CN114025073B - Method and device for extracting hardware fingerprint of camera - Google Patents
Method and device for extracting hardware fingerprint of camera Download PDFInfo
- Publication number
- CN114025073B CN114025073B CN202111374046.6A CN202111374046A CN114025073B CN 114025073 B CN114025073 B CN 114025073B CN 202111374046 A CN202111374046 A CN 202111374046A CN 114025073 B CN114025073 B CN 114025073B
- Authority
- CN
- China
- Prior art keywords
- camera
- hardware fingerprint
- pictures
- quality
- hardware
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/50—Constructional details
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
- H04N23/64—Computer-aided capture of images, e.g. transfer from script file into camera, check of taken image quality, advice or proposal for image composition or decision on when to take image
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
- H04N23/66—Remote control of cameras or camera parts, e.g. by remote control devices
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
- H04N23/66—Remote control of cameras or camera parts, e.g. by remote control devices
- H04N23/661—Transmitting camera control signals through networks, e.g. control via the Internet
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Studio Devices (AREA)
- Collating Specific Patterns (AREA)
Abstract
The disclosure provides a method and a device for extracting a hardware fingerprint of a camera. The method comprises the following steps: acquiring a first group of pictures shot by the camera; dividing the first group of pictures to obtain training pictures and test pictures; extracting a first hardware fingerprint of the camera from the training picture; detecting whether the quality of the first hardware fingerprint is qualified or not by using the test picture; and if the quality of the first hardware fingerprint is unqualified, sending first prompt information to a user of the camera so as to prompt the user to take pictures in a supplementary mode by using the camera.
Description
Technical Field
The disclosure relates to the technical field of hardware fingerprints, in particular to a method and a device for extracting hardware fingerprints of a camera.
Background
The hardware fingerprint of the camera may be extracted from the picture taken using the camera. The quality of camera hardware fingerprint receives the quality influence of picture great. If the quality of the picture is poor, the quality of the extracted hardware fingerprint will also be poor. When a hardware fingerprint with poor quality is used for verification, there is a problem that the false recognition rate (false acceptance rate, FAR) or the rejection rate (false rejection rate, FAR) is high.
Disclosure of Invention
In view of this, the present disclosure provides a method and apparatus for extracting a fingerprint of camera hardware, so as to solve the problem of the influence of the quality of a picture on the quality of the fingerprint of camera hardware.
In a first aspect, the present disclosure proposes a method of extracting a camera hardware fingerprint. The method comprises the following steps: acquiring a first group of pictures shot by the camera; dividing the first group of pictures to obtain training pictures and test pictures; extracting a first hardware fingerprint of the camera from the training picture; detecting whether the quality of the first hardware fingerprint is qualified or not by using the test picture; and if the quality of the first hardware fingerprint is unqualified, sending first prompt information to a user of the camera so as to prompt the user to take pictures in a supplementary mode by using the camera.
Optionally, the detecting whether the quality of the first hardware fingerprint is acceptable by using the test picture includes: calculating the correlation between the first hardware fingerprint and the test picture; if the correlation is greater than a preset threshold, determining that the quality of the first hardware fingerprint is qualified; and if the correlation is smaller than or equal to the preset threshold value, determining that the quality of the first hardware fingerprint is unqualified.
Optionally, the correlation is an average peak correlation energy of the first hardware fingerprint and the test picture.
Optionally, the method further comprises: acquiring a second group of pictures which are complementarily shot by the user by utilizing the camera; updating the training picture and the test picture with the second set of pictures; extracting a second hardware fingerprint of the camera from the updated training picture; determining a difference in fingerprints of corresponding regions of the first hardware fingerprint and the second hardware fingerprint; determining an uncertain region in the visual range of the camera according to the difference of the fingerprints; and sending second prompt information to the user so as to prompt the user to promote the shooting quality of the uncertain region.
Optionally, the method further comprises: calculating target parameters of the uncertain region, wherein the target parameters comprise at least one of brightness, saturation and color phase of the uncertain region; wherein the second prompt information contains information prompting an improvement direction of the target parameter of the uncertain region.
Optionally, the method further comprises: and displaying the uncertain region and the second prompt information on a display screen corresponding to the camera in a visual mode.
Optionally, the method further comprises: and if the quality of the first hardware fingerprint is qualified, storing the first hardware fingerprint to a blockchain system.
Optionally, the extracting the first hardware fingerprint of the camera from the training picture includes: and extracting the first hardware fingerprint from the training picture by adopting a photo response non-uniformity analysis or metric learning mode.
In a second aspect, the present disclosure proposes an apparatus for extracting a hardware fingerprint of a camera, including: the first acquisition unit is used for acquiring a first group of pictures shot by the camera; the dividing unit is used for dividing the first group of pictures to obtain training pictures and test pictures; the first extraction unit is used for extracting a first hardware fingerprint of the camera from the training picture; the detection unit is used for detecting whether the quality of the first hardware fingerprint is qualified or not by using the test picture; and the first sending unit is used for sending first prompt information to a user of the camera to prompt the user to use the camera to supplement and shoot pictures if the quality of the first hardware fingerprint is unqualified.
Optionally, the detection unit includes: the first calculating unit is used for calculating the correlation between the first hardware fingerprint and the test picture; a judgment unit including: if the correlation is greater than a preset threshold, determining that the quality of the first hardware fingerprint is qualified; and if the correlation is smaller than or equal to the preset threshold value, determining that the quality of the first hardware fingerprint is unqualified.
Optionally, the correlation is an average peak correlation energy of the first hardware fingerprint and the test picture.
