Image processing method and device and electronic equipment

Abstract

The invention relates to the technical field of image processing, in particular to an image processing method and device and electronic equipment. In the method, an image to be detected is acquired, and depth features of the image to be detected are extracted. And converting the to-be-detected image from the spatial domain to the frequency domain, and extracting the spectrum features of the to-be-detected image. And extracting statistical characteristics of the to-be-detected image. Wherein the statistical characteristics are used for representing gray level distribution in the image. And adopting a classification model to classify the multi-modal feature vectors, and determining the category of the to-be-detected image. Wherein the multi-modal feature vector is obtained by fusing the depth feature, the spectrum feature and the statistical feature according to a preset fusion rule. The multi-modal feature vector is used for reflecting different dimension information of the to-be-detected image. The classification model is used for distinguishing categories of different images based on the multi-modal feature vectors. According to the scheme, whether the image is a real image or not can be accurately and efficiently identified.

Description

Technical Field

[0001] The present application relates to the field of image processing technology, and in particular to an image processing method, device and electronic device. Background Art

[0002] With the rapid progress of artificial intelligence technology, Generative Adversarial Networks (GANs) have demonstrated remarkable image generation capabilities in the field of deep learning. Through its unique architecture, an adversarial training mechanism consisting of a generator and a discriminator, GANs can gradually optimize the generation model, so that the images it produces have unprecedented visual quality. Under certain conditions, these algorithm-generated images are almost indistinguishable from real images, and it is difficult to distinguish them from real pictures with the naked eye.

[0003] However, the double-edged sword nature of this technology has gradually emerged. On the one hand, GANs have shown great application potential and value in many fields such as artistic creation, entertainment content generation, data enhancement, and auxiliary medical diagnosis, greatly enriching the diversity and creativity of the digital world. On the other hand, its powerful forgery ability has also aroused widespread concern from all walks of life. Malicious users may use GANs to generate false images, conduct online fraud, mislead public opinion, forge evidence and other illegal activities, posing a serious threat to the society's trust system, legal justice and personal privacy security.

[0004] How to determine whether an image is a real image becomes a debatable issue. Summary of the invention

[0005] The embodiments of the present application provide an image processing method, device and electronic device for accurately and efficiently identifying whether an image is a real image.

[0006] In a first aspect, an embodiment of the present application provides an image processing method for acquiring an image to be detected and extracting depth features of the image to be detected, where the depth features are used to represent deep-level features in the image;

[0007] Convert the image to be detected from the spatial domain to the frequency domain, and extract the spectrum features of the image to be detected. The spectrum features are used to represent the distribution and intensity of different frequencies in the image.

[0008] Extract statistical features of the image to be detected, where the statistical features are used to represent the grayscale distribution in the image;

[0009] A classification model is used to classify the multimodal feature vector to determine whether the image to be detected is a real image. The multimodal feature vector is obtained by fusing depth features, spectral features and statistical features according to preset fusion rules. The multimodal feature vector is used to reflect the different dimensional information of the image to be detected. The classification model is used to determine the categories of different images based on the multimodal feature vector.

[0010] Optionally, the deep learning model is used to extract the deep features of the image to be detected, specifically including:

[0011] A convolutional neural network is used to extract features of the image to be detected and obtain the deep features of the image to be detected.

[0012] Optionally, the above-mentioned converting the image to be detected from the spatial domain to the frequency domain to extract the frequency spectrum features of the image to be detected specifically includes:

[0013] Fourier transform is used to convert the image to be detected from the spatial domain to the frequency domain to extract the spectrum characteristics of the image to be detected.

[0014] Optionally, the above statistical features include at least one of a gray level co-occurrence matrix and a color histogram, the gray level co-occurrence matrix is ​​used to describe image texture features, and the color histogram is used to describe the distribution of different gray levels in the image.

[0015] Optionally, the above method further includes:

[0016] The deep features, spectral features and statistical features are concatenated to obtain a multimodal feature vector.

[0017] Optionally, before concatenating the depth features, the spectrum features, and the statistical features to obtain the multimodal feature vector, the method further includes:

[0018] The depth features, spectrum features and statistical features are standardized.

