Image fusion effect evaluation method and system based on binary DCT coefficient

Abstract

A method and system for evaluating image fusion performance based on binarized DCT coefficients includes acquiring a source image and a fused image, and dividing them into blocks according to the matrix size of a quantization table; performing a two-dimensional discrete cosine transform on each block, arranging the transformed coefficients according to the order of the blocks in the original image to obtain transformed coefficient matrices of the source image and the fused image, then quantizing them using a quantization table, and binarizing the elements in the matrices of the quantized DCT coefficients to obtain binarized quantized DCT coefficient matrices of the source image and the fused image; constructing an ideal binary matrix using the binarized quantized DCT coefficient matrix of the source image; calculating the normalized Hamming similarity, normalized reward factor, and penalty factor between the ideal binary matrix and the binarized quantized DCT coefficient matrix of the fused image, and combining different coefficients to obtain the image fusion performance evaluation result. The evaluation results of this invention have high consistency with human visual perception.

Description

Technical Field

[0001] This invention belongs to the field of image processing technology, specifically relating to a method and system for evaluating image fusion effects based on binarized DCT coefficients. Background Technology

[0002] Image fusion technology combines image information from different sources into a single image, integrating complementary information, preserving details, and resulting in a fused image that better conforms to human visual perception characteristics, facilitating image processing. Common applications include multispectral image fusion and image fusion with different focal points. In recent years, various image fusion methods and technologies have emerged, and how to accurately and quickly evaluate the fusion effect of these methods has become an urgent problem to be solved. Summary of the Invention

[0003] The purpose of this invention is to address the problems in the prior art by providing a method and system for evaluating image fusion effects based on binarized DCT coefficients. The evaluation result of the image fusion effect is obtained by computer calculation, the entire calculation process does not require human intervention, and the evaluation result of the obtained image fusion effect is highly consistent with human visual perception.

[0004] To achieve the above objectives, the present invention provides the following technical solution:

[0005] Firstly, a method for evaluating image fusion performance based on binarized DCT coefficients is provided, including:

[0006] Select a quantification table;

[0007] Get source image , ,…, and the merged image and the source image , ,…, and the merged image The image is divided into several blocks according to the matrix size of the quantization table; a two-dimensional discrete cosine transform is performed on each block, and the transformed coefficients are arranged according to the order of the blocks in the original image to obtain the transformed coefficient matrix of the source image. , ,…, and the coefficient matrix after image transformation ;

[0008] The coefficient matrix after transforming the source image , ,…, and the coefficient matrix after image transformation Quantization is performed using the quantization table to obtain the matrix of quantized DCT coefficients of the source image. , ,…, and the matrix after quantizing DCT coefficients of the fused image ;

[0009] The matrix after quantizing the DCT coefficients of the source image , ,…, and the matrix after quantizing DCT coefficients of the fused image Binarizing the matrix elements in the image yields the binarized and quantized DCT coefficient matrices of the source image. , ,…, The matrix after binarizing and quantizing the DCT coefficients of the fused image ;

[0010] The matrix after binarizing and quantizing the DCT coefficients of the source image , ,…, Constructing an ideal binary matrix Each element in the ideal binary matrix is ​​equal to all , Union operation of elements at corresponding positions ;

[0011] Calculate the ideal binary matrix The matrix after binarizing and quantizing the DCT coefficients of the fused image The normalized Hamming similarity, normalized reward factor, and penalty factor between the images are calculated; and the calculated normalized Hamming similarity, normalized reward factor, and penalty factor are combined with different coefficients to obtain the image fusion effect evaluation result value.

[0012] As a preferred embodiment, the normalized Hamming similarity is calculated as follows:

[0013] The ideal binary matrix The matrix after binarizing and quantizing the DCT coefficients of the fused image Perform an XOR operation on corresponding positions between them, and divide the number of non-zero values ​​in the result by the total number of elements in the binary matrix. The normalized Hamming distance between the two matrices is obtained. Then subtract the normalized Hamming distance from 1 to obtain the normalized Hamming similarity between the two matrices.

[0014] .

[0015] As a preferred embodiment, the normalized reward factor is calculated as follows:

[0016] Comparison of ideal binary matrices The matrix after binarizing and quantizing the DCT coefficients of the fused image The low-frequency components between them, if in the same position of the element, The value is 0 and If the value is 1, it is considered that low-frequency gain has been obtained through fusion; sum up the number of such elements and divide by the total number of low-frequency components. Obtain the normalized reward factor :

[0017] .

