Image fusion method and device and electronic equipment
Feature extraction and fusion of infrared and visible images through quantum neural networks has been solved, and more efficient information fusion and more accurate object detection are achieved.
Patent Information
- Application Number
- CN202410986277.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-23
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-07-23
AI Technical Summary
The prior art has poor feature extraction capabilities in the fusion of infrared images and visible light images, resulting in poor image fusion effect.
The quantum neural network is used to extract features through infrared image quantum encoding network and visible light image quantum encoding network, and fuse infrared feature images and visible light feature images using quantum entanglement characteristics, calculate the mean and variance of the fused feature images, build feature vectors, and Enter the quantum decoding network to generate a fused image.
It improves the effect of image fusion, can learn more feature space information, and enhances the accuracy and reliability of image fusion, especially in object detection and object recognition scenarios.
Smart Images

Figure CN118941906A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of image processing technology, and in particular to an image fusion method, device and electronic equipment. Background Art
[0002] Infrared images and visible light images are complementary. Fusion of infrared images and visible light images can obtain richer and more comprehensive information, with higher accuracy and reliability in target detection, target recognition and other scenarios. The existing classical neural network technology has poor feature extraction capabilities in the task of realizing infrared image and visible light image fusion, resulting in poor image fusion effects. Summary of the invention
[0003] The embodiments of the present application provide an image fusion method, device and electronic device to solve the problem of poor image fusion effect in existing image fusion methods.
[0004] In order to solve the above technical problems, this application is implemented as follows:
[0005] In a first aspect, an embodiment of the present application provides an image fusion method, the image fusion method comprising:
[0006] Acquire a first infrared image and a first visible light image;
[0007] Inputting the first infrared image into an infrared image quantum coding network for feature extraction to obtain an infrared feature image, and inputting the first visible light image into a visible light image quantum coding network for feature extraction to obtain a visible light feature image;
[0008] Inputting the infrared feature image and the visible light feature image into a quantum feature fusion network, and fusing the infrared feature image and the visible light feature image using quantum entanglement characteristics to obtain a fused feature image;
[0009] Calculating the mean and variance of the fused feature image, and constructing a feature vector based on the mean and variance of the fused feature image;
[0010] The feature vector is input into a quantum decoding network to obtain a fused image.
[0011] Optionally, the quantum feature fusion network includes a controlled NOT gate and a controlled P gate;
[0012] The step of inputting the infrared feature image and the visible light feature image into a quantum feature fusion network, and fusing the infrared feature image and the visible light feature image using quantum entanglement characteristics to obtain a fused feature image comprises:
[0013] Using the control NOT gate to entangle corresponding bit information of the infrared characteristic image and the visible light characteristic image;
[0014] Using the controlled P gate to perform quantum deposition on the entangled infrared characteristic image and the visible light characteristic image, respectively, so as to transfer each bit information in the entangled infrared characteristic image and the visible light characteristic image to the corresponding auxiliary bit, and obtain a fused quantum state characteristic image;
[0015] Performing quantum measurement on the fused quantum state characteristic image to obtain a fused characteristic image.
[0016] Optionally, acquiring the first infrared image and the first visible light image includes:
[0017] Acquire a second infrared image and a second visible light image;
[0018] Segmenting the second infrared image to obtain a plurality of infrared fragmented images, and segmenting the second visible light image to obtain a plurality of visible light fragmented images;
[0019] Compressing the plurality of infrared slice images respectively, and splicing the compressed plurality of infrared slice images to obtain the first infrared image;
[0020] The plurality of visible light slice images are compressed respectively, and the compressed plurality of visible light slice images are spliced to obtain the first visible light image.
[0021] Optionally, the compressing the plurality of infrared slice images respectively includes:
[0022] Inputting the plurality of infrared fragment images into the infrared image quantum coding network for feature coding to obtain a plurality of infrared fragment feature images;
[0023] The compressing the plurality of visible light fragmented images respectively comprises:
[0024] The multiple visible light slice images are input into the visible light image quantum coding network for feature coding to obtain multiple visible light slice feature images.
[0025] Optionally, the infrared image quantum coding network includes a first quantum coding layer, a first quantum convolution layer, a first quantum pooling layer and a first quantum measurement layer connected in sequence, and the visible light image quantum coding network includes a second quantum coding layer, a second quantum convolution layer, a second quantum pooling layer and a second quantum measurement layer connected in sequence, the convolution kernel of the first quantum convolution layer is larger than the convolution kernel of the second quantum convolution layer, and the number of layers of the second quantum convolution layer is larger than the number of layers of the first quantum convolution layer.
[0026] Optionally, the quantum decoding network includes a plurality of quantum decoding modules;
[0027] The constructing a feature vector based on the mean and variance of the fused feature image includes:
[0028] Converting the mean and variance of the fused feature image into a one-dimensional latent vector;
[0029] The step of inputting the feature vector into a quantum decoding network to obtain a fused image comprises:
[0030] Inputting the one-dimensional potential vectors into the multiple quantum decoding modules for decoding processing respectively to obtain multiple fused sub-images, wherein different quantum decoding modules respectively decode different parts of the fused feature image to generate different parts of the fused image;
[0031] The multiple fused sub-images are spliced together to obtain a fused image.