[0019] In a second aspect, an embodiment of the present application provides an image processing device, the device comprising:

[0020] A transceiver module, used for acquiring an image to be detected;

[0021] A processing module, used for extracting the depth features of the image to be detected, where the depth features are used for representing the deep features in the image;

[0022] The processing module is also used to convert the image to be detected from the spatial domain to the frequency domain, and extract the frequency spectrum features of the image to be detected, where the frequency spectrum features are used to represent the distribution and intensity of different frequencies in the image;

[0023] The processing module is also used to extract statistical features of the image to be detected, and the statistical features are used to represent the grayscale distribution in the image;

[0024] The processing module is also used to classify the multimodal feature vector using a classification model to determine whether the image to be detected is a real image. The multimodal feature vector is obtained by fusing depth features, spectral features and statistical features according to preset fusion rules. The multimodal feature vector is used to reflect the different dimensional information of the image to be detected. The classification model is used to determine the categories of different images based on the multimodal feature vector.

[0025] In a third aspect, an embodiment of the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the computer program is executed by the processor, the processor implements any one of the image processing methods in the first aspect.

[0026] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the image processing method of any one of the first aspects is implemented.

[0027] In a fifth aspect, an embodiment of the present application further provides a computer program product, including a computer program, which is executed by a processor to implement an image processing method as described in any one of the first aspects above.

[0028] The technical effects brought about by any one of the implementation methods in the second to fifth aspects can refer to the technical effects brought about by the corresponding implementation method in the first aspect, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 A schematic diagram of an application scenario of image processing provided in an embodiment of the present application;

[0030] Figure 2 A flow chart of an image processing method provided in an embodiment of the present application;

[0031] Figure 3 An exemplary flowchart of image processing provided in an embodiment of the present application;

[0032] Figure 4 A schematic diagram of an image processing device provided in an embodiment of the present application;

[0033] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0034] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.

[0035] The application scenarios described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided by the embodiments of the present application. It is known to those skilled in the art that with the emergence of new application scenarios, the technical solutions provided by the embodiments of the present application are also applicable to similar technical problems. In the description of the present application, unless otherwise specified, the meaning of "multiple" is two or more.

[0036] With the rapid progress of artificial intelligence technology, Generative Adversarial Networks (GANs) have demonstrated remarkable image generation capabilities in the field of deep learning. Through its unique architecture, an adversarial training mechanism consisting of a generator and a discriminator, GANs can gradually optimize the generation model, so that the images it produces have unprecedented visual quality. Under certain conditions, these algorithm-generated images are almost indistinguishable from real images, and it is difficult to distinguish them from real pictures with the naked eye.

[0037] However, the double-edged sword nature of this technology has gradually emerged. On the one hand, GANs have shown great application potential and value in many fields such as artistic creation, entertainment content generation, data enhancement, and auxiliary medical diagnosis, greatly enriching the diversity and creativity of the digital world. On the other hand, its powerful forgery ability has also aroused widespread concern from all walks of life. Malicious users may use GANs to generate false images, conduct online fraud, mislead public opinion, forge evidence and other illegal activities, posing a serious threat to the society's trust system, legal justice and personal privacy security.

[0038] How to distinguish real images from GANs generated images has become a debatable issue.

[0039] In order to solve the above problems, an embodiment of the present application provides an image processing method. In this method, an image to be detected is obtained, and the depth features of the image to be detected are extracted, and the depth features are used to represent the deep features in the image. The image to be detected is converted from the spatial domain to the frequency domain, and the spectrum features of the image to be detected are extracted, and the spectrum features are used to represent the distribution and intensity of different frequencies in the image. The statistical features of the image to be detected are extracted, and the statistical features are used to represent the grayscale distribution in the image. A classification model is used to classify the multimodal feature vector to determine the category of the image to be detected. The multimodal feature vector is obtained by fusing the depth features, the spectrum features and the statistical features according to the preset fusion rules. The multimodal feature vector is used to reflect the different dimensional information of the image to be detected, and the classification model is used to distinguish the categories of different images based on the multimodal feature vector.

[0040] In this way, by combining the depth features, spectral features and statistical features to obtain a multimodal feature vector, and performing image classification based on the multimodal feature vector, a more comprehensive image information judgment can be achieved, the accuracy of classification can be improved, and efficient and accurate identification of whether the image is a real image can be achieved, which is convenient for subsequent security protection. At the same time, the method of identifying image categories based on multimodal feature vectors makes image classification no longer dependent on a single type of feature. Even if a certain feature is affected by noise, occlusion or deformation, other features can still provide useful information, thereby enhancing the robustness of the present application and being more widely used in various scenarios requiring image classification.

