CN106228503B - The digital watermark embedding and extracting method of satellite remote-sensing image SHP dot files - Google Patents

Abstract

The invention discloses a kind of digital watermark embeddings and extracting method of satellite remote-sensing image SHP dot files, belong to field of information security technology, this method can be while the satellite remote-sensing image to SHP forms carries out product quality supervision, ensures product information safety, retrospect related responsibility, storage redundancy is reduced, and then more watermark information is stored in a small amount of space.

Description

Digital Watermark Embedding and Extraction Method of Satellite Remote Sensing Image SHP Point File

technical field

The invention relates to the technical field of information security, in particular to a method for embedding and extracting digital watermarks of SHP point files of satellite remote sensing images.

Background technique

Satellite remote sensing imagery is an important basic and strategic information resource of the country, an important basis for scientific planning, environmental protection, resource development, engineering construction, military operations, etc., and a basic tool and important tool for informatization construction and sustainable development. Assure. Ensuring the data information security of satellite remote sensing images is an important issue faced by engineers and technicians.

Information security encryption technology is the most commonly used method to protect digital products. The development of this technology is relatively mature and it has been widely used in various fields of the information society. Encryption technology is used to encrypt the file, so that the content of the file becomes ciphertext, so that even if the file is illegally intercepted, its content cannot be obtained, so as to achieve the purpose of protection. However, encryption technology itself has some insurmountable shortcomings. The encrypted data is more likely to attract the attention of attackers. With the rapid development of computers, the ability to decipher passwords is getting stronger and stronger, and the security of conventional passwords is greatly threatened. . What's more serious is that once the password is deciphered by the attacker, the digital product will lose its protection, and it is easy to have piracy, illegal access, malicious tampering and other behaviors.

Digital watermarking technology makes up for the deficiency of traditional encryption technology and develops gradually. A complete digital watermarking scheme generally includes three parts: watermark generation, watermark embedding, watermark extraction or detection. Specifically, digital watermarking technology actually optimizes the key links such as carrier medium analysis, watermark preprocessing, watermark embedding position selection, watermark embedding method design, watermark extraction method design, etc., on the premise of first meeting the basic needs. Seek to solve quasi-optimal design problems under the main constraints of imperceptibility, safety reliability, and robustness. The basic process of digital watermark embedding is shown in Figure 1, the input content includes the original watermark information , raw carrier data and an optional key , the output result is the data containing the watermark . Watermark Generation Algorithm The uniqueness, validity, irreversibility and other attributes of the watermark should be guaranteed. key Can be used to enhance security to avoid unauthorized watermark recovery or watermark extraction. The general process of watermark embedding can be defined by the following formula:

in, Indicates the data after embedding the watermark (that is, the data containing the watermark); Embedding algorithm for watermark; represents raw carrier data; Indicates the original watermark information; Represents a collection of keys. here key It is optional and is generally used for watermark signal extraction.

Figure 2 shows the general process of digital watermark extraction. The process may or may not require the participation of the original carrier image or original watermark. The watermark extraction process in different situations can be described as follows.

Raw carrier data required Time:

original watermark required Time:

When there is no original information:

in, represents the extracted watermark; Algorithm for watermark extraction; Represents data that contains a watermark.

When using the above-mentioned digital watermarking technology to encrypt SHP point files of satellite remote sensing images, the entire image needs to be regarded as watermark information, and the processing of watermark images is mainly limited to generating binary sequences of 0 and 1, which is a waste of time. The imaginary part of the complex array generated during the Fourier transform also increases the additional overhead of the watermark information and increases the space occupation of the watermark information.

Contents of the invention

The purpose of the present invention is to provide a digital watermark embedding and extraction method for SHP point files of satellite remote sensing images. Store redundant information, and then store more watermark information in a small amount of space.

In order to achieve the above object, the technical solutions adopted by the present invention are as follows.

