CN108596820A - a kind of image processing system based on information security - Google Patents

Abstract

The invention belongs to the technical field of image processing, and discloses an image processing system based on information security, including: an image acquisition module, an input module, a main control module, an encryption module, an imaging module, a storage module, a judgment module, a decoding module, and a watermark module . The present invention provides a watermark module that performs non-overlapping blocks according to the size of 2×2 pixels, extracts the average gray value from the image to be processed, and defines the average gray value as watermark information. Since the average gray value is represented by 8 binary bits, the high 6 bits represent the integer part of the average gray value, embedding the watermark information extracted from the image to be processed can represent the average gray value of the image block without error, as long as there is one bit of data change in either the image information part or the embedded watermark information part , the algorithm will detect that the image has been tampered with. At the same time, through the encryption module, the image data is combined with the password, which can improve the security of image viewing.

Description

An Image Processing System Based on Information Security

technical field

The invention belongs to the technical field of image processing, and in particular relates to an image processing system based on information security.

Background technique

Information security mainly includes the following five aspects, that is, the need to ensure the confidentiality, authenticity, integrity, unauthorized copying of information and the security of the parasitic system. Information security itself covers a wide range, including how to prevent the leakage of commercial enterprise secrets, prevent teenagers from browsing bad information, and the leakage of personal information. The information security system in the network environment is the key to ensuring information security, including computer security operating systems, various security protocols, security mechanisms (digital signatures, message authentication, data encryption, etc.), and security systems, such as UniNAC, DLP, etc., as long as A security breach can threaten global security. Information security means that the information system (including hardware, software, data, people, physical environment and its infrastructure) is protected from being damaged, changed, or leaked due to accidental or malicious reasons, and the system operates continuously and reliably. Information services are not interrupted, ultimately achieving business continuity. However, if the existing image is tampered with, it is not easy to find out, which leads to the risk of image dissemination; at the same time, if the image is transferred out of the storage device, anyone can watch the image, and the user's privacy is not fully protected.

To sum up, the problems in the existing technology are: if the existing images are tampered with, it is not easy to find out, resulting in the risk of image dissemination; at the same time, if the images are transferred out of the storage device, anyone can watch the images, and the user's privacy is protected. Not fully protected.

Contents of the invention

Aiming at the problems existing in the prior art, the present invention provides an image processing system based on information security.

The present invention is achieved in that a kind of image processing system based on information security comprises:

Image acquisition module, input module, main control module, encryption module, imaging module, storage module, judgment module, decoding module, watermark module;

An image acquisition module, connected with the main control module, for collecting original pictures;

Said image acquisition module extracts the HoG feature and the GMM feature of the information security image training sample, and combines the HoG feature and the GMM feature as the feature of the information security image; comprises the following steps:

Divide the information security image training sample image into several information security images, convert each pixel in the information security image from the RGB color space to the YC _b C _r color space, and extract the values of its C _b and C _r chrominance components; Wherein, Y represents the luminance component, C _b represents the blue chrominance component, and C _r represents the red chrominance component;

Divide the information security image into U small image blocks that do not overlap with each other, and calculate the mean value vector of each small image block on the C _b , C _r chrominance components: Substitute these mean vectors into the GMM information security image color model trained in step (2), and obtain each Gaussian component ω _i G _i (i∈{1,2,…,K} ), as the color feature of the small image block, the color features of U small image blocks are combined and recorded as

Convert the information security image training sample image into a grayscale image, and perform Gamma correction on the input image;

Calculate the horizontal and vertical gradient values g _x (x, y) and g _y (x, y) of each pixel (x, y) in the grayscale image;

g _x (x,y)=I(x+1,y)-I(x-1,y)

g _y (x,y)=I(x,y+1)-I(x,y-1)

In the formula, I(x, y) represents the gray value at the pixel point (x, y), and the gradient amplitude g(x, y) and the direction α(x, y) at the pixel point (x, y) are calculated according to the following formula ,y);

Divide the grayscale image of the information security image training sample into information security images, and count the gradient histogram on each information security image, which is used to describe the shape information of the target;