[0041] A schematic diagram of an application scenario of an optional image processing method according to an embodiment of the present application. Figure 1 As shown, it includes a server 100 and a terminal 101. The server 100 and the terminal 101 can be communicatively connected through a network to implement the image processing method of the present application.

[0042] The user can use the server 100 to interact with the terminal 101 through the network, such as receiving or sending messages, etc. Various client applications can be installed on the terminal 101, such as program writing applications, web browser applications, search applications, etc.

[0043] It is understandable that in the embodiment of the present application, the server 100 can be implemented as an independent server or a server cluster composed of multiple servers. The terminal 101 can be various electronic devices with a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, desktop computers, etc.

[0044] The server 100 is used to obtain the image to be detected and extract the deep features of the image to be detected. The deep features are used to represent the deep features in the image. The image to be detected is converted from the spatial domain to the frequency domain, and the spectral features of the image to be detected are extracted. The spectral features are used to represent the distribution and intensity of different frequencies in the image. The statistical features of the image to be detected are extracted, and the statistical features are used to represent the grayscale distribution in the image. The classification model is used to classify the multimodal feature vector to determine the category of the image to be detected. The multimodal feature vector is obtained by fusing the deep features, spectral features and statistical features according to the preset fusion rules. The multimodal feature vector is used to reflect the different dimensional information of the image to be detected, and the classification model is used to distinguish the categories of different images based on the multimodal feature vector.

[0045] like Figure 2 As shown, a flowchart of an image processing method provided by an embodiment of the present application can be specifically included in the following steps. The following description is given by taking the server side executing the image processing method of the present application as an example.

[0046] Step S201: Acquire an image to be detected, and extract depth features of the image to be detected.

[0047] Among them, deep features are used to represent deep features in images.

[0048] In an optional embodiment, the server may use a convolutional neural network to extract features of the image to be detected to obtain deep features of the image to be detected.

[0049] In this way, since the deep features extracted by the convolutional neural network have good generalization ability and can be applied to different data sets and scenarios, the accuracy of subsequent classification based on deep features can be improved.

[0050] For example, the convolution layer in a convolutional neural network can extract local features in an image through convolution operations. Each convolution layer contains multiple convolution kernels. The convolution kernel slides on the output feature map of the input image to be detected, calculates the weighted sum of the convolution kernel and the pixels in the corresponding area, and then obtains the output feature map through a nonlinear activation function. The activation layer performs a nonlinear transformation on the output of the convolution layer, increases the nonlinear factor of the network, and enables the network to learn more complex features. Among them, the activation function can use a linear rectified unit (ReLU). The pooling layer can reduce the dimension of the feature map, reduce the number of parameters and the amount of calculation in the network, and retain important features. At the same time, at the end of the convolutional neural network, there are one or more fully connected layers. Each neuron in the fully connected layer is connected to all neurons in the previous layer to integrate the feature information learned by all previous layers. The number of neurons in the output layer corresponds to the number of categories of the classification task. The probability of each category is calculated through functions such as softmax to achieve deep feature extraction of the image.

[0051] Step S202: convert the image to be detected from the spatial domain to the frequency domain, and extract the frequency spectrum features of the image to be detected.

[0052] Among them, spectral features are used to represent the distribution and intensity of different frequencies in the image.

[0053] In an optional embodiment, the server may use Fourier transform to convert the image to be detected from the spatial domain to the frequency domain to extract the frequency spectrum features of the image to be detected.

[0054] In this way, through Fourier transform, the spectral characteristics of the image are intuitively displayed. The spectrum diagram can clearly show the distribution of different frequency components in the image, including low-frequency and high-frequency parts. This makes it easier to identify key features in the image later, making frequency domain analysis more intuitive, improving the efficiency of image processing, and improving image processing effects.

[0055] Step S203: extracting statistical features of the image to be detected.

[0056] Among them, the statistical features are used to represent the grayscale distribution in the image. The statistical features include at least one of the grayscale co-occurrence matrix and the color histogram. The grayscale co-occurrence matrix is ​​used to describe the image texture features. The grayscale co-occurrence matrix reflects the texture information of the image by counting the frequency of the grayscale of two pixels at a certain distance and in a certain direction in the image. The color histogram is used to describe the distribution of different grayscale levels in the image.

[0057] Optionally, the statistical features of the image may also include one or more of the following features:

[0058] 1. Image size

[0059] The size of an image refers to its width and height, usually measured in pixels.