[0032] Optionally, the method further comprises:
[0033] Calculating a reconstruction loss value according to the difference between the fused feature image and the fused image;
[0034] calculating a perception loss value according to a difference between the fused image and the first infrared image;
[0035] Calculating a content loss value according to a difference between the fused image and the first visible light image;
[0036] According to the reconstruction loss value, the perception loss value and the content loss value, the structural parameters of the infrared image quantum coding network, the visible light image quantum coding network and the quantum decoding network are adjusted to optimize the infrared image quantum coding network, the visible light image quantum coding network and the quantum decoding network.
[0037] In a second aspect, an embodiment of the present application further provides an image fusion device, the image fusion device comprising:
[0038] A first acquisition module, used to acquire a first infrared image and a first visible light image;
[0039] A first extraction module is used to input the first infrared image into an infrared image quantum coding network for feature extraction to obtain an infrared feature image, and input the first visible light image into a visible light image quantum coding network for feature extraction to obtain a visible light feature image;
[0040] A first fusion module is used to input the infrared feature image and the visible light feature image into a quantum feature fusion network, and fuse the infrared feature image and the visible light feature image using quantum entanglement characteristics to obtain a fused feature image;
[0041] A first construction module, used for calculating the mean and variance of the fused feature image, and constructing a feature vector based on the mean and variance of the fused feature image;
[0042] The first processing module is used to input the feature vector into a quantum decoding network to obtain a fused image.
[0043] Optionally, the quantum feature fusion network includes a controlled NOT gate and a controlled P gate;
[0044] The first fusion module includes:
[0045] A first processing unit, configured to entangle corresponding bit information of the infrared feature image and the visible light feature image by using the control NOT gate;
[0046] A second processing unit is used to use the controlled P gate to perform quantum deposition on the entangled infrared characteristic image and the visible light characteristic image respectively, so as to transfer each bit information in the entangled infrared characteristic image and the visible light characteristic image to the corresponding auxiliary bit, so as to obtain a fused quantum state characteristic image;
[0047] The third processing unit is used to perform quantum measurement on the fused quantum state characteristic image to obtain a fused characteristic image.
[0048] Optionally, the first acquisition module includes:
[0049] A first acquisition unit, used to acquire a second infrared image and a second visible light image;
[0050] a first segmentation unit, configured to segment the second infrared image to obtain a plurality of infrared segmented images, and segment the second visible light image to obtain a plurality of visible light segmented images;
[0051] a fourth processing unit, configured to compress the plurality of infrared slice images respectively, and to stitch the compressed plurality of infrared slice images to obtain the first infrared image;
[0052] The fifth processing unit is used to compress the multiple visible light segment images respectively, and splice the compressed multiple visible light segment images to obtain the first visible light image.
[0053] Optionally, the fourth processing unit includes:
[0054] A first encoding subunit is used to input the plurality of infrared fragment images into the infrared image quantum encoding network for feature encoding to obtain a plurality of infrared fragment feature images;
[0055] The fifth processing unit comprises:
[0056] The second encoding subunit is used to input the multiple visible light slice images into the visible light image quantum coding network for feature encoding to obtain multiple visible light slice feature images.
[0057] Optionally, the infrared image quantum coding network includes a first quantum coding layer, a first quantum convolution layer, a first quantum pooling layer and a first quantum measurement layer connected in sequence, and the visible light image quantum coding network includes a second quantum coding layer, a second quantum convolution layer, a second quantum pooling layer and a second quantum measurement layer connected in sequence, the convolution kernel of the first quantum convolution layer is larger than the convolution kernel of the second quantum convolution layer, and the number of layers of the second quantum convolution layer is larger than the number of layers of the first quantum convolution layer.
[0058] Optionally, the quantum decoding network includes a plurality of quantum decoding modules;
[0059] The first building block comprises:
[0060] A first conversion unit, configured to convert the mean and variance of the fused feature image into a one-dimensional latent vector;
[0061] The first processing module comprises:
[0062] A first decoding unit is used to input the one-dimensional potential vector into the multiple quantum decoding modules for decoding processing to obtain multiple fused sub-images, wherein different quantum decoding modules respectively decode different parts of the fused feature image to generate different parts of the fused image;
[0063] The first stitching unit is used to stitch the multiple fused sub-images to obtain a fused image.
[0064] Optionally, the device further comprises:
[0065] A first calculation module, used for calculating a reconstruction loss value according to a difference between the fused feature image and the fused image;
[0066] a second calculation module, configured to calculate a perception loss value according to a difference between the fused image and the first infrared image;
[0067] A third calculation module, configured to calculate a content loss value according to a difference between the fused image and the first visible light image;
[0068] The first adjustment module is used to adjust the structural parameters of the infrared image quantum coding network, the visible light image quantum coding network and the quantum decoding network according to the reconstruction loss value, the perception loss value and the content loss value, so as to optimize the infrared image quantum coding network, the visible light image quantum coding network and the quantum decoding network.
[0069] In a third aspect, an embodiment of the present application further provides an electronic device, the electronic device comprising a processor, the processor being configured to:
[0070] Acquire a first infrared image and a first visible light image;
[0071] Inputting the first infrared image into an infrared image quantum coding network for feature extraction to obtain an infrared feature image, and inputting the first visible light image into a visible light image quantum coding network for feature extraction to obtain a visible light feature image;
[0072] Inputting the infrared feature image and the visible light feature image into a quantum feature fusion network, and fusing the infrared feature image and the visible light feature image using quantum entanglement characteristics to obtain a fused feature image;
[0073] Calculating the mean and variance of the fused feature image, and constructing a feature vector based on the mean and variance of the fused feature image;
[0074] The feature vector is input into a quantum decoding network to obtain a fused image.
