JP2011151770A - Image encrypting system for output of encrypted images subjected to undefining treatment of degree according to authorized browsing person - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To enable image browsing while keeping privacy by encrypting and storing images captured by a monitor camera and by displaying the images decrypted and subjected to undefining treatment according to browsing persons. <P>SOLUTION: A video is encrypted by exclusive software and a single key is prepared to decrypt the encrypted video. Decryption software uses the key to display the images subjected to undefining treatment such as a predetermined degree of mosaic treatment. A browsing person is allowed to browse only the images subjected to a defined degree of undefining treatment by distributing software available for the browsing person depending upon the degree of browing permitted for the browsing person concerning the captured images. Different image browsing keys are distributed to browsing persons who can encrypt and browse the video with use of exclusive software, and decryption software is distributed to the browsing persons. The decryption software displays images subjected to a degree of undefining treatment according to the contents of the image browsing keys. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a surveillance camera system, and more specifically, according to a person (viewing right holder) who can view stored images of surveillance cameras installed in various places. This is a system that changes the intensity of smearing processing such as mosaic processing.

  Surveillance cameras are now widespread because they are recognized for their effectiveness in deterring crime and arresting criminals. On the other hand, surveillance cameras always take pictures of people around the house, public places such as parks and shopping streets, and save the images, so all the pictures of people passing in front of the camera are taken and saved. It will be. The image is intended to be used for tracking offenders, but if the security camera owner sees and enjoys the image with interest, it can be a violation of privacy. Therefore, a non-profit organization e self-security network study group in which the inventors organize a surveillance camera system that can be viewed only by people and organizations that have an apparatus that can encrypt and store the image and decrypt the encryption has been proposed. .

  However, this eliminates various uses of surveillance camera video and loses its convenience. In other words, surveillance camera images have various uses other than crime prevention and criminal tracking. For example, you can check the congestion of the park by watching the video of the surveillance camera installed in the park, but if you encrypt the image, you can not see the image itself unless you know the decoding method, It is impossible to know how busy the park is, and if you tell the unspecified number about the decryption method, the meaning of encryption will be lost in the first place, and everyone can know who is in the park. If this happens, it will be a violation of privacy. So, by providing an image with a moderate intensity mosaic, you can see how crowded the park is, but you don't know who is there, without infringing on privacy, and information on the congestion of the park Can be provided.

  Surveillance camera images are useful when an elderly person or a young child gets lost, but the captured images may also show people other than the elderly or child being searched. Viewing images can violate privacy. However, if the search target is known, even if the image has some mosaic, it can be searched if the search target is reflected in the image. You can't see what a person is doing or what he is doing. In this way, by using an appropriate mosaic, it is possible to effectively use the video of the surveillance camera while preventing invasion of privacy, and display the image of the surveillance camera by changing the intensity of the mosaic according to the viewing rights holder This is effective in terms of effective use of surveillance cameras.

  A new security camera operation characterized by separating the person who owns and manages the image (owner) and the person who can view the image (viewing right holder) by encrypting the image. The concept is proposed at the e-surveillance network study group of the non-profit organization organized by the inventors. As a result, the encrypted image is passed from the owner to the viewing-right holder only when both the owner and the viewing-right holder admit that it is necessary, such as when an incident occurs, and the viewing right holder decrypts the view. Will be done. As a result, no one can view the images except in an unnecessary case. This concept was proposed to promote the spread of security cameras throughout the local community by reducing the risk of privacy infringement and, at the same time, reducing unnecessary mental loads that can be imparted to third parties. It is done.

  In this way, privacy infringement and concerns can be greatly reduced by separating the video camera owner from the viewing rights holder. The inability to browse the video at all is very small from the owner's point of view. It can also be an obstacle to the spread of surveillance camera systems based on the above-mentioned concept. Therefore, this problem can be solved by letting the owner or other viewers view the image by applying a mosaic of moderate strength to the image without making the owner browse the image completely.

  In this way, if there are multiple viewing rights holders or multiple groups, and it is necessary to encrypt and save the images of the surveillance camera, if it is necessary to present images with different mosaic strengths according to each viewing rights holder, This solves the problem of camera privacy violation and improves the convenience of surveillance cameras.