[0060] 2. Grayscale average

[0061] Grayscale mean refers to the arithmetic mean of the grayscale values ​​of all pixels in an image, which reflects the average brightness of the image.

[0062] 3. Grayscale standard deviation

[0063] Grayscale standard deviation is the standard deviation of the grayscale value distribution of the image, which reflects the degree of dispersion of the grayscale value of the image relative to the grayscale mean.

[0064] 4. Covariance matrix and correlation coefficient

[0065] The covariance matrix is ​​a measure of the degree of correlation between two images. When the covariance matrix is ​​0, it means that the two images are independent of each other; otherwise, it means that the two images are dependent on each other.

[0066] 5. Other statistical characteristics

[0067] Range: The range is the difference between the maximum and minimum grayscale values ​​in the image, which reflects the extension of the grayscale value of the image. Median: The median is the number in the middle after the grayscale values ​​of all pixels in the image are arranged in order of size. Mode: The mode is the grayscale value that appears most frequently in the image, which reflects the central tendency of the grayscale values ​​in the image.

[0068] Optionally, after extracting the deep features, spectrum features, and statistical features, the server can perform standardization on the deep features, spectrum features, and statistical features, which helps to improve the convergence speed and performance of the classification model later.

[0069] For example, the server can scale the depth features, spectrum features, and statistical features to a uniform range, such as between 0 and 1, to achieve standardization.

[0070] After the deep features, spectral features and statistical features are standardized, the deep features, spectral features and statistical features are fused according to preset fusion rules to obtain a multimodal feature vector.

[0071] In one possible scenario, the server may concatenate the deep features, spectral features, and statistical features to obtain a multimodal feature vector.

[0072] For example, assuming that the depth feature, spectrum feature, and statistical feature are one-dimensional arrays respectively, the server can connect the depth feature, spectrum feature, and statistical feature along the preset axis direction. After concatenating the three one-dimensional arrays, a new one-dimensional array with a higher dimension can be obtained, namely, the multimodal feature vector.

[0073] In another possible scenario, the server may also use a weighted average method to assign different weights to the depth features, spectral features, and statistical features, and then add them together according to the weights to obtain a fused multimodal feature vector.

[0074] It is understandable that the weight allocation may be preset by those skilled in the art based on the importance of image features, and may be changed according to different application scenarios.

[0075] Optionally, before extracting and fusing features on the image to be detected, the server may also perform preprocessing on the image to be detected, such as scaling, cropping, normalization, etc., to ensure the consistency and validity of the input data.

[0076] Step S204: using a classification model to classify the multimodal feature vector to determine whether the image to be detected is a real image.

[0077] Among them, the multimodal feature vector is obtained by fusing the depth features, spectral features and statistical features according to the preset fusion rules. The multimodal feature vector is used to reflect the different dimensional information of the image to be detected, and the classification model is used to distinguish the categories of different images based on the multimodal feature vector.

[0078] The classification model is described below:

[0079] In the embodiments of the present application, the classification model can adopt models such as support vector machine (SVM), random forest, etc. In this way, it is convenient for the subsequent classification model to automatically learn complex feature representations. During the training process, a labeled data set can be used to train the classification model. During the training process, the classification model can autonomously learn how to map the multimodal feature vector to the corresponding category label. Among them, the data set includes images of known categories and their corresponding multimodal feature vectors. The category label of the image of the known category can be "real" or "GANs generated".

[0080] like Figure 3 As shown, an exemplary flowchart of an image processing method provided in an embodiment of the present application may include the following steps:

[0081] Step S301: using a convolutional neural network to extract features of the image to be detected, and obtaining deep features of the image to be detected;

[0082] Step S302: convert the image to be detected from the spatial domain to the frequency domain using Fourier transform to extract the frequency spectrum features of the image to be detected;

[0083] Step S303, extracting statistical features of the image to be detected;

[0084] Step S304: standardize the depth features, spectrum features, and statistical features;

[0085] Step S305: concatenate the depth features, spectrum features and statistical features to obtain a multimodal feature vector;

[0086] Step S306: Use a classification model to classify the multimodal feature vector to determine whether the image to be detected is a real image.

[0087] Figure 4 Schematic diagram of the structure of an image processing device provided in an embodiment of the present application. Figure 4 As shown, the device includes: a transceiver module 401 and a processing module 402.

