR using a DVR-specific function without being integrated with the video signal and stored using a DVR-specific function. In the case of extracting video, the DVR selects and exports only the video of the desired information from the hard disk, but it does not need to integrate the video signal.

The image output from the DVR 930 is transmitted through a network or copied to a file via a diskette or the like, and the image information is sent to a device on which the image authentication apparatus 1000 is installed. In this way, the image signal input to the image input unit 1010 determines whether the image authentication unit 1020 forgery, and displays the result through the image authentication result output unit 1030. Detailed description thereof is the same as in FIG. 6 and will be omitted.

As described above, after acquiring the image through the camera through the network camera and the network camera server and inserting a watermark into the obtained image, preventing a forgery, and performing a series of procedures for authenticating the original, the image is transmitted through the network. The method and apparatus were described as an example, and another example was described in which a watermark was inserted into the acquired image and stored in a digital video storage device (DVR), followed by a series of methods and apparatus for authenticating the original and confirming the forgery. In addition, the method and apparatus of the present invention can be applied to a system for image acquisition, storage, and transmission.

In order to perform the series of watermarking procedures, an independent device (a box form or a chip or chipset form) that performs this function may be implemented. That is, a series of processes of embedding a watermark in the image may be implemented in a video transmission camera, or may be implemented as a separate box or the like to be mounted outside the camera or in a DVR box.

Although the invention has been particularly shown and described with reference to the above embodiments, it has been used for the purpose of illustration and those of ordinary skill in the art to which the invention pertains have the spirit and scope of the invention as defined in the appended claims. Various modifications can be made without departing.

As described above, in a network camera and a camera server device having a watermark embedding function and a camera and a video storage device having a watermark embedding function, an original image of the image is extracted through a process of extracting the inserted watermark. And according to the present invention for authenticating whether or not forgery, at the same time to prevent the illegal operation of the video that can be applied to the image captured for surveillance, and when such surveillance video is illegally manipulated, Has an effect.

In addition, it has an authentication effect of checking the authenticity of the images of the network camera and the camera server with the watermark embedding function and the image recorded on the storage device. In other words, the image taken by the camera contains information about the operator and the image, and the image is authenticated by simultaneously inserting a strong watermark that survives any external attack and a soft watermark that damages the watermark even under minute external manipulation or deformation. If necessary, it has the effect of not only authentic operator and image authentication but also whether the image is forged or not.

In the present invention, a watermarking method capable of detecting a forgery of an image as well as a strong watermark, that is, by inserting a soft watermark together, can increase the effectiveness of authentication of the image through two authentications.

Claims (32)