[0075] In a fourth aspect, an embodiment of the present application further provides an electronic device, comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program implements the steps of the above-mentioned image fusion method when executed by the processor.
[0076] In a fifth aspect, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above-mentioned image fusion method are implemented.
[0077] In a sixth aspect, a computer program product is provided, comprising computer instructions, which, when executed by a processor, implement the steps of the image fusion method as described in the first aspect.
[0078] The image fusion method of the embodiment of the present application includes acquiring a first infrared image and a first visible light image; inputting the first infrared image into an infrared image quantum coding network for feature extraction to obtain an infrared feature image, and inputting the first visible light image into a visible light image quantum coding network for feature extraction to obtain a visible light feature image; inputting the infrared feature image and the visible light feature image into a quantum feature fusion network, and using the quantum entanglement characteristics to fuse the infrared feature image and the visible light feature image to obtain a fused feature image; calculating the mean and variance of the fused feature image, and constructing a feature vector based on the mean and variance of the fused feature image; and inputting the feature vector into a quantum decoding network to obtain a fused image. This method is based on the special data expression and data processing capabilities of quantum neural networks, uses different quantum coding networks to extract features from the first infrared image and the first visible light image, and uses the quantum entanglement characteristics of the quantum feature fusion network to fuse the infrared feature image and the visible light feature image. Compared with traditional classical neural networks, quantum neural networks can learn more information in the feature space, thereby improving the effect of image fusion. BRIEF DESCRIPTION OF THE DRAWINGS
[0079] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments of the present application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.
[0080] Figure 1 This is one of the flow charts of the image fusion method provided in the embodiment of the present application;
[0081] Figure 2 It is a quantum feature fusion circuit diagram of a quantum feature fusion network provided in an embodiment of the present application;
[0082] Figure 3 is a schematic diagram of a quantum coding layer provided in an embodiment of the present application;
[0083] Figure 4 It is a schematic diagram of the quantum convolution layer and the quantum pooling layer provided in the embodiment of the present application;
[0084] Figure 5 is a schematic diagram of a quantum circuit provided in an embodiment of the present application;
[0085] Figure 6 This is the second flowchart of the image fusion method provided in the embodiment of the present application;
[0086] Figure 7 is a schematic diagram of an image fusion method provided in an embodiment of the present application;
[0087] Figure 8 is a structural diagram of an image fusion device provided in an embodiment of the present application;
[0088] Fig. 9 It is a structural diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0089] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0090] Infrared images and visible light images are image data in two different wavelength ranges. Infrared images can capture the thermal radiation information of objects and are not affected by light, but often lose detailed texture information; visible light images reflect visual information such as the shape, details and texture of objects, but are easily affected by light, and the imaging effect is poor under conditions such as night. Therefore, the two are complementary. Fusion of infrared images and visible light images can obtain richer and more comprehensive information, and have higher accuracy and reliability in scenes such as target detection and target recognition. The image fusion method of the embodiment of the present application is mainly based on quantum neural networks. Quantum neural networks use quantum mechanical phenomena such as quantum superposition and quantum entanglement to learn features beyond classical neural networks in certain fields, and achieve better simulation of the distribution law between objects. The application of quantum neural networks in the field of image processing has shown some potential advantages, such as quantum image processing, image compression and image generation. This application breaks through the classical neural network technology and designs an image fusion method based on quantum neural networks.
[0091] The present application embodiment provides an image fusion method. Figure 1 , Figure 1 is a flow chart of the image fusion method provided in the embodiment of the present application, such as Figure 1 As shown, the following steps are included:
[0092] Step 101: Acquire a first infrared image and a first visible light image;
[0093] In this step, an initial infrared image is first acquired by an infrared camera, and then image preprocessing is performed on the initial infrared image. In order to prevent the image from being too large to directly perform quantum encoding on all pixels in the image, the initial infrared image after image preprocessing needs to be compressed to obtain a first infrared image. Similarly, a first visible light image is obtained.
[0094] Step 102: input the first infrared image into an infrared image quantum coding network for feature extraction to obtain an infrared feature image, and input the first visible light image into a visible light image quantum coding network for feature extraction to obtain a visible light feature image;
[0095] In this step, since the first infrared image and the first visible light image contain different feature information, two quantum coding networks are constructed respectively, namely, an infrared image quantum coding network and a visible light image quantum coding network. The infrared image quantum coding network is used to extract features of the first infrared image to obtain an infrared feature image; the visible light image quantum coding network is used to extract features of the first visible light image to obtain a visible light feature image.
[0096] Step 103: input the infrared feature image and the visible light feature image into a quantum feature fusion network, and fuse the infrared feature image and the visible light feature image using quantum entanglement characteristics to obtain a fused feature image;
[0097] Although the infrared feature image and the visible light feature image contain different types of image feature information, since these feature information come from the same object, they are just from different perspectives (one is visible light and the other is infrared). Therefore, there is a certain correlation between the infrared feature image and the visible light feature image. In this step, the infrared feature image and the visible light feature image are input into the quantum feature fusion network, and the two feature images are fused using the two quantum coding circuits and the auxiliary bit circuit in the quantum feature fusion network to obtain a fused feature image.