[0018] As a preferred embodiment, the penalty factor comprises two parts: the frequency components lost during the fusion process and the added high-frequency noise;

[0019] The calculation process for the first part of the penalty factor is as follows: compare the ideal binary matrix. The matrix after binarizing and quantizing the DCT coefficients of the fused image All elements between, if elements at the same position, The value is 1 and If the value is 0, it is considered a frequency component lost during fusion. The number of such elements is summed and divided by the total number of elements in the binary matrix to obtain the first part of the penalty factor. :

[0020] ;

[0021] The calculation process for the penalty factor in the second part is as follows: Based on specific needs, the high-frequency band is divided, and an ideal binary matrix is ​​compared. The matrix after binarizing and quantizing the DCT coefficients of the fused image The high-frequency components between them, if in the same position of the element, The value is 0 and If the value is 1, it is considered as high-frequency noise added during fusion. The number of such elements is summed up and divided by the total number of high-frequency components. Obtain the second part of the penalty factor. :

[0022] .

[0023] As a preferred approach, the calculated normalized Hamming similarity, normalized reward factor, and penalty factor are combined using different coefficients according to the following formula to obtain the image fusion effect evaluation result:

[0024]

[0025] In the formula, ,and ;

[0026] The higher the image fusion effect evaluation result value, the better the image fusion effect, and vice versa.

[0027] As a preferred embodiment, the fused image It is the source image , ,…, Based on the image fusion method, each fused image is obtained. They are all the same size.

[0028] As a preferred option, the quantization table is selected from existing quantization tables used for JPEG compression, or designed independently based on human visual perception.

[0029] Secondly, an image fusion effect evaluation system based on binary DCT coefficients is provided, including:

[0030] The quantization table selection module is used to select the quantization table;

[0031] The image segmentation and transformation module is used to acquire the source image. , ,…, and the merged image and the source image , ,…, and the merged image The image is divided into several blocks according to the matrix size of the quantization table; a two-dimensional discrete cosine transform is performed on each block, and the transformed coefficients are arranged according to the order of the blocks in the original image to obtain the transformed coefficient matrix of the source image. , ,…, and the coefficient matrix after image transformation ;

[0032] The quantization module is used to transform the coefficient matrix of the source image. , ,…, and the coefficient matrix after image transformation Quantization is performed using the quantization table to obtain the matrix of quantized DCT coefficients of the source image. , ,…, and the matrix after quantizing DCT coefficients of the fused image ;

[0033] The binarization module is used to quantize the matrix of DCT coefficients from the source image. , ,…, and the matrix after quantizing DCT coefficients of the fused image Binarizing the matrix elements in the image yields the binarized and quantized DCT coefficient matrices of the source image. , ,…, The matrix after binarizing and quantizing the DCT coefficients of the fused image ;

[0034] The ideal binary matrix construction module is used to construct a matrix from the source image after binarizing and quantizing the DCT coefficients. , ,…, Constructing an ideal binary matrix Each element in the ideal binary matrix is ​​equal to all , Union operation of elements at corresponding positions ;

[0035] The image fusion effect evaluation result calculation module is used to calculate the ideal binary matrix. The matrix after binarizing and quantizing the DCT coefficients of the fused image The normalized Hamming similarity, normalized reward factor, and penalty factor between the images are calculated; and the calculated normalized Hamming similarity, normalized reward factor, and penalty factor are combined with different coefficients to obtain the image fusion effect evaluation result value.

[0036] Thirdly, an electronic device is provided, comprising:

[0037] Memory, storing at least one instruction; and

[0038] The processor executes the instructions stored in the memory to implement the image fusion effect evaluation method based on binarized DCT coefficients.

[0039] Fourthly, a computer-readable storage medium is provided, the computer-readable storage medium storing a computer program, which, when executed by a processor, implements the image fusion effect evaluation method based on binarized DCT coefficients.