There are many patent documents, not limited to Patent Document 1, Patent Document 2, Patent Document 3, and the like, to date, for image encryption and decryption methods, not limited to surveillance cameras. Patent Document 1 proposes a method for encrypting an entire image or a part of an image. Patent Document 2 proposes a method for encrypting and decrypting printed materials and digital images, and a method for decrypting multiple encrypted parts. Patent Document 3 proposes an image encryption method between a sensor and an image processing device in a sensor network. Many image encryption methods have been proposed, but no method has been proposed for creating and encrypting images with multiple mosaics from one image.
Japanese Patent No. 4348381 JP 2009-232333 A Japanese Patent No. 4112509

  From such a background, the present invention encrypts and saves the video shot by the surveillance camera, and decrypts the encrypted video according to the viewer when decrypting it for viewing. By displaying an image that has been subjected to smearing processing such as mosaic processing, it is possible to view the image while protecting privacy.

  When the video captured by the surveillance camera is encrypted and stored, and decrypted for viewing the encrypted video, mosaicing is performed on the decrypted image according to the viewer and the decryption key. The above problem is solved by performing the blurring process and displaying it.

  As a method for making it possible to browse only images with mosaics having different intensities for each viewing-right holder, the viewer uses the strength of blurring processing such as mosaic processing of displayed images. There is a method of changing every. In other words, the video taken by the surveillance camera is encrypted by a dedicated device, and one key is provided to decrypt it. The encrypted video is decrypted using a dedicated device that decrypts using this decryption key, and this dedicated device displays an image that has been subjected to a blurring process such as a mosaic processing with a predetermined intensity. . By distributing the available devices depending on how far the viewer can view the captured images, only images that have been subjected to smearing processing such as mosaic processing with a predetermined intensity can be viewed.

  Another method is to distribute different image browsing keys for each viewer and use the same decryption device. In other words, the video captured by the surveillance camera is encrypted by a dedicated device, and a different image browsing key is distributed to each person who can view it, and a decryption device is distributed. The viewer inputs the encrypted image to the decryption device and also inputs the distributed image browsing key. The decoding device decodes the image based on the image browsing key and displays the image subjected to smearing processing such as mosaic processing with the intensity based on the content of the image browsing key.

  There are two ways to implement this method. One is to encrypt and save the video shot by the surveillance camera. When viewing an image, the decryption device extracts the decryption key from the image viewing key and decrypts the image using this decryption key. After that, data representing the intensity of blurring processing such as mosaic processing is extracted from the image browsing key, and a mosaiced image is output or displayed based on the data.

  Another method is to create images that have been subjected to smearing processing such as mosaic processing with various intensities on images taken with a surveillance camera, encrypt them individually, and connect the encrypted images to create a single image. Save as encrypted image data. This image data consists of multiple encrypted data of images that have undergone smearing processing such as mosaic processing. In the case of decoding, the image browsing key distributed to each viewer is used as a key for decoding an image that has been subjected to smearing processing such as any mosaic processing included in the image data. By doing so, an image can be obtained.

  As the smearing process, various existing methods can be used in addition to the mosaic process. In addition, it is also effective to set the intensity depending on the location in the blurring process over the entire screen. For example, it is effective to set the mosaic intensity low on the site and increase the mosaic intensity outside the site. Furthermore, it is also effective to detect moving objects and to apply a smearing process with a specific intensity to moving objects. It is also effective to identify moving objects and apply mosaic processing according to the object such as men, women, and cars. Furthermore, it is effective to replace the identified object with an abstraction and replace it with a figure.

  It is also effective to incorporate analysis results such as various sensor information and identification analysis results (male, female, adult, child, car, bicycle, etc.) into the encrypted image. As sensor information, reading information of IC tag, sound, atmospheric pressure, temperature, etc. are useful. In particular, sound recording is often useful.

  In addition, the image can be either a still image or a moving image separated by appropriate time intervals. When putting in the sound picked up by the microphone, it may be only within the image acquisition interval and acquisition period, or it may be convenient to overlap each other and record it in the extra eye.

  Allowing the original image file to be handled only by a dedicated device is effective in preventing tampering and proving that there is no tampering. At that time, it is also effective to record the history of decryption and blurring processing for the original image of the encrypted image file.

An image restoration device that prepares multiple encryption keys and outputs different mosaic strengths for each key is effective. The software itself is also encrypted, so it is effective even if the encrypted image can be restored only by this device. Also, prepare several kinds of keys, for example, as follows. Everyone can have the same device or separate devices.
・ ESelf-camera owner: Only images with a strong mosaic are output. (Same as a monitor)
・ Contractors, residents' associations, etc .: Images with weak mosaics are output.
・ Police: A clear image without a mosaic is output.

The present invention is particularly effective when used in operation of e-self-camera cameras and e-warning lights. The e-Self-Camera Camera and e-Self-Light are products developed by the NPO e-Self Network Research Group.