Optionally, the apparatus further comprises: the second acquisition unit is used for acquiring a second group of pictures which are complementarily shot by the user by using the camera; an updating unit configured to update the training picture and the test picture with the second set of pictures; the second extraction unit is used for extracting a second hardware fingerprint of the camera from the updated training picture; a comparison unit for determining a difference in fingerprints of corresponding regions of the first and second hardware fingerprints; the determining unit is used for determining an uncertain region in the visual range of the camera according to the difference of the fingerprints; and the second sending unit is used for sending second prompt information to the user so as to prompt the user to improve the shooting quality of the uncertain region.
Optionally, the apparatus further comprises: a second calculation unit configured to calculate a target parameter of the uncertain region, the target parameter including at least one of brightness, saturation, and color phase of the uncertain region; wherein the second prompt information contains information prompting an improvement direction of the target parameter of the uncertain region.
Optionally, the apparatus further comprises: and the display unit is used for displaying the uncertain region and the second prompt information on a display screen corresponding to the camera in a visual mode.
Optionally, the apparatus further comprises: and the storage unit is used for storing the first hardware fingerprint into the blockchain system if the quality of the first hardware fingerprint is qualified.
Optionally, the first extraction unit includes: and the third extraction unit is used for extracting the first hardware fingerprint from the training picture in a photo response non-uniformity analysis or measurement learning mode.
In a third aspect, the present disclosure provides an apparatus comprising a memory having executable code stored therein and a processor configured to execute the executable code to implement the method of the first aspect.
According to the method, in the process of extracting the hardware fingerprints of the camera, the quality of the hardware fingerprints can be synchronously judged, and when the quality is unqualified, a user is prompted to supplement and shoot pictures, so that the quality of the pictures for extracting the hardware fingerprints is improved, and the high-quality hardware fingerprints of the camera can be extracted.
Drawings
Fig. 1 is a schematic flowchart of a method for extracting a hardware fingerprint of a camera according to an embodiment of the disclosure.
Fig. 2 is a schematic flowchart of a method for determining whether a hardware fingerprint quality is acceptable according to an embodiment of the disclosure.
Fig. 3 is a schematic flowchart of a method for prompting a user to supplement a picture according to an embodiment of the present disclosure.
Fig. 4 is an exemplary diagram for dividing a camera visual range according to an embodiment of the present disclosure.
Fig. 5 is an exemplary diagram of a visual cue corresponding to the region shown in fig. 4.
Fig. 6 is a schematic flowchart of another method for extracting a hardware fingerprint of a camera according to an embodiment of the disclosure.
Fig. 7 is a schematic structural diagram of an apparatus for extracting a hardware fingerprint of a camera according to an embodiment of the disclosure.
Fig. 8 is a schematic structural diagram of another device for extracting a hardware fingerprint of a camera according to an embodiment of the disclosure.
Detailed Description
The following description of the technical solutions in the embodiments of the present disclosure will be made clearly and completely with reference to the accompanying drawings of the embodiments of the present disclosure, and it is apparent that the described embodiments are only some embodiments of the present disclosure, not all embodiments.
Cameras are a common type of image acquisition device that can be used to acquire photographs or videos. The camera is widely applied to intelligent equipment such as monitoring probes, mobile phones or tablet computers.
The camera can retain various noise while imaging. The noise may include hardware noise of the camera, software processing noise, and the like. Some of the noises can be used as identification marks of one camera and other cameras, namely hardware fingerprints of the cameras. Noise that may be a camera hardware fingerprint may include hardware noise or noise of a software process, or the like. The hardware noise may include: natural deviations of the sensor or random white noise, etc. The software-processed noise may be, for example, brought in by image processing software, which may be understood as a filter of some kind.
The hardware fingerprint of the camera has wide application. For example, a hardware fingerprint of a camera may solve the problem of trustworthiness of the data source. That is, the camera, as an image capturing device, is a source of image data, and can verify the authenticity of an image by using a hardware fingerprint. Alternatively, the hardware fingerprint of the camera may also be used for device identity authentication. As an embodiment, the hardware fingerprint of the camera may be used in an authentication scenario such as mobile phone Application (APP), so as to improve the security level.
The hardware fingerprint application for a camera typically includes an extraction phase and a verification phase.
In the extraction stage, the hardware fingerprint of the camera can be extracted based on the picture shot by the camera. In the verification stage, the extracted hardware fingerprint can be compared with the picture to be verified, so that whether the picture to be verified is shot by the camera or not is verified, and functions of verification, identity authentication and the like are realized.
The hardware fingerprint of the camera can be extracted based on photo response non-uniformity analysis (photo response non-uniformity, PRNU) or metric learning.
The Toothpic company provides a method for extracting the hardware fingerprint of the camera based on the PRNU. The method comprises the following steps: first, a picture in RAW format is subjected to high-frequency filtering, thereby extracting high-frequency components of PRNU. The high frequency component of the PRNU can be used as a hardware fingerprint of the camera. Secondly, compressing the hardware fingerprint subset after the isolation points are removed, wherein the compressed hardware fingerprint can be used as a comparison basis for identity consistency. The method has the advantage that the original fingerprint cannot be reversely deduced.
Metric learning is a learning framework based on deep learning. The main process of metric learning consists in learning a difference function between two information representations. For the camera hardware fingerprint extraction method based on metric learning, the main process is to construct a distance function of hardware fingerprint similarity.