[0098] Step 104: Calculate the mean and variance of the fused feature image, and construct a feature vector based on the mean and variance of the fused feature image;
[0099] In this step, the mean and variance of the fused feature image are calculated. Specifically, the mean of the fused feature image is calculated as the sum of the pixel values of all pixels in the fused feature image divided by the total number of pixels in the fused feature image; the variance of the fused feature image is calculated as the square of the difference between the pixel value of each pixel in the fused feature image and the mean, and then the square of the difference of all pixels is added to get the total sum and divided by the total number of pixels. After calculating the mean and variance of the eigenvalues, the reparameterization technique is used to convert the mean and variance of the fused feature map into a feature vector.
[0100] Step 105: Input the feature vector into a quantum decoding network to obtain a fused image.
[0101] In order to use quantum neural network for image fusion, starting from step 102, the first infrared image is input into the infrared image quantum coding network, and the first visible light image is input into the visible light image quantum coding network, so that the classical image data is converted into quantum state data. The image data of the fused image to be obtained in the end should be classical image data, so the feature vector obtained in step 104 needs to be input into the quantum decoding network to convert the quantum state image data into classical image data.
[0102] In one implementation, a first infrared image and a first visible light image are first acquired, and then, based on the difference in feature information contained in the first infrared image and the first visible light image, a feature extraction is performed on the first infrared image using an infrared image quantum coding network to obtain an infrared feature image, and a feature extraction is performed on the first visible light image using a visible light image quantum coding network to obtain a visible light feature image. The obtained infrared feature image and visible light feature image are then input into a quantum feature fusion circuit, and the infrared feature image and the visible light feature image are feature fused using quantum entanglement features to obtain a fused feature image. A feature vector is constructed based on the mean and variance of the fused feature image. Finally, the feature vector is input into a quantum decoding network to obtain a fused image.
[0103] This implementation is based on the special data expression and data processing capabilities of quantum neural networks, uses different quantum coding networks to extract features from the first infrared image and the first visible light image, and uses the quantum entanglement characteristics of the quantum feature fusion network to fuse the infrared feature image and the visible light feature image. Compared with traditional classical neural networks, quantum neural networks can learn more information in the feature space, thereby improving the effect of image fusion. In addition, quantum also has the ability to process data quickly, and using quantum neural networks for image fusion is conducive to improving the speed of image fusion.
[0104] Optionally, the quantum feature fusion network includes a controlled NOT gate and a controlled P gate;
[0105] The step of inputting the infrared feature image and the visible light feature image into a quantum feature fusion network, and fusing the infrared feature image and the visible light feature image using quantum entanglement characteristics to obtain a fused feature image comprises:
[0106] Using the control NOT gate to entangle corresponding bit information of the infrared characteristic image and the visible light characteristic image;
[0107] Using the controlled P gate to perform quantum deposition on the entangled infrared characteristic image and the visible light characteristic image, respectively, so as to transfer each bit information in the entangled infrared characteristic image and the visible light characteristic image to the corresponding auxiliary bit, and obtain a fused quantum state characteristic image;
[0108] Performing quantum measurement on the fused quantum state characteristic image to obtain a fused characteristic image.
[0109] In one embodiment, see Figure 2 , Figure 2 The quantum feature fusion circuit diagram of the quantum feature fusion network includes an auxiliary circuit, an infrared feature circuit and a visible light feature circuit. First, the infrared feature image is encoded on the infrared feature circuit, and convolution feature extraction is performed on it, and the visible light feature image is encoded on the visible light feature image, and convolution feature extraction is performed on it; then, the circuit corresponding bit information of the two feature images is entangled by using a controlled NOT gate to achieve strong entanglement at the feature space level; then, the entangled infrared feature image and the visible light feature image are quantum deposited by using a controlled P gate to transfer the information of each bit in the entangled infrared feature image and the visible light feature image to the corresponding auxiliary bit, and obtain a fused quantum state feature image; finally, the fused quantum state feature image is quantum measured to convert the quantum state image data into classical image data to obtain a fused feature image.
[0110] This implementation utilizes different characteristic circuits in the quantum characteristic fusion network to fuse the infrared characteristic image and the visible light characteristic image, thereby improving the image fusion effect.
[0111] Optionally, acquiring the first infrared image and the first visible light image includes:
[0112] Acquire a second infrared image and a second visible light image;
[0113] Segmenting the second infrared image to obtain a plurality of infrared fragmented images, and segmenting the second visible light image to obtain a plurality of visible light fragmented images;
[0114] Compressing the plurality of infrared slice images respectively, and splicing the compressed plurality of infrared slice images to obtain the first infrared image;
[0115] The plurality of visible light slice images are compressed respectively, and the compressed plurality of visible light slice images are spliced to obtain the first visible light image.
[0116] In one embodiment, the second infrared image is an image obtained after image preprocessing is performed on an initial infrared image acquired by an infrared camera, and the second visible light image is an image obtained after image preprocessing is performed on an initial visible light image acquired by a visible light camera. Specifically, the image preprocessing process includes:
[0117] (1) performing pixel-level registration on the acquired initial infrared image and initial visible light image using an image registration algorithm;
[0118] (2) Since the initial infrared image often contains a lot of noise, it is necessary to filter and reduce noise on the initial infrared image. At the same time, the image size of the denoised initial infrared image and the image size of the initial visible light image are uniformly bilinearly interpolated to 64*64 to obtain the second infrared image and the second visible light image.