The digital watermark embedding process of satellite remote sensing image SHP point file includes the following steps:

Step 1. Scrambling the bmp file as a watermark image. The scrambling process is to multiply the horizontal and vertical coordinates of each pixel of the original image by a matrix to obtain new coordinates, and then divide the new coordinates by the length of the image Take the remainder of the sum and width to get the new coordinates of each pixel.

in are the horizontal and vertical coordinates of the original image, Respectively, the horizontal and vertical coordinates after multiplying the matrix, are the newly obtained horizontal and vertical coordinates after scrambling, is the length and width of the bmp image, in pixels;

Step 2: Transform the converted binary bmp image into a vector image by finding all points with pixel values of 0 (that is, black) in the image, and recording the coordinates of these points in turn ;

Step 3: Align the coordinates to shrink the coordinates Multiply by a fixed coefficient (divided by the number of pixels in the length and width of the watermark image), so that all values are less than 1;

Among them, 32 is the pixel value of the length and width of the watermark image set in this paper, M is the number of black pixels in the bmp image after scrambling, is the coordinate sequence of all black points after scaling;

Step 4: Take the abscissa of the original point file data sequence as the real part of the complex number, and the ordinate as the imaginary part of the complex number, generate an array of complex numbers, perform Fourier transform, and obtain the coefficients after Fourier transform

in, is the original point file data sequence, is the data sequence after Fourier transform, N is the total number of point entities in the point file, Respectively, the horizontal and vertical coordinates of the point entity, is the coefficient after Fourier transform;

Step 5: The coordinate sequence of the black point obtained by preprocessing remember to do , do Fourier transform to get the transformed sequence;

in, As the real part of the sequence of complex numbers, As the imaginary part of a sequence of complex numbers, is the sequence obtained after Fourier transform;

Step 6: Sum the dot file data sequence obtained after Fourier transform and the coordinate sequence of black dots to obtain the embedded watermark sequence, and perform discrete Fourier inverse transform on the sequence to obtain the dot file coordinates after embedding the watermark sequence;

in, For the post-summation sequence, is the embedded watermark sequence obtained after inverse discrete Fourier transform, That is, the horizontal and vertical coordinates of the point after the watermark is embedded, and the watermark embedding is completed.

The digital watermark extraction process of the satellite remote sensing image SHP point file includes the following steps:

Step 1: To the original image , embedded watermarked image , respectively for fast Fourier transform,

in, represents the original vector image, is the original vector image coordinates, represents the transformed image, is the transformed image coordinates, Indicates the image after embedding the watermark, is the image coordinates embedded with the watermark, Indicates the converted watermarked image, is the transformed image coordinates embedded with watermark, N and M are the number of point entities in the original image and the watermarked image respectively;

Step 2: Calculate the difference between the two Fourier-transformed sequences in the previous step, use the complex number array obtained after the Fourier transform of the watermarked image, and subtract it from the complex number array after the Fourier transform of the original image, and obtain the difference It is the array after the Fourier transform of the watermark information, take the smaller number of N and M as the number of subtraction, and then perform the inverse fast Fourier transform to obtain the vector information of the watermark,

,

in, is the array obtained after subtraction, is its coordinate value, is the number of subtractions, is the vector information of the watermark, its coordinate value;

Step 3: put the obtained Multiply by the coefficient of the watermark image preprocessing (the number of pixels of the length and width of the watermark image, that is, 32), the decimal places of the point coordinates are rounded, the points according to the coordinates stored in the vector information are black, and the rest of the coordinate points are white The rules to construct a bmp binary image;

Step 4: Scrambling the bmp binary image constructed in step 3, the obtained image is the watermark image, and the watermark extraction is completed.

In the present invention, the watermark picture is regarded as a vector diagram in a logical sense, and the plural arrays respectively store the x and y coordinates of the points whose pixels are black, and the amount of stored information remains unchanged, but the occupied space is greatly reduced, and can be stored in a small amount of space. Store more watermark information in .

Since the watermark information that undergoes Fourier transform is stored in a coordinate sequence, the coordinate values of x and y may be relatively large, which will have some impact on the accuracy of the original vector image. In the present invention, the coordinate sequence is divided by the length and width pixels of the watermark picture, so that the obtained coordinate sequence is a sequence less than 1, so that the use value of the original image will not be greatly affected.

Description of drawings

Figure 1 Basic process of digital watermark embedding.

Figure 2 The general process of digital watermark extraction.

Figure 3 SHP file watermark embedding specific process.

Figure 4 SHP file watermark extraction specific process.