[0040] Compared with the prior art, the present invention has at least the following beneficial effects:

[0041] This invention's image fusion effect evaluation method based on binarized DCT coefficients does not require constructing an ideal fused image. Instead, it only requires constructing binarized quantized coefficient matrices for different DCT frequency bands, which serve as the ideal binarized matrix for comparison and calculation with the actual fused image's binarized quantized DCT coefficient matrix in subsequent processing. The normalized Hamming similarity, normalized reward factor, and penalty factor between the ideal binarized matrix and the matrix obtained from the binarized quantized DCT coefficients of the fused image are combined using different coefficients to obtain the image fusion effect evaluation result. Both the reward factor and penalty factor calculations are based on human visual characteristics. The reward fusion method increases the gain of low-frequency components that are sensitive to human vision, while the penalty fusion process loses frequency component information and adds high-frequency noise. The coefficients can be selected according to specific circumstances, as long as they meet the range of values ​​for all coefficients and the algebraic sum is 1. The entire calculation process of this invention's evaluation method requires no manual intervention, and the evaluation results are highly consistent with human visual perception. Attached Figure Description

[0042] Figure 1 Flowchart of the image fusion effect evaluation method based on binarized DCT coefficients of this invention;

[0043] Figure 2 Focusing on fused images obtained from different source images and fusion methods:

[0044] (a) is the right-side focused image; (b) is the left-side focused image; (c) is the fused image after mean fusion (AVE); (d) is the fused image after discrete wavelet transform (DWT); (e) is the fused image after gray scale pyramid (GRP); (f) is the fused image after Lappass pyramid (LP). Detailed Implementation

[0045] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, those skilled in the art can obtain other embodiments without creative effort.

[0046] See Figure 1 An image fusion effect evaluation method based on binarized DCT coefficients according to an embodiment of the present invention includes:

[0047] Step 1) Select a suitable quantification table. Quantification table It can be an existing form or one designed based on human visual characteristics. When quantification tables... Once selected, it cannot be changed throughout the entire processing; for example, the following quantization tables with different values:

[0048] Table 1

[0049]

[0050] Table 2

[0051]

[0052] Table 3

[0053]

[0054] Step 2) Perform Discrete Cosine Transform (DCT) on all images.

[0055] Get source image , ,…, and the merged image These images are divided into small blocks, each the same size as the quantization table matrix selected in step 1). Any row or column margins are padded with zeros. A two-dimensional discrete cosine transform (DCT) is then performed on each of these blocks, and the transformed coefficients are arranged according to the order of the blocks in the original image to obtain the transformed coefficient matrix of the source image. , ,…, and the coefficient matrix after image transformation ;

[0056] Step 3), quantification.

[0057] Pass all the coefficient matrices obtained in step 2) through the quantization table in step 1). Quantization yields the matrix of DCT coefficients from the source image. , ,…, and the matrix after quantizing DCT coefficients of the fused image ;

[0058] Step 4) Binarization.

[0059] Binarize the matrix elements after quantizing the DCT coefficients in step 3), setting all non-zero values ​​to 1 and all zero coefficients to 0, to obtain the matrix of the source image after binarization and quantization of the DCT coefficients. , ,…, The matrix after binarizing and quantizing the DCT coefficients of the fused image ;

[0060] Step 5) Construct the ideal binary matrix.

[0061] The matrix obtained in step 4) after binarizing and quantizing the DCT coefficients of the source image , ,…, Constructing an ideal binary matrix In an ideal binary matrix, each element is equal to all... , Union operation on elements at corresponding positions: That is, only one The element at this position has a value of 1. The value of the element at this position is 1, only all The element at that position has a value of 0. The element at this position has a value of 0;

[0062] Step 6) Calculate the normalized Hamming similarity between the ideal binary matrix and the binarized and quantized DCT coefficient matrix of the fused image. (The ideal binary matrix...) Binarization and quantization of the DCT coefficient matrix of the fused image Perform an XOR operation at the corresponding positions, and divide the number of non-zero values ​​in the result by the total number of elements in the binary matrix. The normalized Hamming distance between the two matrices is obtained. Subtracting this distance from 1 gives the normalized Hamming similarity between the two matrices. ;

[0063] Step 7) Calculate the reward factor, which is the low-frequency gain obtained during the fusion process. Divide the low-frequency band according to specific needs and compare it with the ideal binary matrix. Binarization and quantization of the DCT coefficient matrix of the fused image The low-frequency components, in the elements at the same position, The value is 0 and If the value is 1, it is considered that low-frequency gain has been obtained through fusion. Add up the number of such elements and divide by the total number of low-frequency components. Obtain the normalized reward factor , ;

[0064] Step 8) Calculate the penalty factor.