[0119] In view of the quantum circuit scale limitation in the noisy intermediate-scale quantum (NISQ) era, directly using large image information as feature vectors for quantum encoding will exceed the quantum circuit scale limitation. Therefore, it is necessary to segment the second infrared image to obtain multiple infrared slice images, and segment the second visible light image to obtain multiple visible light images.
[0120] Each of the plurality of infrared slice images is compressed, and then the compressed plurality of infrared slice images are spliced to obtain a first visible light image; and each of the plurality of visible light slice images is compressed, and then the compressed plurality of visible light slice images are spliced to obtain a first infrared image. The first infrared image has a smaller image size than the second infrared image, and the first visible light image has a smaller image size than the second visible light image, which is conducive to subsequent quantum encoding of the first infrared image and the first visible light image.
[0121] Optionally, the compressing the plurality of infrared slice images respectively includes:
[0122] Inputting the plurality of infrared fragment images into the infrared image quantum coding network for feature coding to obtain a plurality of infrared fragment feature images;
[0123] The compressing the plurality of visible light fragmented images respectively comprises:
[0124] The multiple visible light slice images are input into the visible light image quantum coding network for feature coding to obtain multiple visible light slice feature images.
[0125] In one embodiment, a plurality of infrared slice images are compressed and processed using an infrared image quantum coding network. Specifically, a plurality of infrared slice images are input into an infrared image quantum coding network, and the infrared image quantum coding network includes a first quantum coding layer, a first quantum convolution layer, a first quantum pooling layer, and a first quantum measurement layer connected in sequence. First, each infrared slice image is input into the first quantum coding layer to encode the classical image data into the quantum system to form quantum state image data; then the quantum state image data is input into the first quantum convolution layer, and the convolution kernel of the quantum convolution layer is used to extract features of the quantum state image data to obtain a quantum state feature image; then the quantum state feature image is input into the quantum pooling layer to reduce the size of the feature image; finally, the reduced quantum state feature image is input into the quantum measurement layer to convert the quantum state image data into classical image data to obtain a plurality of infrared slice feature images. Based on the same processing principle, a plurality of visible light slice images are input into the visible light image quantum coding network for feature encoding to obtain a plurality of visible light slice feature images, and the specific process is not repeated here.
[0126] In this implementation, multiple infrared slice images are input into an infrared image quantum coding network for compression processing, and multiple visible light images are input into a visible light image quantum coding network for compression processing, which is beneficial for obtaining multiple infrared slice feature images and multiple visible light slice feature images of appropriate image size.
[0127] Optionally, the infrared image quantum coding network includes a first quantum coding layer, a first quantum convolution layer, a first quantum pooling layer and a first quantum measurement layer connected in sequence, and the visible light image quantum coding network includes a second quantum coding layer, a second quantum convolution layer, a second quantum pooling layer and a second quantum measurement layer connected in sequence, the convolution kernel of the first quantum convolution layer is larger than the convolution kernel of the second quantum convolution layer, and the number of layers of the second quantum convolution layer is larger than the number of layers of the first quantum convolution layer.
[0128] In one embodiment, see Figure 3 , Figure 3 Schematic diagram of quantum coding layer (including first quantum coding layer and second quantum coding layer), quantum coding layer is used to encode classical image data into quantum state image data. This application adopts probability amplitude encoder to realize quantum coding, which can reduce the number of quantum bits required for encoding exponentially. n The dimensional classical features are normalized and then encoded into the quantum circuit of the image quantum coding network through the quantum encoder to achieve n quantum bit encoding 2 n Dimensional classic features, Figure 3 Shown is 2 quantum bits encoding 4-dimensional classical features.
[0129] See also Figure 4 , Figure 4 It is a schematic diagram of the quantum convolution layer (including the first quantum convolution layer and the second quantum convolution layer) and the quantum pooling layer (including the first quantum pooling layer and the second quantum pooling layer). The quantum convolution layer can realize convolution kernels of different sizes through different parameter quantities, and use parameterized quantum circuits to perform matrix operations on the quantum state image data of the quantum coding layer. Quantum convolution has unique advantages that classical convolution does not have. The convolution kernel of quantum convolution is orthogonal, so it can better regularize the features and is not prone to gradient explosion and other effects. Therefore, the number of convolutions can be appropriately increased to extract rich features.
[0130] Since the edges and details of visible light images are more prominent, the small receptive field of the 2*2 convolution kernel is used to extract the details, and the details are further extracted through deeper convolutions; infrared images have more semantic information and global features, so the large receptive field of the 3*3 convolution kernel is used to extract global features, and the global features are further extracted through fewer convolutions. Two quantum circuits can be constructed by different convolution kernel sizes (one quantum circuit in the infrared image quantum coding network, and one quantum circuit in the visible light image quantum coding network) to achieve feature extraction of the two quantum circuits acting on different images. Further, it should be noted that the quantum circuit for feature extraction of infrared images has two quantum convolution layers for convolution, and the quantum circuit for feature extraction of visible light images has four convolution layers for convolution, so as to achieve the extraction of different features.