In the verification stage, the picture to be verified can be verified through indexes such as confidence level and the like. The confidence may include, for example: peak correlation energy (peak to correlation energy, PCE). The higher the PCE, the higher the confidence, and the higher the likelihood that the picture to be verified is taken by the camera.
Taking a mobile phone with a camera as an example, in the extraction stage, a user can take a plurality of pictures or a video through the mobile phone. The mobile phone can extract the hardware fingerprint of the mobile phone camera through PRNU or metric learning and other technologies. In the verification stage, the mobile phone can calculate the picture to be verified and the PCE of the hardware fingerprint extracted in the extraction stage. If the value of the PCE is greater than a certain threshold, determining that the picture to be verified is shot by the camera, and if the value of the PCE is less than the certain threshold, determining that the picture to be verified is not shot by the camera.
The quality of the camera hardware fingerprint extracted in the extraction stage directly influences the effect of subsequent verification. The effectiveness of verification is typically measured using an index such as a false recognition rate or a rejection rate. The false recognition rate may represent the rate of false recognition of fingerprints by camera hardware during the verification phase. The rejection rate may represent the proportion of false rejects of fingerprints by the camera hardware during the verification phase. It will be appreciated that the effect of the verification may reflect the quality of the extracted hardware fingerprint. For example, the higher the false recognition rate or rejection rate, the poorer the quality of the extracted hardware fingerprint.
The quality of the hardware fingerprint is greatly affected by the picture quality used to extract the hardware fingerprint. The picture quality is affected by picture parameters. The parameters may include, for example: brightness, saturation, hue, etc. When the parameters of the picture for extracting the hardware fingerprint are proper, the quality of the picture is higher for the camera, and the quality of the extracted hardware fingerprint is higher. When the parameters of the picture used to extract the hardware fingerprint are inappropriate, the quality of the picture is low for the camera, and the quality of the hardware fingerprint is low.
The parameters of the picture from which the high quality hardware fingerprint can be extracted are different for different cameras. That is, for one camera, a high quality hardware fingerprint may be obtained based on some pictures, but for another camera, the quality of the hardware fingerprint extracted from pictures taken with the same parameters may be poor. It will be appreciated that the pictures used to extract the hardware fingerprints are difficult to be just high quality pictures suitable for the camera, and the use of these pictures for hardware fingerprint extraction tends to create the problem of poor quality of the hardware fingerprints.
The present disclosure provides a method for extracting a hardware fingerprint of a camera to improve quality of extracting the hardware fingerprint.
Fig. 1 is a schematic flow chart of a method for extracting a camera hardware fingerprint according to an embodiment of the disclosure.
The method shown in fig. 1 may be implemented by a device having computing capabilities. For example, the device may be a camera with computing capabilities, or a cell phone, computer, etc. that includes a camera.
The method shown in fig. 1 includes steps S110 to S150.
Step S110, a first group of pictures shot by a camera is acquired.
The present disclosure does not limit the format of the picture. For example, the format of the picture may be RAW or JPEG.
The first set of pictures may include a plurality of pictures. Since hardware fingerprints are typically very weak noise signals. Therefore, the hardware fingerprints of the camera are extracted through a plurality of pictures, so that the extraction of the hardware fingerprints is more reliable.
The picture can be a photo shot by the camera, or can be a video frame in a video shot by the camera.
The present disclosure does not limit the manner in which the picture is obtained. For example, the pictures can be obtained in real time during the shooting process of the user by using the camera, and also can be obtained from the pictures stored in the history. As an embodiment, in the process of fingerprint extraction, the camera is started to prompt a user to use the camera to shoot, and the picture shot by the camera can be acquired in the process of shooting by the user by using the camera.
Step S120, dividing the first group of pictures to obtain training pictures and test pictures.
The training pictures may include one or more pictures, and the test pictures may also include one or more pictures, as this disclosure is not limited in this regard.
The dividing mode of the pictures can be flexibly selected according to actual conditions. For example, the segmentation may be performed randomly and/or in proportion. As an example, a proportion (e.g., 80%) of the pictures in the first set of pictures may be randomly selected, the pictures divided into training pictures, and the remainder (e.g., 20%) divided into test pictures.
Step S130, a first hardware fingerprint of the camera is extracted from the training pictures.
The present disclosure is not limited to methods of extracting camera hardware fingerprints. For example, the PRNU-based algorithm described above may be used, or a metric learning-based method may be used.
Step S140, detecting whether the quality of the first hardware fingerprint is acceptable or not by using the test picture.
The quality of a hardware fingerprint of a camera may refer to whether the hardware fingerprint can accurately identify pictures taken by most cameras. When the quality of the hardware fingerprint of the camera is qualified, the hardware fingerprint can correctly identify most pictures shot by the camera. When the quality of the hardware fingerprint of the camera is unqualified, the hardware fingerprint cannot correctly identify most of pictures shot by the camera. That is, if the first hardware fingerprint can accurately verify that the test picture is taken by the camera, it can be determined that the quality of the first hardware fingerprint is qualified. Otherwise, it may be determined that the quality of the first hardware fingerprint is not acceptable.
As one implementation, fig. 2 is a schematic flow chart of a method of determining whether a quality of a first hardware fingerprint is acceptable based on a correlation of the first hardware fingerprint and a test picture. Step S140 may include steps S141 to S143 as shown in fig. 2.
In step S141, a correlation between the first hardware fingerprint and the test picture is calculated.
The correlation may be an indicator of whether the hardware fingerprint is acceptable. It can be understood that the evaluation of whether the picture is an index taken by the camera in the verification stage can be used to calculate the correlation.