In both cases where the image is subjected to the blurring process according to the viewing right person or when the image whose degree of smearing is changed according to the viewing right person is presented, a key relating to the viewing right person is required. By encrypting the key together with the image, it is possible to embed a key corresponding to the viewing right holder, that is, a key related to the strength of the viewing right, in the image itself.
For example, a key related to a plurality of viewing rights is attached to an image, and it is encrypted. By giving the viewing right person a key for decryption and a key relating to the viewing right person according to the strength of the authority of the viewing right person, the viewing right person has a key relating to the viewing right person. When browsing an encrypted image, the viewing right person inputs a key for decryption and a key related to the viewing right person to the image viewing apparatus. The image browsing device decrypts the encrypted image with the decryption key, and checks whether the input key related to the viewing right user matches one of the keys related to the viewing right attached to the image, An image that has been subjected to the smearing process on the image according to the matched key is presented to the viewing right holder.

When a key is attached to an image for encryption, various methods can be considered for attaching the key. In addition to the method of simply attaching a key related to the viewing right holder to the image data, there is a method of inserting it inside the image data, and there are many ways to make it difficult to guess the key related to the image data or the viewing right holder. .

Another method for making it possible to browse only images that have been subjected to a blurring process with different intensities for each viewing right holder is to perform the blurring process on the images in stages. When there are multiple viewing rights holders, the blurring process is performed step by step according to the number of people. For example, if an image is composed of 640x480 pixels and there are two viewing rights holders A and B, A can only see the image that has been subjected to the blurring process, and B is not subjected to the blurring process. Suppose you can see an image. At this time, two keys Ka and Kb are prepared, and one key Kb is blurred by a reversible method such as replacing each pixel component, for example, the RGB value of each pixel, for each 10x10 pixel of 640x480 pixels. Process. Next, with another key Ka, the image is further smeared by a reversible method such as replacing the pixel components for the entire image. If necessary, an image to be saved can be obtained by further encrypting the image obtained by the two-stage blurring process.
The method of browsing images is as follows. If the image is encrypted, it is assumed that viewing rights holders A and B have decryption keys, A has keys Ka, and B has keys Ka and Kb. If the image is encrypted, A decrypts it with the decryption key, and then restores the components of the replaced pixel with the key Ka. This allows A to view an image in which the components of the pixel are swapped within each 10x10 pixel. Since B also has a key Kb, it is possible to undo the replacement of the pixel components within each 10x10 pixel of this image, and B can see an image that has not been blurred. it can.
If there are more than 3 viewing rights holders, the number of viewing rights is 2 by first replacing each 10x10 pixel, then 100x100 pixels, and finally the entire pixel component As with, the viewing rights holder can browse images with different levels of blurring processing according to the viewing rights holder.

  It is possible to cleanly remove the part of the security camera that was regarded as the necessary evil of “risk of privacy infringement” and to apply a smearing process such as appropriate mosaicing to the image. Convenience can be maintained. With the increasing threat of terrorism, surveillance cameras are becoming increasingly popular as a necessary evil.

  If the risk of privacy infringement is reduced and convenience is ensured, there will be no obstacles to the spread of surveillance cameras, and many areas will be the surveillance range of security cameras. Criminals can no longer escape out of surveillance by security cameras, which has a tremendous effect on arrest of criminals, resulting in a reduction in crime.

  With the system of the present invention, it becomes possible for each household to monitor not only the house but also public places such as the road in front of the house, thereby improving the security of each house. In addition, by applying a smearing process such as appropriate mosaicing, it is possible to use video from surveillance cameras for purposes other than crime prevention without infringing on privacy. In addition, because public places can be monitored by security cameras installed by each household, police and government can increase the safety and security of the town without financial burden. The present invention is a key idea for e-security camera (= security camera system with privacy protection function by encryption).

  Figure 1 shows an example of how to change the intensity of the mosaic by changing the image browsing device distributed according to the viewer. The surveillance camera 1 takes an image, encrypts the image, and saves it. Distributing image browsing devices that display mosaiced images of different strengths to the viewing-right holders who can view the encrypted image 2. The viewing right holders 3, 4, and 5 in FIG. 1 have decryption keys 6, 8, and 10, respectively, and have image display devices 7, 9, and 11, respectively. The decryption keys 6, 8, and 10 are the same, but the image display devices 7, 9, and 11 are devices that display different images. Each viewing-right holder inputs the image he wants to browse and the decryption key he / she has into his / her image display device. The device decodes the image and outputs an image with a predetermined mosaic intensity.