[0088] The transceiver module 401 is used to obtain the image to be detected;

[0089] A processing module 402 is used to extract the depth features of the image to be detected, where the depth features are used to represent the deep features in the image;

[0090] The processing module 402 is further used to convert the image to be detected from the spatial domain to the frequency domain, and extract the frequency spectrum features of the image to be detected, where the frequency spectrum features are used to represent the distribution and intensity of different frequencies in the image;

[0091] The processing module 402 is also used to extract statistical features of the image to be detected, where the statistical features are used to represent the grayscale distribution in the image;

[0092] The processing module 402 is also used to classify the multimodal feature vector using a classification model to determine whether the image to be detected is a real image. The multimodal feature vector is obtained by fusing depth features, spectral features and statistical features according to preset fusion rules. The multimodal feature vector is used to reflect the different dimensional information of the image to be detected, and the classification model is used to determine the categories of different images based on the multimodal feature vector.

[0093] Optionally, the deep learning model is used to extract the depth features of the image to be detected, and the processing module 402 is specifically used for:

[0094] A convolutional neural network is used to extract features of the image to be detected and obtain the deep features of the image to be detected.

[0095] Optionally, the above-mentioned image to be detected is converted from the spatial domain to the frequency domain to extract the frequency spectrum characteristics of the image to be detected, and the processing module 402 is specifically used for:

[0096] Fourier transform is used to convert the image to be detected from the spatial domain to the frequency domain to extract the spectrum characteristics of the image to be detected.

[0097] Optionally, the above statistical features include at least one of a gray level co-occurrence matrix and a color histogram, the gray level co-occurrence matrix is ​​used to describe image texture features, and the color histogram is used to describe the distribution of different gray levels in the image.

[0098] Optionally, the processing module 402 is further configured to:

[0099] The deep features, spectral features and statistical features are concatenated to obtain a multimodal feature vector.

[0100] Optionally, before the depth features, spectrum features and statistical features are concatenated to obtain a multimodal feature vector, the processing module 402 is further used to:

[0101] The depth features, spectrum features and statistical features are standardized.

[0102] Based on the same technical concept, an electronic device is also provided in an embodiment of the present application, and the electronic device can realize the functions of the aforementioned image processing device.

[0103] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0104] At least one processor 501, and a memory 502 connected to the at least one processor 501. The specific connection medium between the processor 501 and the memory 502 is not limited in the embodiment of the present application. Figure 5 In the example, the processor 501 and the memory 502 are connected via a bus 500. The bus 500 is Figure 5 The connections between other components are shown in bold lines, and are not intended to be limiting. The bus 500 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 5 Only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus. Alternatively, the processor 501 can also be called a controller, and there is no limitation on the name.

[0105] In the embodiment of the present application, the memory 502 stores instructions that can be executed by at least one processor 501. The at least one processor 501 can execute an image processing method discussed above by executing the instructions stored in the memory 502. The processor 501 can implement Figure 4 The functions of each module in the device shown.

[0106] Among them, the processor 501 is the control center of the device, and can use various interfaces and lines to connect the various parts of the entire control device. By running or executing instructions stored in the memory 502 and calling data stored in the memory 502, the various functions of the device and processing data, the device can be monitored as a whole.

[0107] In one possible design, the processor 501 may include one or more processing units, and the processor 501 may integrate an application processor and a modem processor, wherein the application processor mainly processes an operating system, a driver interface, and application programs, and the modem processor mainly processes wireless communications. It is understandable that the modem processor may not be integrated into the processor 501. In some embodiments, the processor 501 and the memory 502 may be implemented on the same chip, and in some embodiments, they may also be implemented separately on separate chips.

[0108] The processor 501 may be a general-purpose processor, such as a central processing unit (CPU), a digital signal processor, an application-specific integrated circuit, a field programmable gate array or other programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component, and may implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of the present application. A general-purpose processor may be a microprocessor or any conventional processor, etc. The steps of an image processing method disclosed in the embodiments of the present application may be directly embodied as being executed by a hardware processor, or may be executed by a combination of hardware and software modules in the processor.

[0109] The memory 502 is a non-volatile computer-readable storage medium that can be used to store non-volatile software programs, non-volatile computer executable programs and modules. The memory 502 may include at least one type of storage medium, such as a flash memory, a hard disk, a multimedia card, a card-type memory, a random access memory (Random Access Memory, RAM), a static random access memory (Static Random Access Memory, SRAM), a programmable read-only memory (Programmable Read Only Memory, PROM), a read-only memory (Read Only Memory, ROM), an electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, EEPROM), a magnetic memory, a disk, an optical disk, etc. The memory 502 is any other medium that can be used to carry or store a desired program code in the form of an instruction or data structure and can be accessed by a computer, but is not limited thereto. The memory 502 in the embodiment of the present application can also be a circuit or any other device that can realize a storage function, for storing program instructions and / or data.