[0065] The penalty factor consists of two parts: the frequency components lost during the fusion process and the added high-frequency noise.

[0066] The calculation process for the penalty factor in the first part is as follows: compare the ideal binary matrix. Binarization and quantization of the DCT coefficient matrix of the fused image All elements, elements at the same position, The value is 1 and If the value is 0, it is considered a frequency component lost during fusion. The sum of the number of such elements, divided by the total number of elements in the binary matrix, yields the first penalty factor. , .

[0067] The second part of the penalty factor calculation process is as follows: Based on specific needs, the high-frequency band is divided, and an ideal binary matrix is ​​compared. Binarization and quantization of the DCT coefficient matrix of the fused image The high-frequency components, in elements at the same position, The value is 0 and If the value is 1, it is considered as high-frequency noise added during fusion. The sum of the number of such elements is then divided by the total number of high-frequency components. Obtain the second part of the penalty factor. , ;

[0068] Step 9) Calculate the evaluation results of the image fusion effect.

[0069] The normalized Hamming similarity between the ideal binary matrix obtained in step 6) and the binarized and quantized DCT coefficient matrix of the fused image is calculated. The normalized reward factor obtained in step 7) The penalty factor obtained in step 8) and By combining different coefficients, the image fusion effect evaluation result is obtained. ,in ,and .

[0070] According to the above-described embodiment of the present invention, an image fusion effect evaluation method based on binarized DCT coefficients is provided, and the fused image... It is the source image , ,…, These images, obtained using a certain image fusion method, are all of the same size.

[0071] Furthermore, in step 5) of the image fusion effect evaluation method based on binarized DCT coefficients of the present invention, it is not necessary to construct an ideal fused image, but only to construct a binarized quantization coefficient matrix of different DCT frequency bands, which serves as the ideal binarized matrix for comparison and calculation with the binarized quantization DCT coefficient matrix of the actual fused image in subsequent processing.

[0072] Furthermore, in the image fusion effect evaluation method based on binarized DCT coefficients of the present invention, the calculation of the reward factor in step 7) and the calculation of the penalty factor in step 8) are both based on human visual characteristics. The reward fusion method improves the gain of low-frequency components that are sensitive to human vision, while the penalty fusion process loses frequency component information and increases high-frequency noise.

[0073] Table 4 lists the... Figure 2 The results of the fusion results obtained by different fusion methods were evaluated. Table 2 is used as the quantification table, and the parameters were selected as follows. The evaluation results indicate that the mean method yields the worst fusion effect, while the Laplace pyramid method yields the best fusion effect, consistent with human visual perception.

[0074] Table 4

[0075]

[0076] Table 5 lists the... Figure 2 The fusion results in each graph are evaluated using different quantization tables, with the parameters selected as follows: As can be seen from the table, as long as the design of the quantification table conforms to human visual characteristics, the evaluation results of the evaluation method are consistent for various fusion methods.

[0077] Table 5

[0078]

[0079] Another embodiment of the present invention proposes an image fusion effect evaluation system based on binarized DCT coefficients, comprising:

[0080] The quantization table selection module is used to select the quantization table;

[0081] The image segmentation and transformation module is used to acquire the source image. , ,…, and the merged image and the source image , ,…, and the merged image The image is divided into several blocks according to the matrix size of the quantization table; a two-dimensional discrete cosine transform is performed on each block, and the transformed coefficients are arranged according to the order of the blocks in the original image to obtain the transformed coefficient matrix of the source image. , ,…, and the coefficient matrix after image transformation ;

[0082] The quantization module is used to transform the coefficient matrix of the source image. , ,…, and the coefficient matrix after image transformation Quantization is performed using the quantization table to obtain the matrix of quantized DCT coefficients of the source image. , ,…, and the matrix after quantizing DCT coefficients of the fused image ;

[0083] The binarization module is used to quantize the matrix of DCT coefficients from the source image. , ,…, and the matrix after quantizing DCT coefficients of the fused image Binarizing the matrix elements in the image yields the binarized and quantized DCT coefficient matrices of the source image. , ,…, The matrix after binarizing and quantizing the DCT coefficients of the fused image ;

[0084] The ideal binary matrix construction module is used to construct a matrix from the source image after binarizing and quantizing the DCT coefficients. , ,…, Constructing an ideal binary matrix Each element in the ideal binary matrix is ​​equal to all , Union operation of elements at corresponding positions ;

[0085] The image fusion effect evaluation result calculation module is used to calculate the ideal binary matrix. The matrix after binarizing and quantizing the DCT coefficients of the fused image The normalized Hamming similarity, normalized reward factor, and penalty factor between the images are calculated; and the calculated normalized Hamming similarity, normalized reward factor, and penalty factor are combined with different coefficients to obtain the image fusion effect evaluation result value.