[0131] The purpose of the quantum pooling layer is to reduce the size of the feature image, thereby reducing the number of parameters. Quantum has a natural advantage in pooling. It only needs to ignore the middle and last quantum bits of the quantum circuit to achieve the purpose of reducing quantum bits and realize a 2*2 pooling dimension reduction operation with a stride of 2. The neglect here refers to the neglect of the middle quantum bits and the last quantum bit in the quantum circuit. For example, the current circuit is an 8-bit quantum circuit, which can achieve a maximum of 16*16 (2 8 ) image encoding, pooling is dimensionality reduction, so after the pooling part, the image should be 8*8, that is, 2 6 , that is, 6 bits. Therefore, for any bit quantum circuit, we only need to ignore the middle and last bits, which is a pooling operation.
[0132] The quantum measurement layer (both the first and second quantum measurement layers can refer to the explanation of the quantum measurement layer) provides classical information about the expectation, variance, or probability of the quantum system. For the quantum expectation measured on the state, see Figure 5 , the desired measurement of the quantum state can be completed through the Pauli Z gate.
[0133] Optionally, the quantum decoding network includes a plurality of quantum decoding modules;
[0134] The constructing a feature vector based on the mean and variance of the fused feature image includes:
[0135] Converting the mean and variance of the fused feature image into a one-dimensional latent vector;
[0136] The step of inputting the feature vector into a quantum decoding network to obtain a fused image comprises:
[0137] Inputting the one-dimensional potential vectors into the multiple quantum decoding modules for decoding processing respectively to obtain multiple fused sub-images, wherein different quantum decoding modules respectively decode different parts of the fused feature image to generate different parts of the fused image;
[0138] The multiple fused sub-images are spliced together to obtain a fused image.
[0139] In one embodiment, after calculating the mean and variance of the fused feature image, the mean and variance of the fused feature image are converted into a one-dimensional latent vector using a reparameterization technique. Since the noise space of the fused feature image is large, the quantum decoding network used in this embodiment is composed of multiple quantum decoding modules, and different quantum decoding modules decode different parts of the fused feature image respectively to generate different parts of the fused image, and then splice the different parts to finally form a fused image. By introducing different quantum decoding modules, the quantum decoding network can become more robust and have better adaptability to noise and changes, thereby improving the quality and clarity of the formed fused image.
[0140] It should be noted that the quantum decoding network also includes quantum coding layer, quantum convolution layer and quantum measurement layer, but there is no quantum pooling layer.
[0141] Optionally, the method further comprises:
[0142] Calculating a reconstruction loss value according to the difference between the fused feature image and the fused image;
[0143] calculating a perception loss value according to a difference between the fused image and the first infrared image;
[0144] Calculating a content loss value according to a difference between the fused image and the first visible light image;
[0145] According to the reconstruction loss value, the perception loss value and the content loss value, the structural parameters of the infrared image quantum coding network, the visible light image quantum coding network and the quantum decoding network are adjusted to optimize the infrared image quantum coding network, the visible light image quantum coding network and the quantum decoding network.
[0146] In one implementation, in order to reduce the difference between the distribution of the fused feature image and the distribution of the fused image and make the clarity of the formed fused image better, the reconstruction loss value can be calculated according to the difference between the fused feature image and the fused image to optimize the structural parameters of the quantum neural network. The specific calculation of the reconstruction loss value can refer to the following calculation formula:
[0147]
[0148] Among them, μ j represents the mean of the jth fusion feature map, σ j Represents the variance of the j-th fused feature map.
[0149] Since the first infrared image has more advanced semantic information, including shape and structure, the perceptual loss value can be calculated based on the difference between the fused image and the first infrared image to optimize the structural parameters of the quantum neural network. The calculation of the perceptual loss value can be specifically referred to the following calculation formula:
[0150] L2(x,y)=E x,y [||Vgg(y)-Vgg(x)|| 1 ]
[0151] Wherein, x represents the first infrared image, y represents the fused image, Vgg(x) represents the features of x calculated by the vgg network, and Vgg(y) represents the features of y calculated by the vgg network.
[0152] Since the first visible light image has rich information such as brightness, contrast, edge and details, the content loss value can be calculated based on the difference between the fused image and the first visible light image to optimize the structural parameters of the quantum neural network. The calculation of the content damage value can be based on the following calculation formula:
[0153]
[0154] Among them, X represents the first visible light image, Y represents the fused image, μ x is the mean of x, μ y is the mean of y, σ xy is the covariance of x and y, σ x 2 and σ y 2 is the variance of x and y, c1 and c 2 is a constant.
[0155] In this implementation, by reconstructing loss values, perception loss values, and content loss values, the structural parameters of the infrared image quantum coding network, the visible light image quantum coding network, and the quantum decoding network can be adjusted, thereby optimizing the infrared image quantum coding network, the visible light image quantum coding network, and the quantum decoding network.
[0156] In one embodiment, see Figure 6 and Figure 7 , perform image preprocessing on the acquired initial infrared image and initial visible light image, and obtain a second infrared image and a second visible light image accordingly. Segment the second infrared image to obtain multiple infrared slice images, and segment the second visible light image to obtain multiple visible light slice images. Use the infrared image feature coding network to compress the multiple infrared slice images and then splice them to obtain a first infrared image, and use the visible light image feature coding network to compress the multiple visible light slice images and then splice them to obtain a first visible light image. At this time, the image sizes of the first infrared image and the first visible light image obtained can be quantum encoded into the quantum circuit.