Detailed ways

The technical scheme of the present invention is described in detail below in conjunction with accompanying drawing:

The specific embedding process of a digital watermark in a satellite remote sensing image SHP point file is shown in Figure 3, and the text description is as follows:

Step 1: Before embedding the watermark, first perform Arnold scrambling (also known as cat face transformation) on the bmp file as the watermark image. The scrambling process is to multiply the horizontal and vertical coordinates of each pixel of the original image by a matrix to obtain The new coordinates, and then divide the new coordinates by the length and width of the image to get the remainder to get the new coordinates of each pixel,

in are the horizontal and vertical coordinates of the original image, Respectively, the horizontal and vertical coordinates after multiplying the matrix, are the newly obtained horizontal and vertical coordinates after scrambling, Be the length and the width of the bmp image, the unit is pixel, and the present invention all takes 31; the number of times of scrambling can be used as a key, and will be used in the process of watermark extraction in the back;

Step 2: Transform the transformed binary bmp image into a vector image by finding all points with pixel values of 0 (that is, black) in the image, and recording the coordinates of these points in turn ;

Step 3: In order to reduce the impact of the embedded watermark on the original image, the coordinates Multiplied by a fixed coefficient, the present invention is the reciprocal of the number of pixels of the length and width of the watermark image, and the present invention takes , let all values be less than 1 to reduce the image information of the image;

Among them, 32 is the pixel value of the length and width of the watermark image set in this paper, M is the number of black pixels in the bmp image after scrambling, is the coordinate sequence of all black points;

Step 4: Take the abscissa of the original point file data sequence as the real part of the complex number, and the ordinate as the imaginary part of the complex number, perform Fourier transform to obtain the coefficient after Fourier transform

in, is the original point file data sequence, is the data sequence after Fourier transform, N is the total number of point entities in the point file, Respectively, the horizontal and vertical coordinates of the point entity, is the coefficient after Fourier transform;

Step 5: The coordinate sequence of the black point obtained by preprocessing remember to do , do the Fourier transform.

in, As the real part of the sequence of complex numbers, As the imaginary part of a sequence of complex numbers, is the sequence obtained after Fourier transform.

Step 6: Sum the dot file data sequence obtained after Fourier transform and the coordinate sequence of the black point using the additive criterion to obtain the embedded watermark sequence, and perform discrete Fourier inverse transform on the sequence to obtain the embedded watermark The coordinate sequence of the point file;

in, For the post-summation sequence, is the embedded watermark sequence obtained after inverse discrete Fourier transform, That is, the horizontal and vertical coordinates of the point after the watermark is embedded, and the watermark embedding ends.

The digital watermark extraction process of a satellite remote sensing image SHP point file is shown in Figure 4, and the text description is as follows:

Step 1: To the original image and the watermarked image Do fast Fourier transform respectively;

in, represents the original image, is the original image coordinates, represents the transformed image, is the transformed image coordinates, Indicates the image after embedding the watermark, is the image coordinates embedded with the watermark, Indicates the converted watermarked image, is the transformed image coordinates embedded with watermark, N and M are the number of point entities in the original image and the watermarked image respectively;

Step 2: Subtract the complex number array obtained after Fourier transform of the watermarked image from the complex number array after Fourier transform of the original image, and the obtained difference is the array after Fourier transform of the watermark information, taking N, The smaller number in M is used as the subtraction number. Then perform inverse fast Fourier transform to obtain the vector information of the watermark

in, is the array obtained after subtraction, is its coordinate value, is the number of subtractions, is the vector information of the watermark, as its coordinate value.

Step 3: Multiply the vector information obtained in step 2 by the coefficient of watermark image preprocessing, and round the decimal places of the point coordinates. According to the rule that the points of the coordinates stored in the vector information are black, and the rest of the coordinate points are white, Construct a bmp binary image;

Step 4: Scrambling the bmp binary image constructed in step 3, the obtained image is the watermark image, and the watermark extraction is completed. The scrambling process (cat face transformation) used in the present invention is a periodic transformation, that is to say, after a certain number of transformations, it can finally be transformed back to the original image. The watermark image used in the present invention is 32*32 pixels, and the transformation period is 24, the bmp binary image constructed in step 3 is passed through times of scrambling, the obtained image is the watermark image, and the watermark extraction is completed.