As one embodiment, the correlation may be computed by the PCE. For example, the test picture comprises a plurality of pictures, and the correlation may be the first hardware fingerprint and an average PCE of the test pictures. That is, PCEs for each of the first hardware fingerprint and the test picture may be calculated separately, resulting in a plurality of PCEs, and the PCEs averaged. The correlation may be the average of the PCEs obtained as described above.
In step S142, if the correlation is greater than the preset threshold, it is determined that the quality of the first hardware fingerprint is acceptable.
In step S143, if the correlation is less than or equal to the preset threshold, it is determined that the quality of the first hardware fingerprint is not acceptable.
Alternatively, the preset threshold may be a fixed value, or may be dynamically selected according to the actual situation, which is not limited in the present disclosure.
And step S150, if the quality of the first hardware fingerprint is unqualified, sending first prompt information to a user of the camera so as to prompt the user to take pictures by using the camera in a supplementing manner.
The additional shot pictures can be used to re-extract the camera hardware fingerprint. The process of re-extracting the hardware fingerprint may be to repeat steps S110 to S150. Multiple iterations of steps S110-S150 may enable multiple rounds of iterative interactions between the user and the device performing the method. Each round of the image taking device can take supplementary images by the user, and the executing device extracts the hardware fingerprints and judges the quality of the hardware fingerprints so as to prompt the user to take the images in the next round of the supplementary images. This is repeated iteratively until the quality of the extracted hardware fingerprint is acceptable.
In order to reduce the number of iterations, the following steps may be included before step S110: the user is prompted with the basic criteria of the picture that needs to be provided. For example, when the picture is completely black, noise associated with the hardware fingerprint of the camera is almost completely filtered, and it is difficult to extract the camera hardware fingerprint. Thus, the user can be prompted to avoid taking a full black picture.
The present disclosure tests the quality of hardware fingerprints at the camera hardware fingerprint extraction stage. And when the quality of the hardware fingerprint is unqualified, prompting the user that the camera needs to be used for supplementing and taking pictures. An "active learning" strategy is implemented. That is, in the process of extracting the hardware fingerprint of the camera, the method can synchronously judge the quality of the hardware fingerprint, and prompt the user to supplement and shoot the picture when the quality is unqualified, so that the quality of the picture for extracting the hardware fingerprint is improved, and the high-quality hardware fingerprint of the camera is extracted.
It will be appreciated that the processing speed of the method proposed by the present disclosure may be fast when the computing power is sufficiently high. In the process of single shooting of the user, repeated iterative interaction can be completed, and the user is prompted for a plurality of times in real time. For a user, only one shooting is needed to extract and match high-quality hardware fingerprints. Therefore, the scheme can improve the experience of the user.
When the quality of the extracted hardware fingerprint is acceptable, the hardware fingerprint may be stored in the blockchain system.
The blockchain system has the characteristic of trusted memory card, so that the blockchain system is widely applied to the storage of key data. The blockchain system may include, for example, an ant chain (antchailstack).
The block chain is used for storing the hardware fingerprints, so that the safety of the hardware fingerprints can be further improved. In combination with the blockchain, the method for extracting the hardware fingerprint of the camera can be applied to a trusted end-edge product based on a blockchain system. The service landing scene applying the method can comprise the following steps: the authenticity of various attributes corresponding to the abstract model on the chain is guaranteed. The attributes of the abstract model may include: financial properties (e.g., properties of the mortgage goods), legal properties (e.g., properties of facts, evidence), value properties (e.g., properties of merchandise and raw materials), and the like.
For a blockchain, the method for extracting the hardware fingerprint provided by the disclosure has important application value for realizing the trusted data acquisition of the blockchain source. In addition, the method provided by the disclosure can communicate asset forms and physical entities defined by the blockchain technology in the digital world, so that corresponding business values can be generated by combining business models of various industries.
When the quality of the hardware fingerprint is unqualified, the user can be prompted to supplement the shot picture, and also can be prompted to adjust shot light, shot content or shot color and the like, so that parameters such as brightness, saturation, hue and the like of the picture are adjusted, the quality of the picture for extracting the camera hardware fingerprint is higher, and further the hardware fingerprint with higher quality is extracted.
As an example, the overall condition of shooting may be prompted. For example, the first prompt message may include a message such as "please adjust shooting light", "please change shooting angle", etc.
As another example, the user may be directed specifically how to take the improved picture. For example, the user may be prompted for an area to adjust, parameters to adjust, how to adjust the parameters, and so on.
In order to purposefully guide the user to take the improved picture, the method for extracting the hardware fingerprint of the camera according to the embodiment of the present disclosure may further include steps S310 to S350 shown in fig. 3.
Step S310, obtaining a second group of pictures which are complementarily shot by the user by using the camera.
It will be appreciated that the second set of pictures, like the first set of pictures, may include a plurality of pictures. The format of the picture can be RAW or JPEG.
Step S320, updating the training picture and the test picture by using the second set of pictures.
The second group of pictures may be divided, and the divided pictures may be updated to the training picture and the test picture, respectively. The present disclosure is not limited to the manner in which the second set of pictures is partitioned, e.g., may be scaled and/or randomly partitioned.
The updating process may be to supplement the second set of pictures into existing training pictures and test pictures. The number of training pictures can be increased by supplementing the training pictures, and the distribution of parameters of the pictures can be more comprehensive, which is also beneficial to improving the quality of hardware fingerprints. The supplement of the test pictures can enable the detection of the quality of the hardware fingerprints to be more reliable, so that the accuracy of the detection of the quality of the hardware fingerprints is improved.