For each information security image, the histogram of 9 bins is used to count the gradient information of the pixels in the information security image, and the gradient size of the pixels in each bin is accumulated to form the gradient histogram of the information security image, and the 9-dimensional feature is used Vector, denoted as h ₁ =[f ₁ ,f ₂ ,…,f ₉ ], where f _i is the cumulative gradient value of the ith bin;

Combine the obtained gradient feature vector h ₁ and the obtained color feature vector h ₂ to form the feature vector [h ₁ , h ₂ ] of the information security image;

Combining infosec images into blocks and normalizing within blocks;

The input module is connected with the main control module and is used for inputting user operation picture information;

The main control module is connected with the image acquisition module, input module, encryption module, storage module, judgment module and watermark module, and is used to control the normal operation of each module;

The information security image stitching line of the main control module adopts a method of eliminating cross-overlapping area occlusion, specifically including:

1) Using the method of threshold segmentation to divide the information security image into two parts, the high value area and the low value area, the threshold cost criterion is defined as formula (2) (3):

Among them, cost (the mismatch measurement matrix obtained by formula (1), δ T] is the image segmentation threshold, where TI is the maximum mismatch value in the overlapping area, which is a fixed constant, δ∈(0; for The binarization matrix;

2) Judging whether the starting point and the ending point of the suture line are both in the low-value area (Tcost_b=0) and located in the same connected component, that is, judging whether there is a path between the starting point and the ending point; if not, it means that there are one or more spanning The occlusion of the overlapping area; find the occlusion across the overlapping area, and gradually reduce the value of the occlusion area until there is no occlusion across the overlapping area, so that the starting point and the ending point are located in the same connected component; at this time, the connected area where the starting point and ending point are located , that is, the least connected domain of the suture line to be solved;

An encryption module is connected with the main control module and is used to encrypt pictures;

The imaging module is connected with the encryption module, and is used to generate an encrypted picture after encrypting the picture;

The storage module is connected with the main control module and is used to store the encrypted pictures;

The judging module is connected with the main control module, and is used for matching and judging the picture operation information input by the user and the encrypted information of the encryption module. If they are consistent, the picture is opened through the decoding module, otherwise it cannot be opened;

The decoding module is connected with the judging module, and is used for judging the result according to the judging module, and if it is consistent, the picture is decrypted;

The watermark module is connected with the main control module and is used for embedding watermarks on pictures.

Further, the embedding method of the watermark module is as follows:

First, the image to be processed is divided into non-overlapping blocks according to the size of 2×2 pixels to obtain several sub-blocks, and the gray-scale mean value of all pixels in the sub-block is calculated, and the gray-scale mean value is defined as watermark information; according to the first key, Encrypting the watermark information using the first pseudo-random binary sequence generated by logistic mapping to obtain the watermark information to be embedded;

Secondly, according to the second key, the second pseudo-random binary sequence generated by logistic mapping is used to encrypt the embedded bit-plane of the watermark information on the image to be processed to obtain an encrypted embedded bit-plane;

Next, embedding the watermark information to be embedded into the image to be processed according to the encrypted embedding bit plane to obtain an embedded watermark;

Then, according to the third key, a chaotic iterative operation is performed to generate a chaotic real-valued sequence, and the chaotic real-valued sequence is divided into a plurality of non-overlapping partitions of the same size, and the chaotic real-valued sequence of each partition is sorted to obtain an ordered sequence, obtaining a permutation embedding address according to the position number of each value in the chaotic real value sequence of each partition in the ordered sequence; permuting the embedding watermark according to the permutation embedding address to obtain a permutation watermark;

Finally, in the unit of partition, one partition containing the permuted watermark is swapped with another partition; the transposed permuted watermark is embedded into the image to be processed.

The advantages and positive effects of the present invention are as follows: the present invention provides the watermark module to perform non-overlapping blocks according to the size of 2×2 pixels, extract the average gray value from the image to be processed, and define the average gray value as watermark information, because the average gray value Represented by 8 binary bits, the upper 6 bits represent the integer part of the gray mean value, and the last 2 bits represent the fractional part of the gray value. Even if only 1 bit of information has been tampered with, it can be detected and the tampered area can be accurately located. Embedding the watermark information extracted from the image to be processed can represent the average gray value of the image block without error. As long as there is a bit of data change in either the image information part or the embedded watermark information part, the algorithm will detect that the image has been tamper. At the same time, through the encryption module, the image data is combined with the password, which can improve the security of image viewing.