Claims (2)

1. A digital watermark embedding method for a satellite remote sensing image SHP point file is characterized by comprising the following steps:

step 1, scrambling the bmp file serving as the watermark picture, wherein the scrambling process is to multiply the horizontal and vertical coordinates of each pixel point of the original image by a matrix to obtain new coordinates, divide the new coordinates by the length and width of the image to obtain the remainder to obtain the new coordinates of each pixel point,

the description shows that D \ electronic application \ cases \ inventions \ dec74a24-241f-4bd8-82bc-9da336c61753\ others \ f1de61a5-9112-4D8f-9D45-3bf2fd5c7D1b \100001\405819dest _ path _ image002.jpg

WhereinRespectively are the horizontal and vertical coordinates of the original image,respectively the horizontal and vertical coordinates after being multiplied by the matrix,respectively are a horizontal coordinate and a vertical coordinate which are newly obtained after scrambling processing,is the length and width of the bmp image, in pixels;

step 2: converting the converted binary bmp image into a vector image by finding out points with all pixel values of 0, namely black in the image and sequentially recording the coordinates of the points；

And step 3: to coordinatePerforming reduction processing to obtain coordinatesMultiplying by a fixed coefficient, namely dividing by the number of pixels of the length and the width of the watermark image, and enabling all values to be smaller than 1;

wherein 32 is water as defined hereinPrinting the pixel values of the length and the width of the image, wherein M is the number of black pixel points of the bmp image after scrambling,the coordinate sequences of all the black points after the zooming processing are obtained;

and 4, step 4: the abscissa of the original point file data sequence is used as the real part of the complex number, the ordinate is used as the imaginary part of the complex number, a complex number array is generated, Fourier transformation is carried out, and the coefficient after Fourier transformation is obtained

Wherein,in the form of an original point file data sequence,is a data sequence after Fourier transform, N is the total number of point entities in the point file,respectively are the horizontal and vertical coordinates of the point entity,the coefficients are Fourier transformed;

and 5: coordinate sequence of black points obtained by preprocessingRecord asPerforming Fourier transform to obtain a transformed sequence;

wherein,as the real part of the complex sequence,as the imaginary part of the sequence of complex numbers,is a sequence obtained after Fourier transform;

step 6: summing the point file data sequence obtained after Fourier transformation and the coordinate sequence of the black point to obtain a watermark embedding sequence, and performing inverse discrete Fourier transformation on the sequence to obtain a point file coordinate sequence embedded with the watermark;

wherein,in order to sum up the sequences of the sequence,for the embedded watermark sequence obtained after the inverse discrete fourier transform,namely the horizontal and vertical coordinates of the point after embedding the watermark, and finishing the watermark embedding.

2. The method for extracting the digital watermark from the SHP point file of the satellite remote sensing image as claimed in claim 1, comprising the following steps:

step 1: for the original imageEmbedding watermarked imagesThe fast Fourier transform is respectively carried out,

wherein,which represents the original vector-based image,is the coordinates of the original vector image,representing the image after the transformation, and,in order to transform the coordinates of the image after transformation,representing the image after the embedding of the watermark,for the image coordinates in which the watermark has been embedded,representing the transformed watermarked image,the transformed image coordinate embedded with the watermark, N and M are the number of point entities in the original image and the watermark-containing image respectively;

step 2: calculating the difference value of the two sequences after Fourier transform in the last step, subtracting the complex number array obtained after Fourier transform of the watermark-containing image from the complex number array obtained after Fourier transform of the original image to obtain the difference, namely the array obtained after Fourier transform of the watermark information, taking the smaller number in N, M as the subtracted number, then performing inverse fast Fourier transform to obtain the vector information of the watermark,

，

wherein,is the array resulting from the subtraction,as a result of the coordinate values thereof,in order to be the number of the subtraction,is the vector information of the watermark and is,is the coordinate value thereof;

and step 3: obtained byMultiplying by a coefficient during the preprocessing of the watermark picture, namely the number of pixels 32 of the length and the width of the watermark picture, rounding the decimal place of the point coordinate, and constructing a bmp binary image according to the rule that the points of the coordinate stored in the vector information are black and the rest coordinate points are white;

and 4, step 4: and (4) scrambling the bmp binary image constructed in the step (3) to obtain an image which is a watermark image, and finishing watermark extraction.