CN112288757B - Encryption domain image segmentation optimization method based on data packing technology - Google Patents
Encryption domain image segmentation optimization method based on data packing technology Download PDFInfo
- Publication number
- CN112288757B CN112288757B CN202011197779.2A CN202011197779A CN112288757B CN 112288757 B CN112288757 B CN 112288757B CN 202011197779 A CN202011197779 A CN 202011197779A CN 112288757 B CN112288757 B CN 112288757B
- Authority
- CN
- China
- Prior art keywords
- image
- image segmentation
- execution server
- service provider
- encrypted
- 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
- 238000003709 image segmentation Methods 0.000 title claims abstract description 160
- 238000000034 method Methods 0.000 title claims abstract description 71
- 238000005516 engineering process Methods 0.000 title claims abstract description 33
- 238000012856 packing Methods 0.000 title claims abstract description 33
- 238000005457 optimization Methods 0.000 title claims abstract description 15
- 238000004364 calculation method Methods 0.000 claims abstract description 27
- 230000008569 process Effects 0.000 claims description 24
- 230000011218 segmentation Effects 0.000 claims description 23
- 238000001914 filtration Methods 0.000 claims description 17
- 238000004806 packaging method and process Methods 0.000 claims description 10
- 230000001186 cumulative effect Effects 0.000 claims description 9
- 230000003993 interaction Effects 0.000 claims description 8
- 238000012986 modification Methods 0.000 claims description 8
- 230000004048 modification Effects 0.000 claims description 8
- 230000005540 biological transmission Effects 0.000 claims description 7
- 238000003708 edge detection Methods 0.000 claims description 4
- 238000010276 construction Methods 0.000 claims description 3
- 238000013139 quantization Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 239000000654 additive Substances 0.000 description 2
- 230000000996 additive effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012886 linear function Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012536 packaging technology Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/12—Edge-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/60—Protecting data
- G06F21/602—Providing cryptographic facilities or services
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T1/00—General purpose image data processing
- G06T1/0021—Image watermarking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioethics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Computer Hardware Design (AREA)
- Computer Security & Cryptography (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Computer And Data Communications (AREA)
- Storage Device Security (AREA)
Abstract
The invention provides an encryption domain image segmentation optimization method based on a data packing technology, relates to the technical field of encryption domain image segmentation optimization, and solves the problems that in the current image segmentation method, space resources are occupied and calculation complexity is high.
Description
Technical Field
The invention relates to the technical field of encryption domain image segmentation optimization, in particular to an encryption domain image segmentation optimization method based on a data packing technology.
Background
Along with the rapid development of social economy, the image segmentation technology has more and more application scenes in our life, including the fields of public security, traffic, hospitals and the like, and improves the efficiency of public safety and information management.
The existing encryption signal processing technology (SPED) provides a cloud server with a large amount of processing capacity for image data while ensuring privacy security. For example, the cloud server may perform the tasks of encrypting the image watermark, denoising, feature extraction, and the like. Edge detection and image segmentation are one of the important issues of image processing, however, image segmentation in a cloud computing scenario may cause privacy disclosure problems. From the aspect of data collection, users store privacy resources such as high-definition satellite images, medical images, traffic conditions and the like which need to be subjected to image segmentation on the terminal. In the process of image segmentation in a cloud environment, a client transmits privacy resources to a cloud server for edge recognition, and in the process, because the data transmitted at present are all encrypted data, the privacy data of a user are always leaked by the cloud server or actively or passively, and the potential safety risk brought by the leakage of the privacy data is serious, so that the privacy data can be used for profit by people, even more for criminals to capture, and used for illegal activities.
Currently, the Paillier cryptographic system and the DGK cryptographic system are the most popular additive homomorphic encryption systems, an image segmentation technology based on homomorphic encryption is widely cited, such as 3 months in 2018, zhang Minqing, li Tianxue, dirichness et al published an article (50 (001): 8-14) of a reversible information hiding algorithm based on the Paillier homomorphic public key encryption system in Zhengzhou university, an image owner encrypts a segmented image plane through a public key, then performs combination, an intermediate operator re-segments an image in a ciphertext domain, embeds the segmented image based on homomorphic multiplication, combines the embedded image, and transmits the segmented image to an image receiver for decryption, under the method, the intermediate operator can embed data through homomorphic multiplication without using a private key, and the image receiver decrypts the image through the private key received by a security channel, so that the privacy security of the image in the transmission segmentation process cannot be obtained by illegal hands, but on the one hand, the computation of the homomorphic encryption itself occupies a large calculation resource, the matrix operation is time-consuming under the encryption domain, and the requirement of the dimension cannot be met easily; on the other hand, in this method, although the private key is transferred by means of the secure channel, the encrypted image is not the same person when decrypted, and there is also a risk of privacy disclosure.
Disclosure of Invention
In order to solve the problems of large space resource occupation and high calculation complexity in the segmentation process of the traditional homomorphic encryption-based image segmentation method, the invention provides the encryption domain image segmentation method based on the data packing technology, which reduces the calculation complexity and the space resource occupation in the image segmentation process while guaranteeing the privacy security, thereby optimizing the segmentation speed and meeting the real-time requirement.
In order to achieve the technical effects, the technical scheme of the invention is as follows:
an encryption domain image segmentation optimization method based on a data packing technology at least comprises the following steps:
s1, a privacy service provider generates a public key pk and a private key sk, both the public key pk and the private key sk are sent to a client, and the private key sk is sent to an image segmentation execution server;
s2, the client encrypts and packages the image by using the public key pk, compresses the size of the encrypted image, and sends the encrypted and packaged image to the image segmentation execution server;
s3, the image segmentation execution server and the privacy service provider interact through multiparty security calculation and a messy code circuit technology to obtain an encrypted segmentation image;
s4, the image segmentation execution server sends the encrypted segmentation image to the client, and the client decrypts the encrypted segmentation image to obtain a final edge image.
In the technical scheme, the client encrypts the image by using the public key pk and packages the data, compresses the size of the encrypted image, reduces the computational complexity and the space resource occupation degree by using the data packaging technology, improves the image segmentation speed, and ensures the privacy requirements of the image segmentation execution server and the client by decrypting the encrypted image only by the image owner through homomorphic encryption.
Preferably, the data packing in step S2, the process of compressing the encrypted image size is:
the client encrypts the image by using the public key pk to form an encrypted imageSize s m ×s n Encryption image +.>Split into L-block pictures of the same size, denoted as: i 0 ，I 1 ，…，I L-1 The formula is performed:
wherein ,pair I using public key pk P Packaging into a size +.>Compressed image +.> Representing compressed image +.>The size parameters t, Q of (a) each represent a packing participation parameter,/->I k And (i, j) represents the pixel value of the image with the index of (i, j) in the kth image, the size of the image is compressed by a data packing technology, the space resource occupation degree is reduced, and the image segmentation speed is improved.
Preferably, after step S2, before step S3, the method further comprises: image segmentation execution server compresses imageGaussian filtering is performed, and the result after Gaussian filtering is expressed as +.> The method meets the following conditions:
reserving multiplication operation results, expressed as:
wherein m=0, 1, …, s m ；n＝0，1，…，s n All represent image size parameters, G is Gaussian filter template, Q is Gaussian filter template quantization parameter, s g Is the gaussian filtered template size.
In order to avoid unnecessary calculation, the calculation cost caused by division operation is reduced, the calculation complexity is reduced, the segmentation speed is increased, and the multiplication result is directly reserved without division operation during Gaussian filtering.
Preferably, the process of interaction between the image segmentation execution server and the privacy service provider through multiparty secure computation and scrambling circuit technology in step S3 includes: for Gaussian filtered encrypted packed dataThe steps of comparison are as follows:
s31, sobel filtering template G in horizontal direction x For example, the image segmentation execution server builds two temporary templates: the positive template coefficient part is expressed asAnd the part with negative template coefficient is denoted +.>The construction formula is as follows:
wherein , and />The parts of the horizontal Sobel convolution kernel where the coefficients are positive and negative,
s32, generating two random numbers alpha by the image segmentation execution server + 、α - And executing the formula to pack the two random numbers:
wherein ,representing a random number alpha + Performing data packaging to obtain a value; />Representing a random number alpha - The value after data packing is carried out, and based on the random number after packing, the image segmentation execution server generates intermediate ciphertext +_>Vertical->The following respectively satisfy:
the image segmentation execution server uses the intermediate ciphertextIs->Sending to a privacy service provider;
s33, the privacy service provider uses the private key sk to intermediate ciphertextIs->Decryption is carried out, segmentation is carried out, and ∈ -> and />And using a random number alpha 1 Generating a set { delta }, any one value delta in the set { delta }, and i the method meets the following conditions:
where i=0, 1, …, L-1, c is an additional variable, c e {0,1} is randomly chosen for the privacy service provider;
s34, according to the random number alpha + 、α - Image segmentation execution server calculates intermediate parameter sigma 0 Sigma (sigma) 1 The method comprises the following steps: sigma (sigma) 0 ＝α + -α - ，σ 1 ＝α - -α + Then by a random number alpha 2 Modification sigma i The formula is:
wherein ,representing modified sigma i I=0, 1; after performing a confusing transmission between the image division execution server and the privacy service provider, the image division execution server holds +.>Privacy service provider holds delta i ；
S35, creating a messy code circuit by the image segmentation execution server, generating an encrypted messy code circuit table GCT corresponding to the comparison circuit C with additional input, and decoding the messy code circuit table GCT by the other party to hold valuesDelta i Calculating the output +.>Encryption packaging data based on comparison scrambling code circuit>The image segmentation execution server obtains the comparison result r e {0,1}.
Here, since the comparison of the encrypted values cannot be realized only by using homomorphic encryption, the interaction between the image segmentation execution server and the privacy service provider is realized by introducing a disorder code circuit technology, and the linear function which cannot be calculated by calculating the additive homomorphic encryption is realized.
Preferably, the image segmentation execution server holds in the process of performing the confusing transmission between the image segmentation execution server and the privacy service provider and />The privacy service provider holds c E {0,1}, the image segmentation execution server cannot obtain the value of c, and the privacy service provider cannot obtain +.>Privacy service provider generation->And will->Transmitting to an image segmentation execution server, wherein the image segmentation execution server subtracts alpha 2 Acquisition->The privacy security is ensured.
Preferably, in step S3, the process of interaction between the image segmentation execution server and the privacy service provider through multiparty secure computation and scrambling circuit technology further includes: for Gaussian filtered encrypted packed dataPerforming Sobel filtering edge detection to obtain an encrypted image +.>The method comprises the following steps:
privacy service provider encrypts set delta to obtain encrypted setAnd sending the image to an image segmentation execution server;
for encrypted collectionsEach of +.>The image segmentation execution server with the comparison result r executes the formula to obtain +.>The formula is:
wherein ,α2 The random number generated by the execution server for the image segmentation in step S3 is based onThe image segmentation execution server may acquire horizontal component +.>And vertical component->Then according to the formula
Constructing Sobel filtering result as encrypted image
Preferably, in step S3, the process of interaction between the image segmentation execution server and the privacy service provider through multiparty secure computation and scrambling circuit technology further includes: acquiring an encrypted imageThe encryption threshold of (2) comprises the following steps:
s301, setting an encrypted imageThe current threshold is +.>The image segmentation execution service first sets an encrypted imageMaximum pixel value +.>And minimum pixel value +.>Then generating an initial threshold value using a division protocol>The formula is:
wherein SecDiv represents a division protocol;
s302, the image segmentation execution server performs segmentation according to the current threshold valueEncryption of images +.>Is divided into a foreground part and a background part, wherein the foreground part refers to a pixel with a value greater than T i The background portion means that the value of the pixel is less than or equal to T i ；
S303, utilizing a scrambling code circuit to perform current threshold valueAnd encryption pixel value +.>Comparing to obtain a comparison result { lambda };
s304, setting background cumulative value asThe foreground integrated value is +.>The background count value is +.>The foreground technical value is->Image segmentation execution server initializing background cumulative value +.>Foreground cumulative value +.>Background count value->Foreground count value->And performing pixel calculation, wherein the formula is as follows:
calculate the next thresholdThe formula is:
s305, the image segmentation execution server performs comparison of an absolute value scrambling code circuit together with a privacy service provider to obtainThe image segmentation execution server will->And delta+r 'is given to the privacy service provider, r' is a random value for modification, the privacy service provider judges whether the threshold calculation reaches a threshold standard epsilon, and if yes, the threshold iterative calculation is finished; otherwise, the process returns to step S304.
Preferably, step S303 is performed to the current threshold value by using a scrambling code circuitAnd encryption pixel value +.>The comparison process is as follows:
A. image segmentation execution server generates random number gamma 1 And beta epsilon {0,1}, based on random number gamma 1 Sum beta e {0,1}, calculateAnd will->To the privacy service provider, the formula is:
B. the privacy service provider willObtaining t 1 Simultaneously generating c epsilon {0,1}, holding t 1 And c privacy service provider and holding random number gamma 1 The image segmentation execution server of (1) executes a comparison scrambling code circuit with additional input, and the privacy service provider obtains modified result +.>
C. According toAnd c, the privacy service provider calculates the encryption pair t 2 The formula is:
the privacy service provider will t 2 Transmitting to an image segmentation execution server;
D. the image segmentation execution server confirms a final comparison result lambda according to beta: if β=1, the image segmentation execution server will t 2 Then, as a result λ, if β=0, the image segmentation execution server sets t 2 As a result lambda.
Preferably, in step S305, the image segmentation execution server compares with 0 when executing the absolute value scrambling circuit together with the privacy service provider.
Preferably, the image segmentation execution server has an encrypted imageEncryption image->Encryption threshold T of (2) i After that, the image segmentation execution server and the privacy service provider encrypt the image using the comparison scrambling circuit +.>Is binarized to obtain an encrypted segmented image +.>
Compared with the prior art, the technical scheme of the invention has the beneficial effects that:
the invention provides an encryption domain image segmentation optimization method based on a data packing technology, which comprises the steps that firstly, a privacy service provider generates a public key and a private key and sends the public key to a client, the private key is sent to an image segmentation execution server, then the client encrypts an image and packs data, the size of the encrypted image is compressed, the reduction of calculation complexity and space resource occupation is realized based on the data packing technology, the image segmentation speed is improved, the image segmentation execution server and the privacy service provider interact through multiparty security calculation and a messy code circuit technology, an encrypted segmentation image is obtained, the image segmentation execution server sends the encrypted segmentation image to the client for decryption, the encrypted image can only be decrypted by an image owner, the requirements of the image segmentation execution server and the client on privacy are guaranteed, and the image segmentation is ensured under the condition of privacy security.
Drawings
Fig. 1 shows a flow diagram of an encryption domain image segmentation optimization method based on a data packing technology according to an embodiment of the present invention;
FIG. 2 is a block diagram of a basic comparison scrambling code circuit according to an embodiment of the present invention;
FIG. 3 is a block diagram of a comparison scrambling code circuit with additional inputs as proposed in an embodiment of the present invention;
fig. 4 is a schematic diagram showing experimental results of encryption image segmentation by using an encryption domain image segmentation optimization method based on a data packing technology according to an embodiment of the present invention.
Detailed Description
The drawings are for illustrative purposes only and are not to be construed as limiting the present patent;
for better illustration of the present embodiment, some parts of the drawings may be omitted, enlarged or reduced, and do not represent actual dimensions;
it will be appreciated by those skilled in the art that some well known descriptions in the figures may be omitted.
The technical scheme of the invention is further described below with reference to the accompanying drawings and examples.
Example 1
Referring to fig. 1, as shown in a flowchart of an encryption domain image segmentation optimization method based on a data packing technology in fig. 1, the encryption domain image segmentation optimization method based on the data packing technology includes the following steps:
s1, a privacy service provider generates a public key pk and a private key sk, both the public key pk and the private key sk are sent to a client, and the private key sk is sent to an image segmentation execution server;
s2, the client encrypts and packages the image by using the public key pk, compresses the size of the encrypted image, and sends the encrypted and packaged image to the image segmentation execution server;
s3, the image segmentation execution server and the privacy service provider interact through multiparty security calculation and a messy code circuit technology to obtain an encrypted segmentation image;
s4, the image segmentation execution server sends the encrypted segmentation image to the client, and the client decrypts the encrypted segmentation image to obtain a final edge image.
In this embodiment, the process of packing the data and compressing the size of the encrypted image in step S2 is as follows:
the client encrypts the image by using the public key pk to form an encrypted imageSize of the deviceIs s m ×s n Encryption image +.>Split into L-block pictures of the same size, denoted as: i 0 ，I 1 ，…，I L-1 The formula is performed:
wherein ,pair I using public key pk P Packaging into a size +.>Compressed image +.> Representing compressed image +.>The size parameters t, Q of (a) each represent a packing participation parameter,/->s m 、s n Representing an encrypted image +.>Dimensional parameters of I k The pixel value of the image with the index (i, j) in the kth image is represented by (i, j), 2 bits are additionally reserved for preventing overflow, and the size of the image is compressed by a data packing technology, so that the space resource occupation degree is reduced, and the image segmentation speed is improved.
In this embodiment, after step S2, before step S3, the method further includes: image segmentationThe execution server compresses the imageGaussian filtering is performed, and the result after Gaussian filtering is expressed as +.>The method meets the following conditions:
reserving multiplication operation results, expressed as:
wherein m=0, 1, …, s m ；n＝0，1，…，s n All represent image size parameters, G is Gaussian filter template, Q is Gaussian filter template quantization parameter, s g For the size of the Gaussian filter template, namely, in order to avoid unnecessary calculation, the calculation cost caused by division operation is reduced, the calculation complexity is reduced, the segmentation speed is increased, and the division operation is not performed during Gaussian filtering, but the multiplication operation result is directly reserved.
In this embodiment, the process of interaction between the image segmentation execution server and the privacy service provider through the multiparty secure computation and scrambling circuit technique in step S3 includes: for Gaussian filtered encrypted packed dataThe steps of comparison are as follows:
s31, sobel filtering template G in horizontal direction x For example, the image segmentation execution server builds two temporary templates: the positive template coefficient part is expressed asAnd the part with negative template coefficient is denoted +.>The construction formula is as follows:
wherein , and />The parts of the horizontal Sobel convolution kernel where the coefficients are positive and negative,
s32, generating two random numbers alpha by the image segmentation execution server + 、α - And executing the formula to pack the two random numbers:
wherein ,representing the value of the random number alpha+ after data packaging; />Representing the value of the random number alpha_after data packing, and based on the packed random number, image segmentation execution clothesThe server generates intermediate ciphertext->Vertical->The following respectively satisfy:
the image segmentation execution server uses the intermediate ciphertextIs->Sending to a privacy service provider;
s33, the privacy service provider uses the private key sk to intermediate ciphertextIs->Decryption is carried out, segmentation is carried out, and ∈ -> and />And using a random number alpha 1 Generating a set { delta }, any one value delta in the set { delta }, and i the method meets the following conditions:
where i=0, 1, …, L-1, c is an additional variable, c e {0,1} is randomly chosen for the privacy service provider;
s34, according to the random number alpha + 、α _ Image segmentation execution server calculates intermediate parameter sigma 0 Sigma (sigma) 1 The method comprises the following steps: sigma (sigma) 0 ＝α + -α - ，σ 1 ＝α - -α + Then by a random number alpha 2 Modification sigma i The formula is:
wherein ,representing modified sigma i I=0, 1; after performing a confusing transmission between the image division execution server and the privacy service provider, the image division execution server holds +.>Privacy service provider holds delta i ；
S35, creating a messy code circuit by the image segmentation execution server, generating an encrypted messy code circuit table GCT corresponding to the comparison circuit C with additional input, and decoding the messy code circuit table GCT by the other party to hold valuesDelta i Calculating the output +.>Encryption packaging data based on comparison scrambling code circuit>The image segmentation execution server obtains the comparison result r e {0,1}. Execution confusing between image segmentation execution server and privacy service providerDuring the transmission, the image segmentation execution server holds +.> and />The privacy service provider holds c E {0,1}, the image segmentation execution server cannot obtain the value of c, and the privacy service provider cannot obtain +.>Privacy service provider generation->And will->Transmitting to an image segmentation execution server, wherein the image segmentation execution server subtracts alpha 2 Acquisition->The privacy security is ensured.
In this embodiment, the process of interacting between the image segmentation execution server and the privacy service provider through the multiparty secure computation and scrambling circuit technique in step S3 further includes: for Gaussian filtered encrypted packed dataPerforming Sobel filtering edge detection to obtain an encrypted image +.>The method comprises the following steps:
privacy service provider encrypts set delta to obtain encrypted setAnd sending the image to an image segmentation execution server;
for encrypted collectionsEach of +.>The image segmentation execution server with the comparison result r executes the formula to obtain +.>The formula is:
wherein ,α2 The random number generated by the execution server for the image segmentation in step S3 is based onThe image segmentation execution server may acquire horizontal component +.>And vertical component->Then according to the formula
Constructing Sobel filtering result as encrypted image
In this embodiment, the process of interacting between the image segmentation execution server and the privacy service provider through the multiparty secure computation and scrambling circuit technique in step S3 further includes: acquiring an encrypted imageThe encryption threshold of (2) comprises the following steps:
s301, setting an encrypted imageThe current threshold is +.>The image segmentation execution service first sets an encrypted imageAnd maximum pixel value->And minimum pixel value +.>Then generating an initial threshold value using a division protocol>The formula is:
wherein SecDiv represents a division protocol;
s302, the image segmentation execution server performs segmentation according to the current threshold valueEncryption of images +.>Is divided into a foreground part and a background part, wherein the foreground part refers to a pixel with a value greater than T i The background portion means that the value of the pixel is less than or equal to T i ；
S303, utilizing a scrambling code circuit to perform current threshold valueAnd encryption pixel value +.>Comparing to obtain a comparison result { lambda }; the current threshold value is +.>And encryption pixel value +.>The comparison process is as follows:
A. image segmentation execution server generates random number gamma 1 And beta epsilon {0,1}, based on random number gamma 1 Sum beta e {0,1}, calculateAnd will->To the privacy service provider, the formula is:
B. the privacy service provider willObtaining t 1 Simultaneously generating c epsilon {0,1}, holding t 1 And c privacy service provider and holding random number gamma 1 The image segmentation execution server of (1) executes a comparison scrambling code circuit with additional input, and the privacy service provider obtains modified result +.>
C. According toAnd c, privacy service provider meterCalculating the encryption pair t 2 The formula is:
the privacy service provider will t 2 Transmitting to an image segmentation execution server;
D. the image segmentation execution server confirms a final comparison result lambda according to beta: if β=1, the image segmentation execution server will t 2 Then, as a result λ, if β=0, the image segmentation execution server sets t 2 As a result λ;
for the above-mentioned comparison scrambling code circuit with additional input, two integers x with l-bit length are set (l) and y(l) The scrambling code circuit for comparison is shown in FIG. 2, see FIG. 2, where the client holds the input x (l) (x 1 -x l ) The image segmentation execution server holds the input y (l) (y 1 -y l ) To hide the input of CMP in fig. 2, the image segmentation execution server introduces an additional variable c in the circuit, satisfying:
at this time, the circuit is constructed as shown in FIG. 3, < + >>Representing the output of the scrambling code comparison circuit with additional inputs.
S304, setting background cumulative value asThe foreground integrated value is +.>The background count value is +.>The foreground technical value is->Image segmentation execution server initializing background cumulative value +.>Foreground cumulative value +.>Background count value->Foreground count value->And performing pixel calculation, wherein the formula is as follows:
calculate the next thresholdThe formula is:
s305, the image segmentation execution server performs comparison of an absolute value scrambling code circuit together with a privacy service provider to obtainThe image segmentation execution server will->And delta+r 'is given to the privacy service provider, r' is a random value for modification, the privacy service provider judges whether the threshold calculation reaches a threshold standard epsilon, and if yes, the threshold iterative calculation is finished; otherwise, returning to step S304, the image segmentation execution server compares with 0 when executing the absolute value scrambling circuit together with the privacy service provider.
In the present embodiment, the image segmentation execution server owns the encrypted imageEncryption image->Encryption threshold T of (2) i After that, the image segmentation execution server and the privacy service provider encrypt the image using the comparison scrambling circuit +.>Is binarized to obtain an encrypted segmented image +.>
In order to further verify the effectiveness of the method provided by the invention, the encryption image segmentation experimental effect shown in fig. 4 is taken as an illustration, see fig. 4, wherein column a is an original image, column b is a plaintext image segmentation result, columns c, d and e are encryption domain image segmentation results with the packing sizes of 4×4, 8×8 and 16×16 respectively by utilizing the data packing technology based on the method provided by the invention, and as can be seen from fig. 4, when the method provided by the invention is used for image segmentation, the results of b, c and d column encryption domain image segmentation are basically the same as the plaintext image segmentation result, the encryption domain segmentation ensures privacy safety, and the occupation degree and the calculation complexity of space resources can be reduced by utilizing the data packing technology, so that the comprehensive optimization of the current image segmentation method is achieved.
The positional relationship depicted in the drawings is for illustrative purposes only and is not to be construed as limiting the present patent; it is to be understood that the above examples of the present invention are provided by way of illustration only and are not intended to limit the scope of the invention. Other variations or modifications of the above teachings will be apparent to those of ordinary skill in the art. It is not necessary here nor is it exhaustive of all embodiments. Any modification, equivalent replacement, improvement, etc. which come within the spirit and principles of the invention are desired to be protected by the following claims.
Claims (7)
1. The encryption domain image segmentation optimization method based on the data packing technology is characterized by comprising the following steps of:
s1, a privacy service provider generates a public key pk and a private key sk, both the public key pk and the private key sk are sent to a client, and the private key sk is sent to an image segmentation execution server;
s2, the client encrypts and packages the image by using the public key pk, compresses the size of the encrypted image, and sends the encrypted and packaged image to the image segmentation execution server;
the data packing in step S2, the process of compressing the encrypted image size is:
the client encrypts the image by using the public key pk to form an encrypted imageSize s m ×s n Encryption image +.>Split into L-block pictures of the same size, denoted as: i 0 ，I 1 ，…，I L-1 The formula is performed:
where i=0, 1, …,pair I using public key pk P Packaging into a size +.>Compressed image of (a)Representing compressed image +.>The size parameters t, Q of (a) each represent a packing participation parameter,/->I k (i, j) represents the pixel value of the image indexed (i, j) in the kth block of images;
after step S2, before step S3, the method further comprises: image segmentation execution server compresses imageGaussian filtering is performed, and the result after Gaussian filtering is expressed as +.>The method meets the following conditions:
reserving multiplication operation results, expressed as:
wherein m=0, 1, …, s m ；n＝0，1，…，s n All represent image size parameters, G is Gaussian filter template, Q is Gaussian filter template quantization parameter, s g The size of the template is Gaussian filter;
s3, the image segmentation execution server and the privacy service provider interact through multiparty security calculation and a messy code circuit technology to obtain an encrypted segmentation image;
the process of interaction between the image segmentation execution server and the privacy service provider through multiparty security calculation and scrambling circuit technology in step S3 comprises the following steps: for Gaussian filtered encrypted packed dataThe steps of comparison are as follows:
s31, sobel filtering template G in horizontal direction x For example, the image segmentation execution server builds two temporary templates: the positive template coefficient part is expressed asAnd the part with negative template coefficient is denoted +.>The construction formula is as follows:
wherein , and />Respectively in the horizontal direction Sobel convolution kernelsThe portions where the coefficients are positive and negative,
s32, generating two random numbers alpha by the image segmentation execution server + 、α - And executing the formula to pack the two random numbers:
wherein ,representing a random number alpha + Performing data packaging to obtain a value; />Representing a random number alpha - The value after data packing is carried out, and based on the random number after packing, the image segmentation execution server generates intermediate ciphertext +_>Vertical->The following respectively satisfy:
the image segmentation execution server uses the intermediate ciphertextIs->To the privacy service provider:
s33, the privacy service provider uses the private key sk to intermediate ciphertextIs->Decryption is carried out, segmentation is carried out, and ∈ ->Andand using a random number alpha 1 Generating a set { delta }, any one value delta in the set { delta }, and i the method meets the following conditions:
where i=0, 1, …, L-1, c is an additional variable, c e {0,1} is randomly chosen for the privacy service provider;
s34, according to the random number alpha + 、α - Image segmentation execution server calculates intermediate parameter sigma 0 Sigma (sigma) 1 The method comprises the following steps: sigma (sigma) 0 ＝α + -α - ，σ 1 ＝α - -α + Then by a random number alpha 2 Modification sigma i The formula is:
wherein ,representing modified sigma i I=0, 1; after performing a confusing transmission between the image division execution server and the privacy service provider, the image division execution server holds +.>Privacy service provider holds delta i ；
S35, creating a messy code circuit by the image segmentation execution server, generating an encrypted messy code circuit table GCT corresponding to the comparison circuit C with additional input, and decoding the messy code circuit table GCT by the other party to hold valuesDelta i Calculating the output +.>Encryption packaging data based on comparison scrambling code circuit>The image segmentation execution server obtains a comparison result r epsilon {0,1};
s4, the image segmentation execution server sends the encrypted segmentation image to the client, and the client decrypts the encrypted segmentation image to obtain a final edge image.
2. The method for optimizing image segmentation in encrypted domain based on data packing technology according to claim 1, wherein the image segmentation execution server holds during the process of performing the confusing transmission between the image segmentation execution server and the privacy service provider and />The privacy service provider holds c E {0,1}, the image segmentation execution server cannot obtain the value of c, and the privacy service provider cannot obtain +.>Privacy service provider generation->And will->Transmitting to an image segmentation execution server, wherein the image segmentation execution server subtracts alpha 2 Acquisition->
3. The method for optimizing image segmentation in encrypted domain based on data packing technology according to claim 2, wherein the process of interaction between the image segmentation execution server and the privacy service provider through multiparty security computation and scrambling circuit technology in step S3 further comprises: for Gaussian filtered encrypted packed dataPerforming Sobel filtering edge detection to obtain an encrypted image +.>The method comprises the following steps:
privacy service provider encrypts set delta to obtain encrypted setAnd send to the graphAn image division execution server;
for encrypted collectionsEach of +.>The image segmentation execution server with the comparison result r executes the formula to obtain +.>The formula is:
wherein ,α2 The random number generated by the execution server for the image segmentation in step S3 is based onThe image segmentation execution server may acquire horizontal component +.>And vertical component->Then according to the formula
Constructing Sobel filtering result as encrypted image
4. A data-based packaging technique as claimed in claim 3The encryption domain image segmentation optimization method is characterized in that the process of interaction between the image segmentation execution server and the privacy service provider through multiparty security calculation and messy code circuit technology in the step S3 further comprises the following steps: acquiring an encrypted imageThe encryption threshold of (2) comprises the following steps:
s301, setting an encrypted imageThe current threshold is +.>The image division execution service first sets an encrypted image +.>Maximum pixel value +.>And minimum pixel value +.>Then generating an initial threshold value using a division protocol>The formula is:
wherein SecDiv represents a division protocol;
s302, the image segmentation execution server performs segmentation according to the current threshold valueEncryption of images +.>Is divided into a foreground part and a background part, wherein the foreground part refers to a pixel with a value greater than T i The background portion means that the value of the pixel is less than or equal to T i ；
S303, utilizing a scrambling code circuit to perform current threshold valueAnd encryption pixel value +.>Comparing to obtain a comparison result { lambda };
s304, setting background cumulative value asThe foreground integrated value is +.>The background count value is +.>The foreground technical value is->Image segmentation execution server initializing background cumulative value +.>Foreground cumulative value +.>Background count valueForeground count value->And performing pixel calculation, wherein the formula is as follows:
calculate the next thresholdThe formula is:
s305, the image segmentation execution server performs comparison of an absolute value scrambling code circuit together with a privacy service provider to obtainThe image segmentation execution server will->And delta+r ', wherein r' is a random value for modification, and the privacy service provider judges whether the threshold calculation reaches a threshold standard epsilon or not, if so, the threshold iterative calculation is finished; otherwise, the process returns to step S304.
5. The method for optimizing encryption domain image segmentation based on data packing technique according to claim 4, wherein step S303 is performed by usingThe messy code circuit is opposite to the current threshold valueAnd encryption pixel value +.>The comparison process is as follows:
A. image segmentation execution server generates random number gamma 1 And beta epsilon {0,1}, based on random number gamma 1 Sum beta e {0,1}, calculateAnd will->To the privacy service provider, the formula is:
B. the privacy service provider willObtaining t 1 Simultaneously generating c epsilon {0,1}, holding t 1 And c privacy service provider and holding random number gamma 1 The image segmentation execution server of (1) executes a comparison scrambling code circuit with additional input, and the privacy service provider obtains modified result +.>
C. According toAnd c, the privacy service provider calculates the encryption pair t 2 The formula is:
the privacy service provider will t 2 Transmitting to an image segmentation execution server;
D. the image segmentation execution server confirms a final comparison result lambda according to beta: if β=1, the image segmentation execution server will t 2 Then, as a result λ, if β=0, the image segmentation execution server sets t 2 As a result lambda.
6. The method according to claim 5, wherein in step S305, the image segmentation execution server compares with 0 when executing the absolute value scrambling code circuit together with the privacy service provider.
7. The method for optimizing image segmentation in encrypted domain based on data packing technique according to claim 6, wherein the image segmentation execution server has the encrypted imageEncryption image->Encryption threshold T of (2) i After that, the image segmentation execution server and the privacy service provider encrypt the image using the comparison scrambling circuit +.>Is binarized to obtain an encrypted segmented image +.>
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011197779.2A CN112288757B (en) | 2020-10-30 | 2020-10-30 | Encryption domain image segmentation optimization method based on data packing technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011197779.2A CN112288757B (en) | 2020-10-30 | 2020-10-30 | Encryption domain image segmentation optimization method based on data packing technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112288757A CN112288757A (en) | 2021-01-29 |
| CN112288757B true CN112288757B (en) | 2023-09-19 |
Family
ID=74353418
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011197779.2A Active CN112288757B (en) | 2020-10-30 | 2020-10-30 | Encryption domain image segmentation optimization method based on data packing technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112288757B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113159316B (en) * | 2021-04-08 | 2022-05-17 | 支付宝(杭州)信息技术有限公司 | Model training method, method and device for predicting business |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101409831A (en) * | 2008-07-10 | 2009-04-15 | 浙江师范大学 | Method for processing multimedia video object |
| CN103347170A (en) * | 2013-06-27 | 2013-10-09 | 郑永春 | Image processing method used for intelligent monitoring and high-resolution camera applied in image processing method |
| CN105959770A (en) * | 2016-07-21 | 2016-09-21 | 李勇 | Communication method using data line |
| CN110008717A (en) * | 2019-02-26 | 2019-07-12 | 东北大学 | Support the decision tree classification service system and method for secret protection |
| CN110298363A (en) * | 2019-06-27 | 2019-10-01 | 西安电子科技大学 | A kind of secret protection LBP feature extracting method on encrypted image |
| CN111526148A (en) * | 2020-04-26 | 2020-08-11 | 中山大学 | System and method for safely denoising encrypted audio in cloud computing environment |
| CN111522973A (en) * | 2020-04-16 | 2020-08-11 | 重庆邮电大学 | Privacy protection image retrieval method fusing compressed sensing |
-
2020
- 2020-10-30 CN CN202011197779.2A patent/CN112288757B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101409831A (en) * | 2008-07-10 | 2009-04-15 | 浙江师范大学 | Method for processing multimedia video object |
| CN103347170A (en) * | 2013-06-27 | 2013-10-09 | 郑永春 | Image processing method used for intelligent monitoring and high-resolution camera applied in image processing method |
| CN105959770A (en) * | 2016-07-21 | 2016-09-21 | 李勇 | Communication method using data line |
| CN110008717A (en) * | 2019-02-26 | 2019-07-12 | 东北大学 | Support the decision tree classification service system and method for secret protection |
| CN110298363A (en) * | 2019-06-27 | 2019-10-01 | 西安电子科技大学 | A kind of secret protection LBP feature extracting method on encrypted image |
| CN111522973A (en) * | 2020-04-16 | 2020-08-11 | 重庆邮电大学 | Privacy protection image retrieval method fusing compressed sensing |
| CN111526148A (en) * | 2020-04-26 | 2020-08-11 | 中山大学 | System and method for safely denoising encrypted audio in cloud computing environment |
Non-Patent Citations (2)
| Title |
|---|
| 江南 ; 张帆 ; 刘文予 ; .适合于网络传输的二次置乱图像加密算法.中国图象图形学报.2009,(第05期),第1-3页. * |
| 适合于网络传输的二次置乱图像加密算法;江南;张帆;刘文予;;中国图象图形学报(第05期);第1-3页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112288757A (en) | 2021-01-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP2791855B1 (en) | Preserving image privacy when manipulated by cloud services | |
| Bloisi et al. | Image based steganography and cryptography. | |
| Mukhedkar et al. | Secure non real time image encryption algorithm development using cryptography & steganography | |
| CN108810555B (en) | Thumbnail image privacy protection method based on image compression and elastic resolution | |
| Patel et al. | Lazy wavelet transform based steganography in video | |
| Soleymani et al. | A survey on principal aspects of secure image transmission | |
| Soualmi et al. | A novel blind medical image watermarking scheme based on Schur triangulation and chaotic sequence | |
| CN106941402B (en) | Homomorphic ciphertext domain reversible hiding method | |
| CN112288757B (en) | Encryption domain image segmentation optimization method based on data packing technology | |
| Kuo et al. | A steganographic scheme based on formula fully exploiting modification directions | |
| Shah et al. | Paillier cryptosystem based mean value computation for encrypted domain image processing operations | |
| CN112035847A (en) | Image encryption and decryption method and device, electronic equipment and storage medium | |
| Kuo et al. | Hybrid GEMD Data Hiding. | |
| Kataria et al. | Exploring Advanced Encryption and Steganography Techniques for Image Security | |
| Abhinav et al. | An improved reversible data hiding on encrypted images by selective pixel flipping technique | |
| CN113206926B (en) | Method for balancing privacy and usability of image based on three-pixel-point encryption | |
| Singh et al. | DIP using image encryption and XOR operation affine transform | |
| Khan et al. | Advancements in reversible data hiding in encrypted images using public key cryptography | |
| Shankar et al. | Image encryption for mobile devices | |
| Mahasree et al. | An improved reversible data hiding for encrypted images over cloud | |
| Zhou et al. | Image encryption based on edge information | |
| Ali et al. | Implementation of a cryptograpgy algorithm for image transmission | |
| CN107590369B (en) | Homomorphic encrypted domain reversible information hiding method based on code division multiplexing and value expansion | |
| Upendra Raju et al. | A Steganography Embedding Method Based on CDF-DWT Technique for Reversible Data Hiding Application Using Elgamal Algorithm | |
| Umamageswari et al. | A new cryptographic digital signature for secure medical image communication in telemedicine |
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 |