[0086] Another embodiment of the present invention also provides an electronic device, comprising:

[0087] Memory, storing at least one instruction; and

[0088] The processor executes the instructions stored in the memory to implement the image fusion effect evaluation method based on binarized DCT coefficients.

[0089] Another embodiment of the present invention also proposes a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the image fusion effect evaluation method based on binarized DCT coefficients.

[0090] For example, the instructions stored in the memory can be divided into one or more modules / units. These modules / units are stored in a computer-readable storage medium and executed by the processor to complete the image fusion effect evaluation method based on binarized DCT coefficients according to the present invention. The one or more modules / units can be a series of computer-readable instruction segments capable of performing specific functions, which describe the execution process of the computer program on the server.

[0091] The electronic device may be a smartphone, laptop, PDA, or cloud server, among other computing devices. It may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the electronic device may also include more or fewer components, or combinations of certain components, or different components; for example, it may also include input / output devices, network access devices, buses, etc.

[0092] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0093] The memory can be an internal storage unit of the server, such as a hard drive or RAM. Alternatively, it can be an external storage device, such as a plug-in hard drive, Smart Media Card (SMC), Secure Digital (SD) card, or FlashCard. Furthermore, the memory may include both internal and external storage units. The memory is used to store computer-readable instructions and other programs and data required by the server. It can also be used to temporarily store data that has been output or will be output.

[0094] It should be noted that the information interaction and execution process between the above-mentioned module units are based on the same concept as the method embodiment. For details on their specific functions and technical effects, please refer to the method embodiment section. They will not be repeated here.

[0095] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0096] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying the computer program code to a photographing device / terminal device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0097] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0098] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for evaluating image fusion performance based on binarized DCT coefficients, characterized in that, include: Select a quantification table; Get source image , ,…, and the merged image and the source image , ,…, and the merged image The image is divided into several blocks according to the matrix size of the quantization table; a two-dimensional discrete cosine transform is performed on each block, and the transformed coefficients are arranged according to the order of the blocks in the original image to obtain the transformed coefficient matrix of the source image. , ,…, and the coefficient matrix after image transformation ; The coefficient matrix after transforming the source image , ,…, and the coefficient matrix after image transformation Quantization is performed using the quantization table to obtain the matrix of quantized DCT coefficients of the source image. , ,…, and the matrix after quantizing DCT coefficients of the fused image ; The matrix after quantizing the DCT coefficients of the source image , ,…, and the matrix after quantizing DCT coefficients of the fused image Binarizing the matrix elements in the image yields the binarized and quantized DCT coefficient matrices of the source image. , ,…, The matrix after binarizing and quantizing the DCT coefficients of the fused image ; The matrix after binarizing and quantizing the DCT coefficients of the source image , ,…, Constructing an ideal binary matrix Each element in the ideal binary matrix is ​​equal to all , Union operation of elements at corresponding positions ; Calculate the ideal binary matrix The matrix after binarizing and quantizing the DCT coefficients of the fused image The normalized Hamming similarity, normalized reward factor, and penalty factor between the images are calculated; and the calculated normalized Hamming similarity, normalized reward factor, and penalty factor are combined with different coefficients to obtain the image fusion effect evaluation result value.

2. The image fusion effect evaluation method based on binarized DCT coefficients according to claim 1, characterized in that, The normalized Hamming similarity is calculated as follows: The ideal binary matrix The matrix after binarizing and quantizing the DCT coefficients of the fused image Perform an XOR operation on corresponding positions between them, and divide the number of non-zero values ​​in the result by the total number of elements in the binary matrix. The normalized Hamming distance between the two matrices is obtained. Then subtract the normalized Hamming distance from 1 to obtain the normalized Hamming similarity between the two matrices. 。 3. The image fusion effect evaluation method based on binarized DCT coefficients according to claim 2, characterized in that, The normalized reward factor is calculated as follows: Comparison of ideal binary matrices The matrix after binarizing and quantizing the DCT coefficients of the fused image The low-frequency components between them, if in the same position of the element, The value is 0 and If the value is 1, it is considered that low-frequency gain has been obtained through fusion; sum up the number of such elements and divide by the total number of low-frequency components. Obtain the normalized reward factor : 。 4. The image fusion effect evaluation method based on binarized DCT coefficients according to claim 3, characterized in that, The penalty factor consists of two parts: the frequency components lost during the fusion process and the added high-frequency noise; The calculation process for the first part of the penalty factor is as follows: compare the ideal binary matrix. The matrix after binarizing and quantizing the DCT coefficients of the fused image All elements between, if elements at the same position, The value is 1 and If the value is 0, it is considered a frequency component lost during fusion. The number of such elements is summed and divided by the total number of elements in the binary matrix to obtain the first part of the penalty factor. : ； The calculation process for the penalty factor in the second part is as follows: Based on specific needs, the high-frequency band is divided, and an ideal binary matrix is ​​compared. The matrix after binarizing and quantizing the DCT coefficients of the fused image The high-frequency components between them, if in the same position of the element, The value is 0 and If the value is 1, it is considered as high-frequency noise added during fusion. The number of such elements is summed up and divided by the total number of high-frequency components. Obtain the second part of the penalty factor. : 。 5. The image fusion effect evaluation method based on binarized DCT coefficients according to claim 4, characterized in that, The calculated normalized Hamming similarity, normalized reward factor, and penalty factor are combined using different coefficients according to the following formula to obtain the image fusion effect evaluation result: In the formulae, and ; The higher the image fusion effect evaluation result value, the better the image fusion effect, and vice versa.

6. The image fusion effect evaluation method based on binarized DCT coefficients according to claim 1, characterized in that, The fused images are source images , …, obtained according to an image fusion method, and each fused image is of the same size.

7. The image fusion effect evaluation method based on binarized DCT coefficients according to claim 1, characterized in that, The quantization table can be selected from existing quantization tables used for JPEG compression, or designed according to human visual perception.

8. An image fusion effect evaluation system based on binarized DCT coefficients, characterized in that, include: The quantization table selection module is used to select the quantization table; The image segmentation and transformation module is used to acquire the source image. , ,…, and the merged image and the source image , ,…, and the merged image The image is divided into several blocks according to the matrix size of the quantization table; a two-dimensional discrete cosine transform is performed on each block, and the transformed coefficients are arranged according to the order of the blocks in the original image to obtain the transformed coefficient matrix of the source image. , ,…, and the coefficient matrix after image transformation ; The quantization module is used to transform the coefficient matrix of the source image. , ,…, and the coefficient matrix after image transformation Quantization is performed using the quantization table to obtain the matrix of quantized DCT coefficients of the source image. , ,…, and the matrix after quantizing DCT coefficients of the fused image ; The binarization module is used to quantize the matrix of DCT coefficients from the source image. , ,…, and the matrix after quantizing DCT coefficients of the fused image Binarizing the matrix elements in the image yields the binarized and quantized DCT coefficient matrices of the source image. , ,…, The matrix after binarizing and quantizing the DCT coefficients of the fused image ; The ideal binary matrix construction module is used to construct a matrix from the source image after binarizing and quantizing the DCT coefficients. , ,…, Constructing an ideal binary matrix Each element in the ideal binary matrix is ​​equal to all , Union operation of elements at corresponding positions ; The image fusion effect evaluation result calculation module is used to calculate the ideal binary matrix. The matrix after binarizing and quantizing the DCT coefficients of the fused image The normalized Hamming similarity, normalized reward factor, and penalty factor between the images are calculated; and the calculated normalized Hamming similarity, normalized reward factor, and penalty factor are combined with different coefficients to obtain the image fusion effect evaluation result value.

9. An electronic device, characterized in that, include: Memory, storing at least one instruction; and The processor executes instructions stored in the memory to implement the image fusion effect evaluation method based on binarized DCT coefficients as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the image fusion effect evaluation method based on binary DCT coefficients as described in any one of claims 1 to 7.

Images