[0110] By programming the processor 501, the code corresponding to the image processing method described in the above embodiment can be fixed into the chip, so that the chip can execute the code when running. Figure 2 The embodiment shown is an image processing method. How to design and program the processor 501 is a technique known to those skilled in the art and will not be described in detail here.

[0111] It should be noted here that the above-mentioned electronic device provided in the embodiment of the present application can implement all the method steps implemented in the above-mentioned method embodiment, and can achieve the same technical effect. The parts and beneficial effects of this embodiment that are the same as the method embodiment will not be described in detail here.

[0112] An embodiment of the present application further provides a computer-readable storage medium, which stores computer-executable instructions. The computer-executable instructions are used to enable a computer to execute an image processing method in the above embodiment.

[0113] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0114] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 Flow or flows and / or boxes Figure 1 A device that performs the functions specified in a box or multiple boxes.

[0115] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 Flow or flows and / or boxes Figure 1 The functions specified in one or more boxes.

[0116] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 Flow or flows and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0117] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. An image processing method, characterized in that: The method comprises: Acquire an image to be detected, and extract depth features of the image to be detected, wherein the depth features are used to represent deep-level features in the image; Converting the image to be detected from the spatial domain to the frequency domain, extracting the frequency spectrum features of the image to be detected, wherein the frequency spectrum features are used to represent the distribution and intensity of different frequencies in the image; Extracting statistical features of the image to be detected, where the statistical features are used to represent grayscale distribution in the image; A classification model is used to classify the multimodal feature vector to determine whether the image to be detected is a real image. The multimodal feature vector is obtained by fusing the depth feature, the spectrum feature and the statistical feature according to a preset fusion rule. The multimodal feature vector is used to reflect the different dimensional information of the image to be detected. The classification model is used to determine the categories of different images based on the multimodal feature vector.

2. The method according to claim 1, characterized in that The extracting the depth features of the image to be detected by using a deep learning model specifically includes: A convolutional neural network is used to extract features of the image to be detected to obtain deep features of the image to be detected.

3. The method according to claim 1, characterized in that: The step of converting the image to be detected from the spatial domain to the frequency domain and extracting the frequency spectrum features of the image to be detected specifically includes: The image to be detected is converted from the spatial domain to the frequency domain by Fourier transform, and the frequency spectrum features of the image to be detected are extracted.

4. The method according to claim 1, characterized in that: The statistical features include at least one of a gray level co-occurrence matrix and a color histogram. The gray level co-occurrence matrix is ​​used to describe image texture features, and the color histogram is used to describe the distribution of different gray levels in an image.

5. The method according to claim 1, characterized in that The method further comprises: The depth feature, the spectrum feature and the statistical feature are concatenated to obtain the multimodal feature vector.

6. The method according to claim 5, characterized in that Before the deep features, the spectrum features and the statistical features are concatenated to obtain the multimodal feature vector, the method further includes: The depth feature, the spectrum feature and the statistical feature are standardized.

7. An image processing device, characterized in that: The device comprises: A transceiver module, used for acquiring an image to be detected; A processing module, used for extracting the depth features of the image to be detected, wherein the depth features are used for representing the deep features in the image; The processing module is further used to convert the image to be detected from the spatial domain to the frequency domain, and extract the frequency spectrum features of the image to be detected, wherein the frequency spectrum features are used to represent the distribution and intensity of different frequencies in the image; The processing module is further used to extract statistical features of the image to be detected, where the statistical features are used to represent the grayscale distribution in the image; The processing module is also used to classify the multimodal feature vector using a classification model to determine whether the image to be detected is a real image. The multimodal feature vector is obtained by fusing the depth feature, the spectrum feature and the statistical feature according to a preset fusion rule. The multimodal feature vector is used to reflect the different dimensional information of the image to be detected. The classification model is used to determine the categories of different images based on the multimodal feature vector.

8. An electronic device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is used to cause the computer to execute the method according to any one of claims 1 to 6.

10. A computer program product, characterized in that When the computer program product is called by a computer, the computer executes the method according to any one of claims 1 to 6.

Images