[0157] The first infrared image is input into the infrared image quantum coding network for feature extraction to obtain an infrared feature image, and the first visible light image is input into the visible light image quantum coding network for feature extraction to obtain a visible light feature image. The infrared feature image and the visible light feature image are then input into the quantum feature fusion network, and the infrared feature image and the visible light feature image are fused using the quantum entanglement characteristics to obtain a fused feature image. The mean and variance of the fused feature image are calculated, and the mean and variance of the fused feature image are converted into a one-dimensional latent vector Z; the one-dimensional latent vector Z and noise are input into the quantum decoding network, and different parts of the fused feature image are decoded using different quantum decoding modules to generate different parts of the fused image.
[0158] See also Figure 8 , Figure 8 is a structural diagram of an image fusion device provided in one embodiment of the present application. Figure 8 As shown, the image fusion device 800 includes:
[0159] A first acquisition module 801 is used to acquire a first infrared image and a first visible light image;
[0160] A first extraction module 802 is used to input the first infrared image into an infrared image quantum coding network for feature extraction to obtain an infrared feature image, and to input the first visible light image into a visible light image quantum coding network for feature extraction to obtain a visible light feature image;
[0161] A first fusion module 803 is used to input the infrared feature image and the visible light feature image into a quantum feature fusion network, and fuse the infrared feature image and the visible light feature image using quantum entanglement characteristics to obtain a fused feature image;
[0162] A first construction module 804 is used to calculate the mean and variance of the fused feature image, and construct a feature vector based on the mean and variance of the fused feature image;
[0163] The first processing module 805 is used to input the feature vector into a quantum decoding network to obtain a fused image.
[0164] Optionally, the quantum feature fusion network includes a controlled NOT gate and a controlled P gate;
[0165] The first fusion module includes:
[0166] A first processing unit, configured to entangle corresponding bit information of the infrared feature image and the visible light feature image by using the control NOT gate;
[0167] A second processing unit is used to use the controlled P gate to perform quantum deposition on the entangled infrared characteristic image and the visible light characteristic image respectively, so as to transfer each bit information in the entangled infrared characteristic image and the visible light characteristic image to the corresponding auxiliary bit, so as to obtain a fused quantum state characteristic image;
[0168] The third processing unit is used to perform quantum measurement on the fused quantum state characteristic image to obtain a fused characteristic image.
[0169] Optionally, the first acquisition module includes:
[0170] A first acquisition unit, used to acquire a second infrared image and a second visible light image;
[0171] a first segmentation unit, configured to segment the second infrared image to obtain a plurality of infrared segmented images, and segment the second visible light image to obtain a plurality of visible light segmented images;
[0172] a fourth processing unit, configured to compress the plurality of infrared slice images respectively, and to stitch the compressed plurality of infrared slice images to obtain the first infrared image;
[0173] The fifth processing unit is used to compress the multiple visible light segment images respectively, and splice the compressed multiple visible light segment images to obtain the first visible light image.
[0174] Optionally, the fourth processing unit includes:
[0175] A first encoding subunit is used to input the plurality of infrared fragment images into the infrared image quantum encoding network for feature encoding to obtain a plurality of infrared fragment feature images;
[0176] The fifth processing unit comprises:
[0177] The second encoding subunit is used to input the multiple visible light slice images into the visible light image quantum coding network for feature encoding to obtain multiple visible light slice feature images.
[0178] Optionally, the infrared image quantum coding network includes a first quantum coding layer, a first quantum convolution layer, a first quantum pooling layer and a first quantum measurement layer connected in sequence, and the visible light image quantum coding network includes a second quantum coding layer, a second quantum convolution layer, a second quantum pooling layer and a second quantum measurement layer connected in sequence, the convolution kernel of the first quantum convolution layer is larger than the convolution kernel of the second quantum convolution layer, and the number of layers of the second quantum convolution layer is larger than the number of layers of the first quantum convolution layer.
[0179] Optionally, the quantum decoding network includes a plurality of quantum decoding modules;
[0180] The first building block comprises:
[0181] A first conversion unit, configured to convert the mean and variance of the fused feature image into a one-dimensional latent vector;
[0182] The first processing module comprises:
[0183] A first decoding unit is used to input the one-dimensional potential vector into the multiple quantum decoding modules for decoding processing to obtain multiple fused sub-images, wherein different quantum decoding modules respectively decode different parts of the fused feature image to generate different parts of the fused image;
[0184] The first stitching unit is used to stitch the multiple fused sub-images to obtain a fused image.
[0185] Optionally, the device further comprises:
[0186] A first calculation module, used for calculating a reconstruction loss value according to a difference between the fused feature image and the fused image;
[0187] a second calculation module, configured to calculate a perception loss value according to a difference between the fused image and the first infrared image;
[0188] A third calculation module, configured to calculate a content loss value according to a difference between the fused image and the first visible light image;
[0189] The first adjustment module is used to adjust the structural parameters of the infrared image quantum coding network, the visible light image quantum coding network and the quantum decoding network according to the reconstruction loss value, the perception loss value and the content loss value, so as to optimize the infrared image quantum coding network, the visible light image quantum coding network and the quantum decoding network.
[0190] It should be noted that this embodiment is Figure 1 The specific implementation of the corresponding device side embodiment can be found in Figure 1 To avoid repeated description, the relevant description of the embodiment shown in is not repeated in this embodiment, and the same beneficial effects can be achieved.