It will be appreciated that the update process may also use the second set of pictures to replace existing training pictures and test pictures.
And step S330, extracting a second hardware fingerprint of the camera from the updated training picture.
Step S340, determining a difference between fingerprints of corresponding regions of the first and second hardware fingerprints.
Alternatively, the difference in values of the corresponding locations of the first and second hardware fingerprints may be compared. Taking the first hardware fingerprint and the second hardware fingerprint as the matrix corresponding to the picture shot by the camera as an example, the difference between the elements at the corresponding positions in the matrix can be calculated to determine the difference of the fingerprints of the corresponding areas.
Step S350, determining an uncertain region within the range of the camera according to the difference of the fingerprints.
An uncertainty region may be understood as a region where improvement is desired. In other words, in the visual range of the camera, parameters of some areas are more proper, parameters of some areas are less proper, and the quality of the shot picture can be improved only by improving the areas with unsuitable parameters.
Alternatively, the fingerprint region of the hardware fingerprint may be divided into a plurality of regions. Among the plurality of regions, it is possible to determine which region or regions is/are the uncertainty region based on the above-described difference. For example, when the fingerprint difference of a certain area is large, it may be determined as an uncertainty area.
The present disclosure does not limit the number of divided regions. It will be appreciated that the uncertainty region may be used to alert the user to the location of the region where improvement is desired. The more regions are divided, the smaller the area of each region, the finer the division of the fingerprint region, and the more accurate the boundary of the uncertainty region is determined.
As one embodiment, the fingerprint region may be divided into a plurality of regions in a grid shape. As shown in fig. 4, the fingerprint area may be divided into 3 rows and 3 columns (3×3) of grid areas (e.g., 9 areas of the thickened frame in fig. 4). For convenience of description, each mesh region is hereinafter denoted by a [ line number, column number ] (line number and column number are marked on the left and upper sides in fig. 4). For example, [1,2] may represent a grid region of row 1 and column 2.
The plurality of regions may be further divided to simplify the computational complexity of determining whether a region is an uncertain region. For example, each region may be further divided into a plurality of lattices. As shown in fig. 4, each mesh region is divided into 3×3 lattices. When the difference in hardware fingerprints in the lattice is large, then the lattice is marked as an uncertain lattice (in fig. 4, the uncertain lattice is marked by a gray lattice). When the number of uncertain lattices in one area is greater than or equal to a preset threshold value, the area may be marked as an uncertain area. For example, the predetermined threshold value may be 3, and the number of uncertain lattices in the [1,2] and [2,1] areas in fig. 4 is equal to 3, and may be determined as an uncertain area.
And step S360, sending second prompt information to the user so as to prompt the user to improve the shooting quality of the uncertain region.
The second prompt message may include a location of the uncertainty region to prompt the user to improve the uncertainty region.
The second prompt information can be visually displayed on a display screen corresponding to the camera so as to intuitively prompt the user. Fig. 5 is an exemplary diagram of a visual cue corresponding to the region shown in fig. 4. As shown in FIG. 5, the uncertainty regions (e.g., the [1,2] and [2,1] regions) of FIG. 4 may be highlighted to prompt the user to refine the region.
The method and the device can prompt the position of the uncertain region to the user through the steps, so that the user is guided to improve the uncertain region pertinently, and further, the user can quickly and efficiently supplement high-quality pictures.
The above describes how the location of the uncertainty region is determined. The application also proposes a method of determining an improved direction of a target parameter of an uncertainty region. By this method, the user can purposefully improve the area.
The target parameter may be, for example, a parameter related to the picture quality of the extracted hardware fingerprint, such as one or more of brightness, saturation, or hue. As an embodiment, the target parameter may be obtained by calculating an average value of brightness of a plurality of pixels of the uncertain region in the picture.
The direction of improvement may be, for example, the direction opposite to the calculated target parameter. For example, when the target parameter of the uncertainty region is low, the direction of improvement is to increase the parameter. When the target parameter of the uncertainty area is high, the direction of improvement is to decrease the parameter. The target parameter may be measured by a preset threshold value, or may be measured based on other determined areas, which is not limited in this disclosure.
The improvement direction of the target parameter of the uncertain region can be prompted to the user through second prompt information.
Continuing with the example of FIG. 4, if the average brightness of the calculated region [1,2] is lower and the hue is colder, the second hint information may hint that: the zones [1,2] need to be brighter and warmer. If the average brightness of the area [2,1] is lower and the hue is warmer, the second prompt message may prompt: the region [2,1] needs to be brighter and cooler.
The direction of improvement of the target parameter can also be prompted in a visual manner. For example, different colors, text, or logos may be used to indicate the direction of improvement of the target parameter.
Continuing with fig. 4 as an example, the second prompt may be as shown in fig. 5. The second hint information may be hinted by using text at the location of the uncertainty area. For example, a text prompt "brighter warmer needed" may be used at the location of zone [1,2 ]. The second hint may also be indicated by a color for the direction of improvement. For example, in a region where warmth is required, a warm color system such as red is used for presentation, or in a region where coolness is required, a cold color system such as blue is used for presentation.
In the following, taking a mobile phone camera as an example, a method for extracting a hardware fingerprint of the camera according to an embodiment of the disclosure is described. Fig. 6 is a method for extracting a hardware fingerprint of a mobile phone camera, which includes steps S610 to S690. The method may be performed by a cell phone comprising a camera.
Step S610, the camera is started.
Step S620, a first group of pictures is shot, and the first group of pictures are divided into two parts of training pictures and test pictures.