  Figure 2 shows a system that encrypts and stores the image itself, distributes the image browsing keys with different mosaic strengths to the viewers, and allows each viewer to browse the images. The surveillance camera 1 takes an image, encrypts the image, and saves it. Distributing an image browsing key for displaying mosaiced images of different strengths to the viewing-right holders who can view the encrypted image 2. The right holders 3, 4, and 5 in FIG. 2 have image browsing keys 6, 8, and 10, respectively, and have image display devices 7, 9, and 11, respectively. The image browsing keys 6, 8 and 10 are composed of a key for decrypting an encrypted image and a portion indicating the strength of the mosaic. The image browsing keys 6, 8, and 10 differ accordingly. The image display devices 7, 9, and 11 are the same device. If each viewer has the right image to browse, the image browsing device decodes the image and outputs a mosaic image with the intensity specified by the image browsing key. The image browsing key used in Fig. 2 consists of a decryption key for decrypting the encrypted image and a portion representing the strength of the mosaic. As shown in Fig. 3, an image browsing key can be created by combining the decryption key and the portion representing the intensity of the mosaic. If the decryption key and the data indicating the mosaic strength can be easily separated, those who can only browse with a strong mosaic using the decryption key will change the data representing the strength of the mosaic to a weaker mosaic and use the image viewing key. There is a possibility that images with weak mosaics can be viewed. For this reason, the configuration of the image browsing key is kept secret, the encryption and decryption method is kept secret, the inside of the encryption and decryption device is kept secret, and data indicating the strength of the mosaic is stored. By representing with very large bits, it is difficult to guess the data indicating the strength of the decryption key and mosaic.

  The modes for carrying out the inventions shown so far are the mode in which a decryption device is distributed by different viewing rights holders and a common decryption key is used, and the distribution of image browsing keys by different viewing rights holders is common. This is a form using a device. There is also a combination of these two. In other words, a different decryption device and a different image browsing key are distributed to each viewing right person, an encrypted image is handed over to each viewing right person, and an image is displayed using the decryption device and the image viewing key that each has. Decrypt. If you use something different from the distributed image browsing key or decryption device, you cannot decrypt it.

  Fig. 4 shows a method of creating a mosaiced image with varying intensity on the image itself captured by the surveillance camera, and creating a single piece of data by creating data in which each mosaic image is encrypted with a different key. ing. In FIG. 4, an image 2 with a very strong mosaic, an image with a strong mosaic 4, and an image with a weak mosaic applied to the image 2 taken by the surveillance camera 1 according to the intensity of the mosaic. 5. Create four images of image 6 without mosaic. These four images were encrypted with different keys, and the encrypted image 7 with a very strong mosaic, the encrypted image 8 with a strong mosaic, and the encrypted weak mosaic were applied. Create four images, image 9, image 10 without an encrypted mosaic, and combine these four encrypted images to create stored data 12.

  FIG. 5 shows a method for decoding data stored by the method of FIG. Decryption device 3 receives stored data 1 and decryption key 2 of any one of A, B, C, and D as inputs. An error occurs if the correct decryption key is not entered. Depending on whether the decryption key 2 is A, B, C, or D, the decryption device 3 extracts the corresponding encrypted data from the stored data 1 and decrypts the image with the decryption key 2. Output.

In the method shown in FIGS. 4 and 5, a decodable image is output from data obtained by concatenating a plurality of encrypted images. One of the methods will be described with reference to FIGS. Fig. 6 shows a method for generating saved data from a captured image. A message digest 3 is obtained from the image 2 that has been subjected to the blurring process obtained by the blurring process on the image 1 taken by the monitoring camera. Encryption is performed on the image 4 that has been subjected to the blurring process and the message digest 3, and the encrypted data 6 is obtained. Encrypted data 5, 7, and 8 are generated by performing the same process while changing the intensity of the blurring process, and the stored data 9 is obtained by combining the encrypted data 6 with the encrypted data 6. Figure 7 shows how to retrieve an image from stored data. In the case of extraction, the stored data 1 and the composite key 2 are given to the decryption device 3. The stored data 1 is composed of a plurality of encrypted data, and in FIG. 7, it is composed of four encrypted data. The decryption device 3 applies the decryption key 2 to each encrypted data included in the stored data 1 to obtain decrypted data 4, 5, 6, and 7. The decrypted data consists of a blurred image and a message digest. The decrypted data 5 is composed of an image 8 and a message digest 9 that have been subjected to a blurring process. A message digest 10 is generated from the image 8 subjected to the smearing process, and compared with the message digest 9. If the message digests 9 and 10 match, the decoding apparatus 3 outputs the image 8 subjected to the smearing process. The decoding apparatus 3 performs the same processing on the decoded data 4, 6, and 7, and extracts the image.
Fig. 8 shows the flow of processing that adds a key corresponding to the viewing rights holder to the image data and encrypts it to save data. A key corresponding to the viewing rights holder is added to the image taken with the camera, and the data is encrypted. Each viewing owner has a decryption key and a viewing owner's key. When the viewing right person browses the image, the viewing right person inputs the decryption key and the viewing right person's key to the image viewing apparatus. The image browsing device decrypts the data with the decryption key, and searches for a match between the input rights holder's key and the rights owner's key included in the data. If there is a match, the image that has been subjected to the smearing process determined by the key of the matching view right holder is displayed. If there is no match, nothing is displayed, or all or part of the image smeared at a preset intensity is output.