An image input unit which receives an image signal captured in real time; An image data processor converting an image signal output from the image input unit into a digital signal; An information generating unit for generating information to be inserted as a watermark; A watermark generator for generating a watermark using information of the information generator; A watermark inserting unit inserting a watermark generated by the watermark generating unit into an image signal output from the image data processing unit; An image compressor which compresses an image signal to which the watermark inserted from the watermark inserter is inserted; And And a network connection unit for transmitting the compressed image signal output from the image compression unit to a network. The method of claim 1, The information generator is a fixed information generator for generating unique information of the network camera device, And a watermark inserted into the video signal by the watermark inserting unit is a robust watermark. An image input unit which receives an image signal captured in real time; An image data processor converting an image signal output from the image input unit into a digital signal; An image compressor which compresses an image signal output from the image data processor; An information generating unit for generating information to be inserted as a watermark; A watermark generator for generating a watermark using information of the information generator; A watermark inserting unit inserting the watermark generated by the watermark generating unit into the compressed image signal output from the image compressing unit; And And a network connection unit which transmits the watermark-embedded video signal outputted from the watermark insertion unit to a network. The method of claim 3, wherein The information generation unit is an arbitrary information generation unit for generating information transmitted from a remote place through the network, And a watermark inserted into the video signal by the watermark inserting unit is a soft watermark. An image input unit which receives an image signal captured in real time; An image data processor converting an image signal output from the image input unit into a digital signal; An information generating unit for generating information to be inserted as a watermark; A watermark generator for generating a watermark using information of the information generator; A first watermark inserting unit inserting a first watermark generated by the watermark generating unit into an image signal output from the image data processing unit; An image compressor which compresses an image signal into which the first watermark is inserted, which is output from the first watermark inserter; A second watermark inserting unit inserting a second watermark generated by the watermark generating unit into the compressed image signal output from the image compressing unit; And And a network connection unit which transmits the video signal inserted with the second watermark inserted from the second watermark inserting unit to a network. The method of claim 5, The information generation unit fixed information generation unit for generating the unique information of the network camera device; And An arbitrary information generating unit for generating information transmitted from a remote place via the network; The first watermark inserted into the video signal by the first watermark embedding unit is a robust watermark, and the second watermark embedded into the video signal by the second watermark embedding unit is soft. Network camera device, characterized in that the watermark (Fragile Watermark). The method according to any one of claims 1 to 6, And a real time operating unit for controlling the insertion of the watermark, the compression of the video signal, and the generation of the arbitrary information in real time. An apparatus for authenticating a watermark embedded image transmitted from a network camera apparatus according to any one of claims 1 to 6, An image input unit configured to receive a compressed video signal having a watermark embedded thereon through a network; A compressed image extension unit for restoring an image signal output from the image input unit to an image signal before compression; An image authenticator which determines the authentication of the image by calculating a correlation between the watermark extracted from the image signal reconstructed by the compressed image extension unit and the watermark generated from the image authentication information; And And an image authentication result output unit configured to output an authentication result of the image authentication unit. The method of claim 8, The image authentication unit includes a strong watermark authentication unit for authenticating the original image by detecting the strong watermark; And And a weak watermark authentication unit for detecting whether the image is forged and the forged position by detecting the weak watermark. In the method of inserting a watermark to the video signal captured by the network camera device to transmit to the network, Converting an image signal input in real time into a digital signal; Inserting a robust watermark containing unique information of a network camera device into the converted video signal in real time; Compressing the video signal having the strong watermark embedded therein; Inserting a soft watermark containing random information transmitted through a network into the compressed video signal in real time; And And transmitting the video signal embedded with the second watermark to a network. A plurality of image data processing units for converting each of the plurality of image signals input in real time from the plurality of cameras into digital signals; An information generation unit for generating information to be inserted as a watermark corresponding to each of the plurality of video signals; A watermark generator for generating respective watermarks corresponding to each of the plurality of video signals by using the information generator; A plurality of watermark inserting units respectively inserting watermarks generated by the watermark generating unit into each of the plurality of image signals output from the plurality of image data processing units; An image compressor which compresses a plurality of image signals into which watermarks output from the plurality of watermark inserters are inserted; An image signal integrator for integrating a plurality of image signals output from the image compressor into one image signal; And And a network connection unit for