The first set of pictures may include a plurality of pictures, and the training pictures and the test pictures may also include a plurality of pictures. Multiple test pictures may constitute a test set and multiple training pictures may constitute a training set.
Step S630, based on the plurality of pictures in the training set, the hardware fingerprint of the camera is extracted, and the first hardware fingerprint is obtained.
Step S640, based on the multiple pictures in the test set, detects whether the hardware fingerprint meets the confidence requirement. That is, using the test picture, it is detected whether the quality of the first hardware fingerprint is acceptable.
If the hardware fingerprint does not meet the confidence requirement, go to step S680 and step S690.
In step S680, the hardware fingerprints satisfying the confidence requirement are registered in the blockchain system.
Step S690, the camera is turned off.
If the hardware fingerprint meets the confidence requirement, steps S650 to S670 are performed.
In step S650, an uncertain region within the camera' S visual range is detected.
Step S660, the uncertain region is prompted. The uncertainty area can be visually displayed on the display of the handset. The content of the hint may include the location of the uncertainty area and/or the direction of improvement of the target parameter of the uncertainty area.
The user can continue taking the supplementary picture according to the prompt.
Step S670, a second group of pictures taken by the user in a supplementary manner is acquired. And cutting the second group of pictures, and respectively supplementing the cut pictures into a training set and a testing set to update the training set and the testing set.
Based on the updated training set and the test set, multiple rounds of iterations may be performed. That is, steps S630 to S670 are repeated a plurality of times until the extracted hardware fingerprint satisfies the confidence requirement.
Method embodiments of the present disclosure are described in detail above in connection with fig. 1-6. Device embodiments of the present disclosure are described in detail below in conjunction with fig. 7-8. It is to be understood that the description of the method embodiments corresponds to the description of the device embodiments, and that parts not described in detail can therefore be seen in the preceding method embodiments.
Fig. 7 is a schematic block diagram of an apparatus 700 for extracting a hardware fingerprint of a camera according to an embodiment of the disclosure. The apparatus 700 may include: a first acquisition unit 710, a division unit 720, a first extraction unit 730, a detection unit 740, and a first transmission unit 750.
The first obtaining unit 710 may be configured to obtain a first set of pictures taken by the camera.
The dividing unit 720 may be configured to divide the first group of pictures to obtain a training picture and a test picture.
The first extraction unit 730 may be configured to extract a first hardware fingerprint of the camera from the training picture.
The detecting unit 740 may be configured to detect whether the quality of the first hardware fingerprint is acceptable by using the test picture.
The first sending unit 750 is configured to send, if the quality of the first hardware fingerprint is not acceptable, first prompt information to a user of the camera to prompt the user to make a supplemental picture with the camera.
Optionally, the detection unit includes: the first calculating unit is used for calculating the correlation between the first hardware fingerprint and the test picture; a judgment unit including: if the correlation is greater than a preset threshold, determining that the quality of the first hardware fingerprint is qualified; and if the correlation is smaller than or equal to the preset threshold value, determining that the quality of the first hardware fingerprint is unqualified.
Optionally, the correlation is an average peak correlation energy of the first hardware fingerprint and the test picture.
Optionally, the apparatus 700 further includes: the second acquisition unit is used for acquiring a second group of pictures which are complementarily shot by the user by using the camera; an updating unit configured to update the training picture and the test picture with the second set of pictures; the second extraction unit is used for extracting a second hardware fingerprint of the camera from the updated training picture; a comparison unit for determining a difference in fingerprints of corresponding regions of the first and second hardware fingerprints; the determining unit is used for determining an uncertain region in the visual range of the camera according to the difference of the fingerprints; and the second sending unit is used for sending second prompt information to the user so as to prompt the user to improve the shooting quality of the uncertain region.
Optionally, the apparatus 700 further includes: a second calculation unit configured to calculate a target parameter of the uncertain region, the target parameter including at least one of brightness, saturation, and color phase of the uncertain region; wherein the second prompt information contains information prompting an improvement direction of the target parameter of the uncertain region.
Optionally, the apparatus 700 further includes: and the display unit is used for displaying the uncertain region and the second prompt information on a display screen corresponding to the camera in a visual mode.
Optionally, the apparatus 700 further includes: and the storage unit is used for storing the first hardware fingerprint into the blockchain system if the quality of the first hardware fingerprint is qualified.
Optionally, the first extraction unit 730 includes: and the third extraction unit is used for extracting the first hardware fingerprint from the training picture by adopting a PRNU or metric learning mode.
Fig. 8 is a schematic view of an apparatus according to another embodiment of the present disclosure. The apparatus 800 may be, for example, a computing device having computing functionality. For example, the apparatus 800 may be a server. The apparatus 800 may include a memory 810 and a processor 820. Memory 810 may be used to store executable code. Memory 810 may also be used to store graph data. Processor 820 may be used to execute executable code stored in memory 810 to implement the steps in the various methods described previously. In some embodiments, the apparatus 800 may further include a network interface 830, and data exchange of the processor 820 with external devices may be achieved through the network interface 830.
In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, produces a flow or function in accordance with embodiments of the present disclosure, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by a wired (e.g., coaxial cable, fiber optic, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains an integration of one or more available media. The usable medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a magnetic tape), an optical medium (e.g., a digital video disc (digital video disc, DVD)), or a semiconductor medium (e.g., a Solid State Disk (SSD)), or the like.
Those of ordinary skill in the art will appreciate that the elements and algorithm steps described in connection with the embodiments disclosed herein can be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present disclosure.