Fig. 9 shows an image encryption method that divides an image into sections that are rougher than the pixels and replaces the pixels in the sections based on the encryption key in accordance with the viewing right holder. Figure 2 illustrates the case where there are two viewing rights holders. The captured image is divided into 10x10 pixel units, and each element of RGB of 100 pixels, a total of 300 values, is replaced based on the key Kb. In addition, each RGB element of the entire image is replaced based on the key Ka. This result is encrypted to save data. The viewing rights holder has a decryption key, one viewing rights holder has the key Ka, and the other viewing rights holder has both the keys Ka and Kb.
When browsing the image, the viewing right person inputs the decryption key and the key that he / she owns into the image browsing device, and after the image browsing device decrypts the image with the decryption key, based on the input key, Undo the replacement of the image components and display the image. If the only key entered is Ka, the replacement inside each 10x10 image cannot be undone, so the image with the components of each 10x10 image still replaced is displayed. When the input keys are Ka and Kb, it is possible to undo the replacement in each 10x10 image, so the captured image is displayed as it is.
If the input key is Ka only, the 10 × 10 pixel averaging process (mosaic processing) for the obtained image is the same as the 10 × 10 pixel averaging process for the original image ( This is the same as that with mosaic processing. This is because each 10x10 pixel has only been replaced, and its average value does not change.

  The present invention is suitable as a surveillance camera system for public places such as residential areas and parks.

It is an explanatory diagram of a method of displaying different mosaic images with different strengths by distributing different decryption devices to the viewing rights holder and inputting the decryption key and the encrypted image. Distributing a different image browsing key for each decryption device and viewing rights to the viewing right holder, and inputting the decryption key and the encrypted image to display a mosaic image with different strengths. It is an explanatory diagram. It is explanatory drawing of the structure of the key for image browsing. It is an explanatory diagram of a method of constructing image data that creates multiple mosaiced images with different intensities from images taken by a surveillance camera, encrypts them with different encryption keys, and combines them. It is explanatory drawing of the method of extracting and extracting an image from the image data produced | generated in FIG. Encryption when using message digest when creating multiple images with mosaics of different strengths from images taken with a surveillance camera and composing them by encrypting and combining each with a different encryption key It is an explanatory diagram of the method of conversion. Decoding when using message digest when creating multiple images with mosaics of different strengths from images taken by a surveillance camera and composing image data by encrypting and combining them with different encryption keys It is an explanatory diagram of the method of conversion. It is an explanatory diagram of the method of adding the key corresponding to the viewing right person to the image data and encrypting it and saving it. It is an explanatory diagram of a method of generating stored data by dividing an image, performing a blurring process for each divided part, and further performing a blurring process on the entire image.

Claims (8)

  An image encryption / decryption method characterized by outputting a blurred image according to a predetermined rule when restoring an encrypted image with a release key.   2. The image encryption and decryption according to claim 1, wherein there are a plurality of devices for restoring an encrypted image with a release key, and each device outputs an image subjected to a blurring process having different intensities. Method.   2. The image encryption according to claim 1, wherein after the encrypted image is restored with the release key, another key is required for browsing, and the image is subjected to a smearing process with intensity according to the key and output. And decoding method.   2. The image encryption according to claim 1, wherein there are a plurality of decryption keys capable of decrypting the encrypted image, and images having different blurring intensities are restored according to the decryption keys. Decryption method.   5. The image encryption and decryption method according to claim 1, wherein the blurring intensity is adjusted according to the location of the image.   6. The image encryption and decryption method according to claim 1, wherein mosaicing is used as a blurring technique. The image encryption and decryption method according to claim 1, 3, 4, 5, or 6, wherein a key related to the strength of viewing rights is embedded in an image to be encrypted. The image is divided into sections that are rougher than the pixels, and the pixels are replaced based on the encryption key in the sections to perform blurring or encryption. 7. Image encryption and decryption method.