Description of drawings

FIG. 1 is a structural block diagram of an image processing system based on information security provided by an embodiment of the present invention.

In the figure: 1. Image acquisition module; 2. Input module; 3. Main control module; 4. Encryption module; 5. Imaging module; 6. Storage module; 7. Judgment module; 8. Decoding module; 9. Watermark module.

Detailed ways

In order to further understand the content, features and effects of the present invention, the following examples are given, and detailed descriptions are given below with reference to the accompanying drawings.

The structure of the present invention will be described in detail below in conjunction with the accompanying drawings.

As shown in Figure 1, the image processing system based on information security provided by the present invention includes: image acquisition module 1, input module 2, main control module 3, encryption module 4, imaging module 5, storage module 6, judgment module 7, decoding Module 8, watermark module 9.

The image acquisition module 1 is connected with the main control module 3 for collecting original pictures;

The input module 2 is connected with the main control module 3, and is used to input user operation picture information;

The main control module 3 is connected with the image acquisition module 1, the input module 2, the encryption module 4, the storage module 6, the judgment module 7, and the watermark module 9, and is used to control the normal operation of each module;

Encryption module 4, is connected with main control module 3, is used for encrypting picture;

The imaging module 5 is connected with the encryption module 4, and is used to generate an encrypted picture after encrypting the picture;

Storage module 6, is connected with main control module 3, is used for storing encrypted picture;

Judgment module 7 is connected with main control module 3, and is used for matching and judging the picture operation information input by the user and the encrypted information of encryption module 4. If consistent, the picture is opened by decoding module 8, otherwise it cannot be opened;

The decoding module 8 is connected with the judging module 7, and is used to judge the result according to the judging module 7, if consistent, then decrypt the picture;

The watermark module 9 is connected with the main control module 3, and is used for embedding a watermark on the picture.

Said image acquisition module extracts the HoG feature and the GMM feature of the information security image training sample, and combines the HoG feature and the GMM feature as the feature of the information security image; comprises the following steps:

Divide the information security image training sample image into several information security images, convert each pixel in the information security image from the RGB color space to the YC _b C _r color space, and extract the values of its C _b and C _r chrominance components; Wherein, Y represents the luminance component, C _b represents the blue chrominance component, and C _r represents the red chrominance component;

Divide the information security image into U small image blocks that do not overlap with each other, and calculate the mean value vector of each small image block on the C _b , C _r chrominance components: Substitute these mean vectors into the GMM information security image color model trained in step (2), and obtain each Gaussian component ω _i G _i (i∈{1,2,…,K} ), as the color feature of the small image block, the color features of U small image blocks are combined and recorded as

Convert the information security image training sample image into a grayscale image, and perform Gamma correction on the input image;

Calculate the horizontal and vertical gradient values g _x (x, y) and g _y (x, y) of each pixel (x, y) in the grayscale image;

g _x (x,y)=I(x+1,y)-I(x-1,y)

g _y (x,y)=I(x,y+1)-I(x,y-1)

In the formula, I(x, y) represents the gray value at the pixel point (x, y), and the gradient amplitude g(x, y) and the direction α(x, y) at the pixel point (x, y) are calculated according to the following formula ,y);

Divide the grayscale image of the information security image training sample into information security images, and count the gradient histogram on each information security image, which is used to describe the shape information of the target;

For each information security image, the histogram of 9 bins is used to count the gradient information of the pixels in the information security image, and the gradient size of the pixels in each bin is accumulated to form the gradient histogram of the information security image, and the 9-dimensional feature is used Vector, denoted as h ₁ =[f ₁ ,f ₂ ,…,f ₉ ], where f _i is the cumulative gradient value of the ith bin;

Combine the obtained gradient feature vector h ₁ and the obtained color feature vector h ₂ to form the feature vector [h ₁ , h ₂ ] of the information security image;

Combining infosec images into blocks and normalizing within blocks;

The information security image stitching line of the main control module adopts a method of eliminating cross-overlapping area occlusion, specifically including:

1) Using the method of threshold segmentation to divide the information security image into two parts, the high value area and the low value area, the threshold cost criterion is defined as formula (2) (3):

where is cost (the mismatch measurement matrix obtained by formula (1), δ T] is the image segmentation threshold, where T] is the maximum mismatch value in the overlapping area, which is a fixed constant, δ∈(0; for The binarization matrix;

2) Judging whether the starting point and the ending point of the suture line are both in the low-value area (Tcost_b=0) and located in the same connected component, that is, judging whether there is a path between the starting point and the ending point; if not, it means that there are one or more spanning The occlusion of the overlapping area; find the occlusion across the overlapping area, and gradually reduce the value of the occlusion area until there is no occlusion across the overlapping area, so that the starting point and the ending point are located in the same connected component; at this time, the connected area where the starting point and ending point are located , that is, the least connected domain of the suture line to be solved.

The embedding method of watermark module 9 provided by the present invention is as follows:

First, the image to be processed is divided into non-overlapping blocks according to the size of 2×2 pixels to obtain several sub-blocks, and the gray-scale mean value of all pixels in the sub-block is calculated, and the gray-scale mean value is defined as watermark information; according to the first key, Encrypting the watermark information using the first pseudo-random binary sequence generated by logistic mapping to obtain the watermark information to be embedded;

Secondly, according to the second key, the second pseudo-random binary sequence generated by logistic mapping is used to encrypt the embedded bit-plane of the watermark information on the image to be processed to obtain an encrypted embedded bit-plane;

Next, embedding the watermark information to be embedded into the image to be processed according to the encrypted embedding bit plane to obtain an embedded watermark;

Then, according to the third key, a chaotic iterative operation is performed to generate a chaotic real-valued sequence, and the chaotic real-valued sequence is divided into a plurality of non-overlapping partitions of the same size, and the chaotic real-valued sequence of each partition is sorted to obtain an ordered sequence, obtaining a permutation embedding address according to the position number of each value in the chaotic real value sequence of each partition in the ordered sequence; permuting the embedding watermark according to the permutation embedding address to obtain a permutation watermark;

Finally, in the unit of partition, one partition containing the permuted watermark is swapped with another partition; the transposed permuted watermark is embedded into the image to be processed.

During the processing of the present invention, the original picture is collected by the image acquisition module 1; the main control module 3 transfers the encryption module 4 to encrypt the picture; then, the encrypted picture is generated after the image is encrypted by the imaging module 5; the encrypted picture is encrypted by the storage module 6 Storage; the user inputs the user operation picture information through the input module 2; the main control module 3 transfers the judgment module 7 to match and judge the picture operation information input by the user and the encrypted information of the encryption module 4, if they are consistent, the picture is opened by the decoding module 8, otherwise Unable to open; Embedding watermark to the picture through watermark module 9.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention in any form. Any simple modifications made to the above embodiments according to the technical essence of the present invention, equivalent changes and modifications, all belong to this invention. within the scope of the technical solution of the invention.

Claims (2)