In the several embodiments provided in the present disclosure, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of the units is merely a logical function division, and there may be additional divisions when actually implemented, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in each embodiment of the present disclosure may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit.
The foregoing is merely specific embodiments of the disclosure, but the protection scope of the disclosure is not limited thereto, and any person skilled in the art can easily think about changes or substitutions within the technical scope of the disclosure, and it is intended to cover the scope of the disclosure. Therefore, the protection scope of the present disclosure shall be subject to the protection scope of the claims.
Claims (15)
1. A method of extracting a hardware fingerprint of a camera, comprising:
acquiring a first group of pictures shot by the camera;
dividing the first group of pictures to obtain training pictures and test pictures;
extracting a first hardware fingerprint of the camera from the training picture;
detecting whether the quality of the first hardware fingerprint is qualified or not by using the test picture;
if the quality of the first hardware fingerprint is unqualified, sending first prompt information to a user of the camera so as to prompt the user to take pictures in a supplementary mode by using the camera;
acquiring a second group of pictures which are complementarily shot by the user by utilizing the camera;
updating the training picture and the test picture with the second set of pictures;
extracting a second hardware fingerprint of the camera from the updated training picture;
determining a difference in fingerprints of corresponding regions of the first hardware fingerprint and the second hardware fingerprint;
determining an uncertain region in the visual range of the camera according to the difference of the fingerprints;
and sending second prompt information to the user so as to prompt the user to promote the shooting quality of the uncertain region.
2. The method of claim 1, the detecting whether the quality of the first hardware fingerprint is acceptable using the test picture, comprising:
calculating the correlation between the first hardware fingerprint and the test picture;
if the correlation is greater than a preset threshold, determining that the quality of the first hardware fingerprint is qualified;
and if the correlation is smaller than or equal to the preset threshold value, determining that the quality of the first hardware fingerprint is unqualified.
3. The method of claim 2, the correlation being an average peak correlation energy of the first hardware fingerprint and the test picture.
4. The method of claim 1, the method further comprising:
calculating target parameters of the uncertain region, wherein the target parameters comprise at least one of brightness, saturation and color phase of the uncertain region;
wherein the second prompt information contains information prompting an improvement direction of the target parameter of the uncertain region.
5. The method of claim 1, the method further comprising:
and displaying the uncertain region and the second prompt information on a display screen corresponding to the camera in a visual mode.
6. The method of claim 1, the method further comprising:
and if the quality of the first hardware fingerprint is qualified, storing the first hardware fingerprint to a blockchain system.
7. The method of claim 1, the extracting a first hardware fingerprint of the camera from the training picture, comprising:
and extracting the first hardware fingerprint from the training picture by adopting a photo response non-uniformity analysis or metric learning mode.
8. An apparatus for extracting a hardware fingerprint of a camera, comprising:
the first acquisition unit is used for acquiring a first group of pictures shot by the camera;
the dividing unit is used for dividing the first group of pictures to obtain training pictures and test pictures;
the first extraction unit is used for extracting a first hardware fingerprint of the camera from the training picture;
the detection unit is used for detecting whether the quality of the first hardware fingerprint is qualified or not by using the test picture;
the first sending unit is used for sending first prompt information to a user of the camera to prompt the user to use the camera to supplement and shoot pictures if the quality of the first hardware fingerprint is unqualified;
the second acquisition unit is used for acquiring a second group of pictures which are complementarily shot by the user by using the camera;
an updating unit configured to update the training picture and the test picture with the second set of pictures;
the second extraction unit is used for extracting a second hardware fingerprint of the camera from the updated training picture;
a comparison unit for determining a difference in fingerprints of corresponding regions of the first and second hardware fingerprints;
the determining unit is used for determining an uncertain region in the visual range of the camera according to the difference of the fingerprints;
and the second sending unit is used for sending second prompt information to the user so as to prompt the user to improve the shooting quality of the uncertain region.
9. The apparatus of claim 8, the detection unit comprising:
the first calculating unit is used for calculating the correlation between the first hardware fingerprint and the test picture;
the judging unit is used for determining that the quality of the first hardware fingerprint is qualified if the correlation is larger than a preset threshold value;
and if the correlation is smaller than or equal to the preset threshold value, determining that the quality of the first hardware fingerprint is unqualified.
10. The device of claim 9, the correlation being an average peak correlation energy of the first hardware fingerprint and the test picture.
11. The apparatus of claim 8, the apparatus further comprising:
a second calculation unit configured to calculate a target parameter of the uncertain region, the target parameter including at least one of brightness, saturation, and color phase of the uncertain region;
wherein the second prompt information contains information prompting an improvement direction of the target parameter of the uncertain region.
12. The apparatus of claim 8, the apparatus further comprising:
and the display unit is used for displaying the uncertain region and the second prompt information on a display screen corresponding to the camera in a visual mode.
13. The apparatus of claim 8, the apparatus further comprising:
and the storage unit is used for storing the first hardware fingerprint into the blockchain system if the quality of the first hardware fingerprint is qualified.
14. The apparatus of claim 8, the first extraction unit comprising:
and the third extraction unit is used for extracting the first hardware fingerprint from the training picture in a photo response non-uniformity analysis or measurement learning mode.