transmitting the integrated video signal output from the video signal integrator to a network. The method of claim 11, The information generation unit is a fixed information generation unit for generating unique information of the plurality of cameras or the network camera server, And a watermark inserted respectively corresponding to the plurality of video signals by the plurality of watermark inserting units is a robust watermark. A plurality of image data processing units for converting each of the plurality of image signals input in real time from the plurality of cameras into digital signals; An image compressor which compresses image signals output from the plurality of image data processors; An information generation unit for generating information to be inserted as a watermark corresponding to each of the plurality of video signals; A watermark generator configured to generate respective watermarks corresponding to each of the plurality of image signals by using the information generator; A plurality of watermark inserting units respectively inserting watermarks generated by the watermark generating unit into each of the plurality of image signals output from the image compressing unit; An image signal integrating unit for integrating a plurality of image signals embedded with watermarks outputted from the plurality of watermark embedding units into one image signal; And And a network connection unit for transmitting the integrated video signal output from the video signal integrator to a network. The method of claim 13, The information generation unit is an arbitrary information generation unit for generating information transmitted from a remote place through the network, And a watermark inserted respectively corresponding to the plurality of video signals by the plurality of watermark inserting units is a soft watermark. A plurality of image data processing units for converting each of the plurality of image signals input in real time from the plurality of cameras into digital signals; An information generation unit for generating information to be inserted as a watermark corresponding to each of the plurality of video signals; A watermark generator configured to generate respective watermarks corresponding to each of the plurality of image signals by using the information generator; A plurality of first watermark inserting units respectively inserting first watermarks generated by the watermark generating unit into each of the plurality of image signals output from the plurality of image data processing units; An image compressor which compresses a plurality of image signals into which watermarks output from the plurality of first watermark inserters are inserted; A plurality of second watermark inserting units respectively inserting second watermarks generated by the watermark generating unit into each of the plurality of image signals output from the image compressing unit; An image signal integrating unit integrating a plurality of image signals embedded with watermarks outputted from the plurality of second watermark embedding units into one image signal; And And a network connection unit for transmitting the integrated video signal output from the video signal integrator to a network. The method of claim 15, The information generation unit fixed information generation unit for generating the unique information of the plurality of cameras or the network camera server; And An arbitrary information generating unit for generating information transmitted from a remote place via the network; The first watermarks respectively inserted in correspondence with the plurality of video signals in the plurality of first watermark embedding units are robust watermarks, and the plurality of images in the plurality of second watermark embedding units. And the second watermarks respectively inserted corresponding to the signals are fragile watermarks. The method according to any one of claims 11 to 16, And a real time operating unit which controls the insertion of the watermark, the compression of the video signal, and the generation of the arbitrary information in real time. An apparatus for authenticating a watermark embedded image transmitted from a network camera server of any one of claims 11 to 16, A video input unit which receives a video signal having a watermark inserted therein and integrated in a compressed state through a network; An image signal separation unit for dividing the integrated image signal output from the image input unit into an image signal corresponding to each camera; A plurality of image storage units respectively storing the image signals separated by the image signal separator; An image signal selection unit for selecting an image signal requiring authentication from among the image signals stored in the image storage unit; A compressed image extension unit for restoring a video signal selected by the video signal selection unit to a video signal before compression; An image authenticator which determines the authentication of the image by calculating a correlation between the watermark extracted from the image signal reconstructed by the compressed image extension unit and the watermark generated from the image authentication information; And And an image authentication result output unit configured to output an authentication result of the image authentication unit. The method of claim 18, The image authentication unit includes a strong watermark authentication unit for authenticating the original image by detecting the strong watermark; And And a weak watermark authentication unit for detecting whether the image is forged and the forged position by detecting the weak watermark. A method of inserting a watermark into a plurality of video signals input from a plurality of cameras and transmitting the watermark to a network, Converting a plurality of image signals input in real time from the plurality of cameras into digital signals; Inserting, in each of the plurality of converted video signals, a strong watermark containing unique information of the plurality of cameras or a network camera server in real time; Compressing a plurality of video signals each having the strong watermark embedded therein; Inserting in each of the plurality of compressed video signals a soft watermark containing random information transmitted through a network in real time; Integrating the plurality of video signals including the soft watermark into one video signal; And Transmitting the integrated video signal to a network. A digital video storage device (DVR) including a watermark embedding apparatus for embedding a watermark into a plurality of video signals input from a plurality of cameras, The watermark embedding apparatus A plurality of image data processing units for converting each of the plurality of image signals input in real time from the plurality of cameras into digital signals; An information generation unit for generating information to be inserted as a watermark corresponding to each of the plurality of video signals; A watermark generator configured to generate respective watermarks corresponding to each of the plurality of image signals by using the information generator; A plurality of first watermark inserting units respectively inserting first watermarks generated by the watermark generating unit into each of the plurality of image signals output from the plurality of image data processing units; And A video signal integrating unit for integrating a plurality of video signals embedded with watermarks output from the plurality of first watermark embedding units into a single video signal, And a video signal output from the video signal integrator is compressed and recorded. The method of claim 21, The watermark embedding apparatus And a plurality of second watermark inserting units respectively inserting second watermarks generated by the watermark generating unit into each of the plurality of image signals output from the plurality of first watermark inserting units. Digital video storage. The method of claim 22, The information generating unit comprises a fixed information generating unit for generating unique information of each of the plurality of cameras; And An arbitrary information generating unit for generating information transmitted from a remote place via the network; The first watermarks respectively inserted in correspondence with the plurality of video signals in the plurality of first watermark embedding units are robust watermarks, and the plurality of images in the plurality of second watermark embedding units. And the second watermarks respectively inserted corresponding to the signals are fragile watermarks. The method according to any one of claims 21 to 23, The watermark embedding apparatus And a real-time operating unit for controlling the insertion of the watermark and the generation of the arbitrary information in real time. The method according to any one of claims 21 to 23, And the watermark embedding apparatus is integrated into the digital video storage device in hardware or as a software module. An apparatus for authenticating a watermark embedded image from the digital video apparatus according to any one of claims 21 to 23, An image input unit configured to receive an image signal having a watermark inserted therein and integrated into one signal in a form of a transmission or a file through a network; An image signal separation unit for dividing an image signal integrated into one signal output from the image input unit into an image signal corresponding to each camera; An image authenticator configured to determine an authentication of an image by calculating a correlation between a watermark extracted from the image signal from the image signal separator and a watermark generated from the image authentication information; And And an image authentication result output unit configured to output an authentication result of the image authentication unit. A method of inserting a watermark into a plurality of video signals input from a plurality of cameras and recording and storing the watermark, the method comprising: Converting a plurality of image signals input in real time from the plurality of cameras into digital signals; Inserting, in each of the plurality of converted video signals, a robust watermark containing unique information of the plurality of cameras in real time; Inserting, in real time, a weak watermark containing arbitrary information transmitted through a network into each of the plurality of video signals into which the strong watermark is inserted; Integrating the plurality of video signals including the soft watermark into one video signal; And Compressing and recording the integrated video signal. In a digital video storage device (DVR) for recording and storing a plurality of video signals input from a plurality of cameras, The plurality of cameras are provided with a plurality of watermark inserting devices, respectively, Each of the plurality of watermark embedding apparatuses An image data processor converting an image signal input in real time from a corresponding camera into a digital signal; An information generating unit for generating information to be inserted as a watermark; A watermark generator for generating a watermark using information of the information generator; And A first watermark inserting unit inserting a first watermark generated by the watermark generating unit into an image signal output from the image data processing unit, And a video signal output from the first watermark inserting unit is compressed and recorded. The method of claim 28, Each of the plurality of watermark embedding apparatuses And a second watermark embedding unit which embeds a second watermark generated by the watermark generating unit in the image signal output from the first watermark inserting unit. The method of claim 29, The information generator includes a fixed information generator for generating unique information of the corresponding camera; And It includes an arbitrary information generating unit for generating information transmitted from the remote over the network, The first watermark inserted into the video signal by the first watermark embedding unit is a robust watermark, and the second watermark inserted by the second watermark embedding unit corresponding to the video signal. Is a soft watermark (Fragile Watermark). The method according to any one of claims 28 to 30, Each of the plurality of watermark embedding apparatuses And a real-time operating unit for controlling the insertion of the watermark and the generation of the arbitrary information in real time. 31. An apparatus for authenticating a watermark embedded image from the digital video apparatus according to any one of claims 28 to 30. An image input unit which receives an image signal in a form of a transmission or a file through a network; An image authentication unit which determines an authentication of an image by calculating a correlation between a watermark extracted from the image signal output from the image input unit and a watermark generated from the image authentication information; And And an image authentication result output unit configured to output an authentication result of the image authentication unit.