1. An image processing system based on information security, characterized in that, the image processing system based on information security comprises: Image acquisition module, input module, main control module, encryption module, imaging module, storage module, judgment module, decoding module, watermark module; An image acquisition module, connected with the main control module, for collecting original pictures; Said image acquisition module extracts the HoG feature and the GMM feature of the information security image training sample, and combines the HoG feature and the GMM feature as the feature of the information security image; comprises the following steps: Divide the information security image training sample image into several information security images, convert each pixel in the information security image from the RGB color space to the YC _b C _r color space, and extract the values of its C _b and C _r chrominance components; Wherein, Y represents the luminance component, C _b represents the blue chrominance component, and C _r represents the red chrominance component; Divide the information security image into U small image blocks that do not overlap with each other, and calculate the mean value vector of each small image block on the C _b , C _r chrominance components: Substitute these mean vectors into the GMM information security image color model trained in step (2), and obtain each Gaussian component ω _i G _i (i∈{1,2,…,K} ), as the color feature of the small image block, the color features of U small image blocks are combined and recorded as Convert the information security image training sample image into a grayscale image, and perform Gamma correction on the input image; Calculate the horizontal and vertical gradient values g _x (x, y) and g _y (x, y) of each pixel (x, y) in the grayscale image; g _x (x,y)=I(x+1,y)-I(x-1,y) g _y (x,y)=I(x,y+1)-I(x,y-1) In the formula, I(x, y) represents the gray value at the pixel point (x, y), and the gradient amplitude g(x, y) and the direction α(x, y) at the pixel point (x, y) are calculated according to the following formula ,y); Divide the grayscale image of the information security image training sample into information security images, and count the gradient histogram on each information security image, which is used to describe the shape information of the target; For each information security image, the histogram of 9 bins is used to count the gradient information of the pixels in the information security image, and the gradient size of the pixels in each bin is accumulated to form the gradient histogram of the information security image, and the 9-dimensional feature is used Vector, denoted as h ₁ =[f ₁ ,f ₂ ,…,f ₉ ], where f _i is the cumulative gradient value of the ith bin; Combine the obtained gradient feature vector h ₁ and the obtained color feature vector h ₂ to form the feature vector [h ₁ , h ₂ ] of the information security image; Combining infosec images into blocks and normalizing within blocks; The input module is connected with the main control module and is used for inputting user operation picture information; The main control module is connected with the image acquisition module, input module, encryption module, storage module, judgment module and watermark module, and is used to control the normal operation of each module; The information security image stitching line of the main control module adopts a method of eliminating cross-overlapping area occlusion, specifically including: 1) Using the method of threshold segmentation to divide the information security image into two parts, the high value area and the low value area, the threshold cost criterion is defined as formula (2) (3): Among them, cost (the mismatch measurement matrix obtained by formula (1), δ T] is the image segmentation threshold, where T1 is the maximum mismatch value in the overlapping area, which is a fixed constant, δ∈(0; for The binarization matrix; 2) Judging whether the starting point and the ending point of the suture line are both in the low-value area (Tcost_b=0) and located in the same connected component, that is, judging whether there is a path between the starting point and the ending point; if not, it means that there are one or more spanning The occlusion of the overlapping area; find the occlusion across the overlapping area, and gradually reduce the value of the occlusion area until there is no occlusion across the overlapping area, so that the starting point and the ending point are located in the same connected component; at this time, the connected area where the starting point and ending point are located , that is, the least connected domain of the suture line to be solved; An encryption module is connected with the main control module and is used to encrypt pictures; The imaging module is connected with the encryption module, and is used to generate an encrypted picture after encrypting the picture; The storage module is connected with the main control module and is used to store the encrypted pictures; The judging module is connected with the main control module, and is used for matching and judging the picture operation information input by the user and the encrypted information of the encryption module. If they are consistent, the picture is opened through the decoding module, otherwise it cannot be opened; The decoding module is connected with the judging module, and is used for judging the result according to the judging module, and if it is consistent, the picture is decrypted; The watermark module is connected with the main control module and is used for embedding watermarks on pictures. 2. the image processing system based on information security as claimed in claim 1, is characterized in that, described watermark module embedding method is as follows: First, the image to be processed is divided into non-overlapping blocks according to the size of 2×2 pixels to obtain several sub-blocks, and the gray-scale mean value of all pixels in the sub-block is calculated, and the gray-scale mean value is defined as watermark information; according to the first key, Encrypting the watermark information using the first pseudo-random binary sequence generated by logistic mapping to obtain the watermark information to be embedded; Secondly, according to the second key, the second pseudo-random binary sequence generated by logistic mapping is used to encrypt the embedded bit-plane of the watermark information on the image to be processed to obtain an encrypted embedded bit-plane; Next, embedding the watermark information to be embedded into the image to be processed according to the encrypted embedding bit plane to obtain an embedded watermark; Then, according to the third key, a chaotic iterative operation is performed to generate a chaotic real-valued sequence, and the chaotic real-valued sequence is divided into a plurality of non-overlapping partitions of the same size, and the chaotic real-valued sequence of each partition is sorted to obtain an ordered sequence, obtaining a permutation embedding address according to the position number of each value in the chaotic real value sequence of each partition in the ordered sequence; permuting the embedding watermark according to the permutation embedding address to obtain a permutation watermark; Finally, in the unit of partition, one partition containing the permuted watermark is swapped with another partition; the transposed permuted watermark is embedded into the image to be processed.