15. An apparatus for extracting a hardware fingerprint of a camera, comprising a memory having executable code stored therein and a processor configured to execute the executable code to implement the method of any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111374046.6A CN114025073B (en) | 2021-11-18 | 2021-11-18 | Method and device for extracting hardware fingerprint of camera |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111374046.6A CN114025073B (en) | 2021-11-18 | 2021-11-18 | Method and device for extracting hardware fingerprint of camera |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114025073A CN114025073A (en) | 2022-02-08 |
| CN114025073B true CN114025073B (en) | 2023-09-29 |
Family
ID=80065539
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111374046.6A Active CN114025073B (en) | 2021-11-18 | 2021-11-18 | Method and device for extracting hardware fingerprint of camera |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114025073B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2010103672A (en) * | 2008-10-22 | 2010-05-06 | Ricoh Co Ltd | Imaging device |
| CN104639819A (en) * | 2013-11-07 | 2015-05-20 | 中兴通讯股份有限公司 | Portable terminal and method for taking pictures using same |
| CN109190469A (en) * | 2018-07-27 | 2019-01-11 | 阿里巴巴集团控股有限公司 | A kind of detection method and device, a kind of calculating equipment and storage medium |
| CN111861976A (en) * | 2020-05-20 | 2020-10-30 | 西安理工大学 | Digital image source shooting equipment identification method based on hardware fingerprint correlation |
| CN111951254A (en) * | 2020-08-18 | 2020-11-17 | 济南大学 | Source camera identification method and system based on edge-guided weighted average |
| US10868984B1 (en) * | 2018-11-15 | 2020-12-15 | Government of the United States as represented by Director, National Security Agency | Method for estimating an improved camera fingerprint by identifying low-mass pixel positions and correcting corresponding fingerprint values |
| CN113537088A (en) * | 2021-07-20 | 2021-10-22 | 杭州人云数字科技有限公司 | Digital image sensor hardware fingerprint matching method and system based on small sample learning |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3591961A4 (en) * | 2017-03-21 | 2020-03-25 | Huawei Technologies Co., Ltd. | Photographing guide method, device and system |
-
2021
- 2021-11-18 CN CN202111374046.6A patent/CN114025073B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2010103672A (en) * | 2008-10-22 | 2010-05-06 | Ricoh Co Ltd | Imaging device |
| CN104639819A (en) * | 2013-11-07 | 2015-05-20 | 中兴通讯股份有限公司 | Portable terminal and method for taking pictures using same |
| CN109190469A (en) * | 2018-07-27 | 2019-01-11 | 阿里巴巴集团控股有限公司 | A kind of detection method and device, a kind of calculating equipment and storage medium |
| US10868984B1 (en) * | 2018-11-15 | 2020-12-15 | Government of the United States as represented by Director, National Security Agency | Method for estimating an improved camera fingerprint by identifying low-mass pixel positions and correcting corresponding fingerprint values |
| CN111861976A (en) * | 2020-05-20 | 2020-10-30 | 西安理工大学 | Digital image source shooting equipment identification method based on hardware fingerprint correlation |
| CN111951254A (en) * | 2020-08-18 | 2020-11-17 | 济南大学 | Source camera identification method and system based on edge-guided weighted average |
| CN113537088A (en) * | 2021-07-20 | 2021-10-22 | 杭州人云数字科技有限公司 | Digital image sensor hardware fingerprint matching method and system based on small sample learning |
Non-Patent Citations (2)
| Title |
|---|
| 一种基于硬件指纹的混合加密文件保护系统;刘庆俞;陈磊;刘磊;山东农业工程学院学报;第第36卷卷(第第1期期);全文 * |
| 基于相机溯源的潜在不良视频通话预警;马晓晨;韦世奎;蒋翔;李晓飞;;光学精密工程(第11期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114025073A (en) | 2022-02-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Cozzolino et al. | Noiseprint: A CNN-based camera model fingerprint | |
| CN108229120B (en) | Face unlocking method, face unlocking information registration device, face unlocking information registration equipment, face unlocking program and face unlocking information registration medium | |
| WO2020151489A1 (en) | Living body detection method based on facial recognition, and electronic device and storage medium | |
| US20230116801A1 (en) | Image authenticity detection method and device, computer device, and storage medium | |
| EP2866170B1 (en) | Image processing device and image processing method | |
| CN107026842B (en) | Method and device for generating security problem and verifying identity | |
| WO2020259128A1 (en) | Liveness detection method and apparatus, electronic device, and computer readable storage medium | |
| CN106165391A (en) | The image capturing strengthened | |
| US11856285B2 (en) | Systems and methods for adjusting a monitoring device | |
| CN112153373A (en) | Fault identification method and device for bright kitchen range equipment and storage medium | |
| CN111259757B (en) | Living body identification method, device and equipment based on image | |
| CN111767879A (en) | Living body detection method | |
| CN113642639B (en) | Living body detection method, living body detection device, living body detection equipment and storage medium | |
| CN114387548A (en) | Video and liveness detection method, system, device, storage medium and program product | |
| CN114025073B (en) | Method and device for extracting hardware fingerprint of camera | |
| You et al. | Tampering detection and localization base on sample guidance and individual camera device convolutional neural network features | |
| WO2022222957A1 (en) | Method and system for identifying target | |
| CN110135204A (en) | Data guard method and device | |
| CN110084076A (en) | Evidence display methods and device | |
| CN111767829B (en) | Living body detection method, device, system and storage medium | |
| CN115240245A (en) | Face living body detection method, face living body detection device and electronic equipment | |
| CN116152932A (en) | Living body detection method and related equipment | |
| CN113705428A (en) | Living body detection method and apparatus, electronic device, and computer-readable storage medium | |
| US20210067683A1 (en) | Flat surface detection in photographs | |
| JP6148785B1 (en) | Information processing system, information processing apparatus, and program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |