An image complexity acquisition apparatus, method, device and storage medium

Abstract

The application relates to the technical field of image processing, and discloses an image complexity acquisition device, method, equipment and storage medium, which comprise a gradient graph generation module, a gradient amplitude calculation module, a centroid moment calculation module and a complexity acquisition module; the gradient graph generation module is used for generating a first gradient graph in an X direction and a second gradient graph in a Y direction of a to-be-tested image respectively; the gradient amplitude calculation module is used for calculating image edge gradient amplitudes and generating a gradient amplitude graph by using the first gradient graph and the second gradient graph; the centroid moment calculation module is used for calculating image edge centroid moments according to the gradient amplitude graph; and the complexity acquisition module is used for combining the calculated image edge centroid moments and image edge gradient amplitudes with angle entropy to acquire the complexity of the to-be-tested image. The complexity of the image can be qualitatively and quantitatively described, which is beneficial to better guiding and grasping image processing and analysis work and has extremely important significance for various fields of image engineering, and the corresponding complexity can also be acquired for different spatial distributions of the same pixel point of the image.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of image processing, in particular to an image complexity obtaining device, method, equipment and storage medium. BACKGROUND

[0002] Complexity is an inherent characteristic of an objective existence or event which is well known to people but difficult to quantitatively describe. Complexity degree is a degree description of complexity. Qualitative understanding of complexity of a thing is not only beneficial to a general overview of the thing, but also conducive to further grasping the relevant field. Quantitative analysis of complexity of a thing is helpful to more accurately grasping the inherent difficulty degree of performing corresponding work, so as to guide reasonable design and implementation of the work.

[0003] Image complexity is a description of the inherent complexity degree of an image, which can reflect the inherent difficulty degree of whether to complete or complete some operations (such as edge detection, image enhancement, image compression, target automatic extraction, etc.). As for the complexity of an image, scholars at home and abroad have studied it early, and these studies are also different in different research purposes. At present, the commonly used image complexity obtaining methods include information entropy, edge ratio, correlation degree and energy. However, these methods cannot distinguish the complexity when the images have different spatial distributions of the same pixel points. SUMMARY

[0004] Therefore, the purpose of the present application is to provide an image complexity obtaining device, method, equipment and storage medium, which can qualitatively and quantitatively describe the complexity degree of an image, and is also applicable to images with different spatial distributions of the same pixel points. The specific scheme is as follows:

[0005] An image complexity obtaining device comprises:

[0006] A gradient map generation module is configured to generate a first gradient map of a to-be-tested image in an X direction and a second gradient map of the to-be-tested image in a Y direction respectively.

[0007] A gradient amplitude calculation module is configured to calculate image edge gradient amplitudes and generate a gradient amplitude map by using the first gradient map and the second gradient map.

[0008] A centroid moment calculation module is configured to calculate image edge centroid moments according to the gradient amplitude map.

[0009] A complexity obtaining module is configured to combine the calculated image edge centroid moments and the image edge gradient amplitudes with an angle entropy to obtain complexity of the to-be-tested image.

[0010] Preferably, in the image complexity obtaining apparatus provided by the embodiment of the present application, the gradient amplitude calculation module is specifically configured to calculate the image edge gradient amplitude by using the following formula:

[0011]

[0012] wherein, Gf is the image edge gradient amplitude, G x is the gradient value of the first gradient map, G y is the gradient value of the second gradient map.

[0013] Preferably, in the image complexity obtaining apparatus provided by the embodiment of the present application, the image complexity obtaining apparatus further comprises:

[0014] a gradient direction calculation module, configured to calculate the image edge gradient direction by using the first gradient map and the second gradient map;

[0015] an angle histogram obtaining module, configured to normalize the calculated image edge gradient direction to obtain an angle histogram;

[0016] an angle entropy statistical module, configured to obtain the probability of each angle from the angle histogram, and to statistically obtain the angle entropy according to the probability of each angle.

[0017] Preferably, in the image complexity obtaining apparatus provided by the embodiment of the present application, the center of mass calculation module comprises:

[0018] a point set construction unit, configured to construct image edge point set coordinates in the gradient amplitude map;

[0019] a center of mass calculation unit, configured to calculate image edge center of mass coordinates according to the image edge point set coordinates and the point set number;

[0020] a center of mass moment calculation unit, configured to calculate image edge center of mass moment by using the image edge point set coordinates and the image edge center of mass coordinates.

[0021] Preferably, in the image complexity obtaining apparatus provided by the embodiment of the present application, the complexity obtaining module comprises:

[0022] a joint product calculation unit, configured to calculate the joint product of the image edge center of mass moment and the image edge gradient amplitude;

[0023] a complexity obtaining unit, configured to obtain the complexity of the image to be tested according to the calculated joint product and the angle entropy.

[0024] Preferably, in the image complexity obtaining apparatus provided by the embodiment of the present application, the complexity obtaining unit is specifically configured to obtain the complexity of the image to be tested by using the following formula:

[0025] C = e Entropy *log(F(d, Gf));

[0026]

[0027]

[0028] Where C is the complexity of the image to be tested, Gf(x) i y i ) is the point (x i y i The gradient magnitude from the image edge is given by N, where N represents the number of points in the set, P(k) is the probability when the angle is k, and Entropy is the angle entropy.

[0029] Preferably, in the image complexity acquisition device provided in the embodiments of the present invention, the gradient map generation module is specifically used to perform convolution on the image to be tested using an edge detection algorithm to generate a first gradient map of the image to be tested in the X direction and a second gradient map in the Y direction.

[0030] This invention also provides a method for obtaining image complexity, including:

[0031] Generate the first gradient map of the image under test in the X direction and the second gradient map in the Y direction, respectively;

[0032] Using the first gradient map and the second gradient map, the gradient magnitude of the image edge is calculated and a gradient magnitude map is generated;

[0033] Calculate the centroid moment of the image edge based on the gradient magnitude map;

[0034] The complexity of the image under test is obtained by combining the calculated centroid moment of the image edge and the magnitude of the image edge gradient with the angular entropy.

[0035] This invention also provides an image complexity acquisition device, including a processor and a memory, wherein the processor executes a computer program stored in the memory to implement the image complexity acquisition method described above in this invention.

[0036] This invention also provides a computer-readable storage medium for storing a computer program, wherein the computer program, when executed by a processor, implements the image complexity acquisition method described above in this invention.

[0037] From the above technical scheme can be seen, the image complexity acquisition device provided by the application, comprising: gradient map generation module, for generating the first gradient map of the image to be measured in the X direction and the second gradient map in the Y direction respectively; gradient amplitude calculation module, for calculating the image edge gradient amplitude and generating the gradient amplitude map by using the first gradient map and the second gradient map; Centroid moment calculation module, for calculating the image edge centroid moment according to the gradient amplitude map; Complexity acquisition module, for combining the calculated image edge centroid moment and image edge gradient amplitude with angle entropy to acquire the complexity of the image to be measured.

[0038] The image complexity acquisition device provided by the application considers the conditions of image gradient amplitude, image centroid moment and angle entropy to acquire the image complexity through the interaction of the above four modules, which can qualitatively and quantitatively describe the complexity of the image, is beneficial to better guide and grasp the image processing and analysis work, has extremely important significance for the fields of image engineering, can acquire the corresponding complexity for the different spatial distribution of the image with the same pixel, and the acquired image complexity can be used as an evaluation index to reflect the difficulty of image watermarking technology, image compression and image segmentation, be used for evaluating and measuring the quality of the image, be used as a feature of the image to reflect the visual feature of the image, be used for target recognition, image classification and information steganography, and be used as an image visual perception emotion for the research of image cognition and aesthetic calculation.

[0039] In addition, the application also provides a corresponding method, device and computer readable storage medium for the image complexity acquisition device, which further makes the above device more practical, and the method, device and computer readable storage medium have corresponding advantages. BRIEF DESCRIPTION OF DRAWINGS

[0040] In order to more clearly illustrate the technical solutions in the embodiments of the application or the related art, the drawings needed to be used in the embodiments or the related art description will be briefly introduced. Obviously, the drawings in the following description only the embodiments of the application, and for those skilled in the art, without creative labor, can also obtain other drawings according to the provided drawings.

[0041] Figure 1 The structural schematic diagram of the image complexity acquisition device provided by the embodiment of the application;

[0042] Figure 2 The image to be measured provided by the embodiment of the application;

[0043] Figure 3 The first gradient map provided by the embodiment of the application;

[0044] Figure 4A second gradient map provided for an embodiment of the present application;

[0045] Figure 5 A gradient amplitude map provided for an embodiment of the present application;

[0046] Figure 6 A first to-be-measured image example map provided for an embodiment of the present application;

[0047] Figure 7 A second to-be-measured image example map provided for an embodiment of the present application;

[0048] Figure 8 A third to-be-measured image example map provided for an embodiment of the present application;

[0049] Figure 9 A first center of mass moment image example map provided for an embodiment of the present application;

[0050] Figure 10 A second center of mass moment image example map provided for an embodiment of the present application;

[0051] Figure 11 A third center of mass moment image example map provided for an embodiment of the present application;

[0052] Figure 12 A fourth center of mass moment image example map provided for an embodiment of the present application;

[0053] Figure 13 A flow chart of an image complexity acquisition method provided for an embodiment of the present application. DETAILED DESCRIPTION

[0054] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the protection scope of the present application.

[0055] The present application provides an image complexity acquisition device, as shown in Figure 1 , comprising:

[0056] A gradient map generation module 11 is configured to generate a first gradient map in the X direction and a second gradient map in the Y direction of a to-be-measured image, respectively.

[0057] In implementation, the to-be-measured image is processed to generate gradient maps in the X direction and the Y direction, i.e., the first gradient map and the second gradient map. Figure 2 A to-be-measured image is shown, Figure 3 A first gradient map corresponding to the to-be-measured image of Figure 2 is shown,Figure 4 a second gradient map corresponding to the image to be tested. Figure 2

[0058] a gradient amplitude calculation module 12, configured to calculate image edge gradient amplitudes and generate a gradient amplitude map by using the first gradient map and the second gradient map.

[0059] In implementation, the image edge gradient amplitudes can be calculated based on the first gradient map and the second gradient map, and the gradient amplitude map can be generated according to the calculated image edge gradient amplitudes. Figure 5 a gradient amplitude map corresponding to the image to be tested. Figure 2

[0060] a centroid moment calculation module 13, configured to calculate image edge centroid moments according to the gradient amplitude map.

[0061] a complexity acquisition module 14, configured to acquire complexity of the image to be tested by combining the calculated image edge centroid moments and the image edge gradient amplitudes with angular entropy.

[0062] It should be noted that the image pixel gray scale distribution law can reflect the spatial distribution of each gray scale pixel in the image. If there are more target points in the image (without considering noise), generally, the image structure is more complex.

[0063] The more scattered the same number of points are, the more complex the image is, for example Figure 7 the second image to be tested example shown in FIG. 2 is more complex than Figure 6 the first image to be tested example shown in FIG. 1. Therefore, the fraction degree of the target points of the image can be represented by the centroid distance:

[0064] R=d(x i ,y i ); (1)

[0065] From the visual point of view, the higher the contrast (gradient) of the same number of point sets in the same position, the clearer the image is, and the higher the complexity of the image is, for example Figure 7 the second image to be tested example shown in FIG. 2 is more complex than Figure 8 the third image to be tested example shown in FIG. 3. Therefore, the image complexity can consider the centroid moments and the gradient.

[0066] In addition, Figures 9 to 12 a first centroid moment image example, a second centroid moment image example, a third centroid moment image example and a fourth centroid moment image example are shown. Figures 9 to 12 the centroid moment and the edge gradient gray scale product are increasing, but actually the structural complexity of the image is not increasing, Figures 9 to 11 ​​The spatial distribution area tends to be simple. Therefore, the present application uses the angle information entropy of the gradient direction to evaluate the confusion degree of the image edge, and the joint product of the centroid moment and the gradient amplitude to evaluate the dispersion degree of the image edge point.

[0067] From the information entropy, the size of information is related to the probability of random events. The smaller the probability, the greater the amount of information generated by the event, and the greater the probability, the smaller the amount of information generated by the event. Information entropy can also be used as a measure of the complexity of a system. If a system is more complex, the more different situations appear, and the information entropy is larger. If a system is simpler, the number of situations is small, and if the probability is 1, it is considered to occur inevitably, and the information entropy is 0. Therefore, the more dispersed the direction of the image gradient, the more dispersed the corresponding angle, and the greater the information entropy, and the more complex the image structure.

[0068] Based on this, the present application jointly constructs an image complexity model by the joint product of the angle information entropy, the centroid moment and the gradient amplitude.

[0069] In the above image complexity acquisition device provided by the embodiment of the present application, the image gradient amplitude, the image centroid moment, the angle entropy and other conditions are considered to acquire the image complexity through the interaction of the above four modules, which can qualitatively and quantitatively describe the complexity of the image. This is not only conducive to better guiding and grasping the image processing and analysis work, but also has extremely important significance for the fields of image engineering. The complexity of the image can be acquired for different spatial distributions of the same pixel points in the image. The acquired image complexity can be used as an evaluation index to reflect the difficulty of image watermarking technology, image compression and image segmentation, and can be used to evaluate and measure the quality of the image. It can also be used as a feature of the image to reflect the visual feature of the image, and can be used for target recognition, image classification, information steganography. It can also be used as an image visual perception emotion for the research of image cognition and aesthetic calculation.

[0070] Further, in the specific implementation, in the above image complexity acquisition device provided by the embodiment of the present application, the gradient map generation module 11 can be specifically used to convolve the to-be-tested image by using an edge detection algorithm to generate a first gradient map of the to-be-tested image in the X direction and a second gradient map of the to-be-tested image in the Y direction, respectively. It should be noted that the edge detection algorithm of the present application can be Sobel, but it is not limited to Sobel edge extraction operator, and can also be other algorithms, as long as the first gradient map in the X direction and the second gradient map in the Y direction can be extracted respectively.

[0071] In the specific implementation, in the above image complexity acquisition device provided by the embodiment of the present application, the gradient amplitude calculation module 12 can be specifically used to calculate the image edge gradient amplitude by using the first gradient map and the second gradient map by using the following formula:

[0072]

[0073] wherein, Gf is the image edge gradient amplitude, G x is the gradient value of the first gradient map, G y is the gradient value of the second gradient map.

[0074] Further, in the specific implementation, the image complexity obtaining device provided by the embodiment of the present application can further comprise:

[0075] a gradient direction calculation module, configured to calculate the image edge gradient direction by using the first gradient map and the second gradient map;

[0076] In the implementation, the image edge gradient direction (i.e. the angle of each point on the edge) can be calculated by using the following formula:

[0077]

[0078] wherein, Ang is the image edge gradient direction, (x i , y i ) is the coordinate of each point on the edge.

[0079] an angle histogram obtaining module, configured to normalize the calculated image edge gradient direction to obtain an angle histogram;

[0080] an angle entropy statistical module, configured to obtain the probability of each angle from the angle histogram, and to statistically obtain the angle entropy according to the probability of each angle.

[0081] In the specific implementation, the centroid moment calculation module 13 in the image complexity obtaining device provided by the embodiment of the present application can comprise:

[0082] a point set construction unit, configured to construct the image edge point set coordinate in the gradient amplitude map;

[0083] a centroid calculation unit, configured to calculate the image edge centroid coordinate according to the image edge point set coordinate and the point set number;

[0084] In the implementation, the image edge centroid coordinate (C x , C y ) can be calculated according to the image edge point set coordinate (x i , y i ) and the point set number N by using the following formula:

[0085]

[0086] a centroid moment calculation unit, configured to calculate the image edge centroid coordinate (C x , C y by using the image edge point set coordinate and the image edge centroid coordinate (C x , C y ).), the image edge centroid moment of the i-th point is calculated as follows:

[0087] In implementation, the image edge point set coordinates (x i , y i ) and the image edge centroid coordinates are used, and the image edge centroid moment d(x i , y i ) of the i-th point is calculated as follows:

[0088]

[0089] In implementation, the complexity obtaining module 14 in the image complexity obtaining device provided by the embodiment of the present application can include:

[0090] a joint product calculation unit, configured to calculate the joint product of the image edge centroid moment and the image edge gradient amplitude;

[0091] In implementation, the joint product calculation unit can calculate the joint product of the image edge centroid moment d(x i , y i ) and the image edge gradient amplitude Gf(x i , y i ) as follows:

[0092]

[0093] wherein Gf(x i , y i ) is the image edge gradient amplitude of the point (x i , y i ), and N represents the number of points.

[0094] a complexity obtaining unit, configured to obtain the complexity of the image to be measured according to the calculated joint product and the angle entropy;

[0095] In implementation, the complexity obtaining unit can obtain the complexity of the image to be measured according to the calculated joint product and the angle entropy as follows:

[0096] C = e Entropy *log(F(d, Gf)); (7)

[0097]

[0098] wherein C is the complexity of the image to be measured, P(k) is the probability when the angle is k, and Entrop is the angle entropy.

[0099] The image complexity model of the present application is the above-mentioned formula (7). The higher the contrast of an image (referring to the edge gradient of the image), the more chaotic the image (referring to the gradient direction of the image, that is, the angle information entropy of the gradient), the more scattered the target points of the image (referring to the centroid moment), and the more complex the image.

[0100] Based on the same inventive concept, the present application also provides an image complexity acquisition method. Since the principle of the method for solving the problem is similar to the above-mentioned image complexity acquisition device, the implementation of the method can be referred to the implementation of the image complexity acquisition device, and the repeated parts will not be described here.

[0101] In specific implementation, the image complexity acquisition method provided by the present application embodiment, as shown in Figure 13 includes the following steps:

[0102] S1301, respectively generating a first gradient map of the image to be measured in the X direction and a second gradient map in the Y direction;

[0103] S1302, calculating the image edge gradient amplitude and generating a gradient amplitude map by using the first gradient map and the second gradient map;

[0104] S1303, calculating the image edge centroid moment according to the gradient amplitude map;

[0105] S1304, combining the calculated image edge centroid moment and image edge gradient amplitude with angle entropy to acquire the complexity of the image to be measured.

[0106] In the above-mentioned image complexity acquisition method provided by the present application embodiment, the image complexity is acquired by considering the image gradient amplitude, image centroid moment, and angle entropy, which can qualitatively and quantitatively describe the complexity of the image. This is not only conducive to better guiding and grasping the image processing and analysis work, but also has extremely important significance for the fields of image engineering. The complexity can be acquired for different spatial distributions of the same pixel points in the image, and the acquired image complexity can be used as an evaluation index to reflect the difficulty of image watermarking technology, image compression, and image segmentation, for evaluating and measuring the quality of the image, as a feature of the image to reflect the visual feature of the image, for target recognition, image classification, information steganography, as an image visual perception emotion for image cognition and aesthetic calculation research.

[0107] In specific implementation, in the above-mentioned image complexity acquisition method provided by the present application embodiment, the step S1301 respectively generates a first gradient map of the image to be measured in the X direction and a second gradient map in the Y direction, which specifically can include: using an edge detection algorithm to convolve the image to be measured to respectively generate a first gradient map of the image to be measured in the X direction and a second gradient map in the Y direction.

[0108] In the implementation, in the image complexity obtaining method provided by the embodiment of the present application, when the step S1302 of calculating the image edge gradient amplitude is performed, the following steps can also be included: first, the image edge gradient direction is calculated by using the first gradient map and the second gradient map; then, the angle histogram is obtained by normalizing the calculated image edge gradient direction; then, the probability of each angle is obtained from the angle histogram; finally, the angle entropy is calculated according to the probability of each angle.

[0109] In the implementation, in the image complexity obtaining method provided by the embodiment of the present application, the step S1303 of calculating the image edge centroid moment according to the gradient amplitude map can include the following steps: first, the image edge point set coordinates are constructed in the gradient amplitude map; then, the image edge centroid coordinates are calculated according to the image edge point set coordinates and the point set number; finally, the image edge centroid moment is calculated by using the image edge point set coordinates and the image edge centroid coordinates.

[0110] In the implementation, in the image complexity obtaining method provided by the embodiment of the present application, the step S1304 of obtaining the complexity of the image to be tested by combining the calculated image edge centroid moment, the image edge gradient amplitude and the angle entropy can include the following steps: first, the joint product of the image edge centroid moment and the image edge gradient amplitude is calculated; then, the complexity of the image to be tested is obtained according to the calculated joint product and the angle entropy.

[0111] The more specific working process of each step can refer to the corresponding content disclosed in the foregoing embodiments, and will not be described here.

[0112] Correspondingly, the embodiment of the present application also discloses an image complexity obtaining device, which includes a processor and a memory; wherein the processor realizes the image complexity obtaining method disclosed in the foregoing embodiments when executing the computer program stored in the memory. The more specific process of the foregoing method can refer to the corresponding content disclosed in the foregoing embodiments, and will not be described here.

[0113] Further, the present application also discloses a computer readable storage medium for storing a computer program; the computer program is executed by the processor to realize the foregoing disclosed image complexity obtaining method. The more specific process of the foregoing method can refer to the corresponding content disclosed in the foregoing embodiments, and will not be described here.

[0114] The embodiments in the specification are described in a progressive manner, and each embodiment focuses on the difference from other embodiments. The same or similar parts of each embodiment can be referred to each other. For the device, equipment and storage medium disclosed in the embodiments, since they correspond to the method disclosed in the embodiments, the description is relatively simple, and the related parts can refer to the method part.

[0115] Those skilled in the art will further appreciate that the units and algorithms steps of the examples described in connection with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or both. The disclosure is not limited to any particular hardware or software unless specifically stated herein. Where the steps of a method or algorithm have been identified, some or all of those steps can be implemented in hardware, software, or both, depending on the particular application and design constraints imposed on the overall system. Skilled persons can implement the described functionality in varying ways for each particular application, but such implementation does not cause a departure from the scope of the disclosure.

[0116] The steps of a method or algorithm described in connection with the embodiments disclosed herein can be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module can reside in random access memory (RAM), memory, flash memory, read-only memory (ROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0117] In summary, the image complexity acquisition device provided by the embodiments of the present application comprises: a gradient map generation module, configured to generate a first gradient map in an X direction and a second gradient map in a Y direction of a to-be-tested image respectively; a gradient amplitude calculation module, configured to calculate image edge gradient amplitudes and generate a gradient amplitude map by using the first gradient map and the second gradient map; a centroid moment calculation module, configured to calculate image edge centroid moments according to the gradient amplitude map; and a complexity acquisition module, configured to combine the calculated image edge centroid moments and image edge gradient amplitudes with an angular entropy to acquire the complexity of the to-be-tested image. Through the interaction of the above four modules, the image complexity is acquired by considering the conditions of the image gradient amplitude, the image centroid moment, and the angular entropy, which can qualitatively and quantitatively describe the complexity of the image. This is not only conducive to better guiding and grasping the image processing and analysis work, but also has extremely important significance for the fields of image engineering. The complexity can be acquired for different spatial distributions of the same pixel point of an image, the acquired image complexity can be used as an evaluation index to reflect the difficulty of image watermarking technology, image compression, and image segmentation, and can be used to evaluate and measure the quality of an image, and can be used as a feature of an image to reflect the visual feature of the image, and can be used for target recognition, image classification, information steganography, and can be used as an image visual perception emotion for the research of image cognition and aesthetic calculation. In addition, the present application also provides a corresponding method, device, and computer readable storage medium for the image complexity acquisition device, which further makes the above device more practical, and the method, device, and computer readable storage medium have corresponding advantages.

[0118] Finally, it needs to be pointed out that, in this document, relational terms such as first and second and the like can only be intended to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprising", "including", or any other variant thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can also include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without more limitations, an element defined by the statement "comprising a" does not exclude the existence of additional identical elements in the process, method, article, or apparatus including the element.

[0119] The image complexity acquisition device, method, equipment and storage medium provided by the present application are described in detail above, and the principles and implementation manners of the present application are described by applying specific examples in this document. The above description of the embodiments is only used to help understand the method of the present application and its core idea. Meanwhile, for those skilled in the art, according to the idea of the present application, the specific implementation manners and application ranges will be changed, and the above description should not be understood as a limitation of the present application.

Claims

1. An image complexity acquisition apparatus characterized by comprising: The method comprises the following steps: a gradient map generation module is configured to generate a first gradient map in the X direction and a second gradient map in the Y direction of a to-be-tested image respectively; a gradient amplitude calculation module is configured to calculate image edge gradient amplitudes and generate a gradient amplitude map by using the first gradient map and the second gradient map; a centroid moment calculation module is configured to calculate image edge centroid moments according to the gradient amplitude map; a complexity acquisition module is configured to acquire the complexity of the to-be-tested image by combining the calculated image edge centroid moments and the image edge gradient amplitudes with an angle entropy; the complexity acquisition module comprises: a joint product calculation unit configured to calculate the joint product of the image edge centroid moments and the image edge gradient amplitudes; a complexity acquisition unit configured to acquire the complexity of the to-be-tested image according to the calculated joint product and the angle entropy by using the following formula: ； ； ； wherein, is the complexity of the image to be tested, is the centroid moment, is the image edge gradient amplitude, is the image edge point set coordinate, is the image edge centroid moment of the i-th point, is the image edge gradient amplitude of the point N represents the number of point sets, is the probability when the angle is k, is the angle entropy.

2. The image complexity acquisition apparatus according to claim 1, characterized by the gradient amplitude calculation module is specifically configured to calculate the image edge gradient amplitudes by using the following formula: ； wherein is a gradient value of the first gradient map, is a gradient value of the second gradient map.

3. The image complexity acquisition apparatus according to claim 2, characterized by The method further comprises the following steps: a gradient direction calculation module is configured to calculate image edge gradient directions by using the first gradient map and the second gradient map; an angle histogram acquisition module is configured to obtain an angle histogram by performing normalization processing on the calculated image edge gradient directions; an angle entropy statistical module is configured to obtain the probability of each angle from the angle histogram and to statistically obtain the angle entropy according to the probability of each angle.

4. The image complexity acquisition apparatus according to claim 3, characterized by The centroid moment calculation module comprises: a point set construction unit configured to construct image edge point set coordinates in the gradient amplitude map; a centroid calculation unit configured to calculate image edge centroid coordinates according to the image edge point set coordinates and the number of point sets; a centroid moment calculation unit configured to calculate image edge centroid moments by using the image edge point set coordinates and the image edge centroid coordinates.

5. The image complexity acquisition apparatus according to claim 1, characterized by The gradient map generation module is specifically configured to generate the first gradient map in the X direction and the second gradient map in the Y direction of the to-be-tested image by performing convolution on the to-be-tested image by using an edge detection algorithm.

6. An image complexity acquisition method characterized by comprising: The method comprises the following steps: generate a first gradient map in the X direction and a second gradient map in the Y direction of a to-be-tested image respectively; calculate image edge gradient amplitudes and generate a gradient amplitude map by using the first gradient map and the second gradient map; calculate image edge centroid moments according to the gradient amplitude map; acquire the complexity of the to-be-tested image by combining the calculated image edge centroid moments and the image edge gradient amplitudes with an angle entropy, which comprises the following steps: ； ； ； wherein, is the complexity of the image to be tested, is the centroid moment, is the image edge gradient amplitude, is the image edge point set coordinate, is the image edge centroid moment of the i-th point, is the image edge gradient amplitude of the point N represents the number of point sets, is the probability when the angle is k, is the angle entropy.

7. An image complexity acquisition apparatus characterized by comprising: calculate the joint product of the image edge centroid moments and the image edge gradient amplitudes; and acquire the complexity of the to-be-tested image according to the calculated joint product and the angle entropy by using the following formula:

8. A computer-readable storage medium, characterized in that, The method comprises a processor and a memory, wherein the processor executes a computer program stored in the memory to realize the image complexity acquisition method of claim 6. A computer program is stored, wherein the computer program is executed by a processor to realize the image complexity acquisition method of claim 6.

Images