[0191] The present application also provides an electronic device. Since the principle of solving the problem by the electronic device is similar to the image fusion method in the present application, the implementation of the electronic device can refer to the implementation of the method, and the repeated parts will not be repeated. Fig. 9 As shown, the electronic device of the embodiment of the present application includes: a processor 900, which is used to read the program in the memory 920 and execute the following process:
[0192] Acquire a first infrared image and a first visible light image;
[0193] Inputting the first infrared image into an infrared image quantum coding network for feature extraction to obtain an infrared feature image, and inputting the first visible light image into a visible light image quantum coding network for feature extraction to obtain a visible light feature image;
[0194] Inputting the infrared feature image and the visible light feature image into a quantum feature fusion network, and fusing the infrared feature image and the visible light feature image using quantum entanglement characteristics to obtain a fused feature image;
[0195] Calculating the mean and variance of the fused feature image, and constructing a feature vector based on the mean and variance of the fused feature image;
[0196] The feature vector is input into a quantum decoding network to obtain a fused image.
[0197] Among them, Fig. 9In the embodiment, the bus architecture may include any number of interconnected buses and bridges, specifically linking various circuits of one or more processors represented by processor 900 and memory represented by memory 920. The bus architecture may also link various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art and are therefore not further described herein. The bus interface provides an interface. The processor 900 is responsible for managing the bus architecture and general processing, and the memory 920 may store data used by the processor 900 when performing operations.
[0198] Optionally, the quantum feature fusion network includes a controlled NOT gate and a controlled P gate;
[0199] The processor 900 is used to read the program in the memory 920 and execute the following process:
[0200] Using the control NOT gate to entangle corresponding bit information of the infrared characteristic image and the visible light characteristic image;
[0201] Using the controlled P gate to perform quantum deposition on the entangled infrared characteristic image and the visible light characteristic image, respectively, so as to transfer each bit information in the entangled infrared characteristic image and the visible light characteristic image to the corresponding auxiliary bit, and obtain a fused quantum state characteristic image;
[0202] Performing quantum measurement on the fused quantum state characteristic image to obtain a fused characteristic image.
[0203] Optionally, the processor 900 is configured to read a program in the memory 920 and perform the following process:
[0204] Acquire a second infrared image and a second visible light image;
[0205] Segmenting the second infrared image to obtain a plurality of infrared fragmented images, and segmenting the second visible light image to obtain a plurality of visible light fragmented images;
[0206] Compressing the plurality of infrared slice images respectively, and splicing the compressed plurality of infrared slice images to obtain the first infrared image;
[0207] The plurality of visible light slice images are compressed respectively, and the compressed plurality of visible light slice images are spliced to obtain the first visible light image.
[0208] Optionally, the processor 900 is configured to read a program in the memory 920 and perform the following process:
[0209] Inputting the plurality of infrared fragment images into the infrared image quantum coding network for feature coding to obtain a plurality of infrared fragment feature images;
[0210] The compressing the plurality of visible light fragmented images respectively comprises:
[0211] The multiple visible light slice images are input into the visible light image quantum coding network for feature coding to obtain multiple visible light slice feature images.
[0212] Optionally, the infrared image quantum coding network includes a first quantum coding layer, a first quantum convolution layer, a first quantum pooling layer and a first quantum measurement layer connected in sequence, and the visible light image quantum coding network includes a second quantum coding layer, a second quantum convolution layer, a second quantum pooling layer and a second quantum measurement layer connected in sequence, the convolution kernel of the first quantum convolution layer is larger than the convolution kernel of the second quantum convolution layer, and the number of layers of the second quantum convolution layer is larger than the number of layers of the first quantum convolution layer.
[0213] Optionally, the quantum decoding network includes a plurality of quantum decoding modules;
[0214] The processor 900 is used to read the program in the memory 920 and execute the following process:
[0215] Converting the mean and variance of the fused feature image into a one-dimensional latent vector;
[0216] The processor 900 is used to read the program in the memory 920 and execute the following process:
[0217] Inputting the one-dimensional potential vectors into the multiple quantum decoding modules for decoding processing respectively to obtain multiple fused sub-images, wherein different quantum decoding modules respectively decode different parts of the fused feature image to generate different parts of the fused image;
[0218] The multiple fused sub-images are spliced together to obtain a fused image.
[0219] Optionally, the processor 900 is configured to read a program in the memory 920 and perform the following process:
[0220] Calculating a reconstruction loss value according to the difference between the fused feature image and the fused image;
[0221] calculating a perception loss value according to a difference between the fused image and the first infrared image;
[0222] Calculating a content loss value according to a difference between the fused image and the first visible light image;
[0223] According to the reconstruction loss value, the perception loss value and the content loss value, the structural parameters of the infrared image quantum coding network, the visible light image quantum coding network and the quantum decoding network are adjusted to optimize the infrared image quantum coding network, the visible light image quantum coding network and the quantum decoding network.
[0224] The embodiment of the present application also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, each process of the above-mentioned image fusion method embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it is not repeated here. The computer-readable storage medium is, for example, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.
[0225] The present application also provides a computer program product, including computer instructions, which, when executed by a processor, implement the above Figure 1 The various processes of the method embodiment shown can achieve the same technical effect, and will not be described again here to avoid repetition.
[0226] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.
[0227] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present application.
[0228] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.
Claims
1. An image fusion method, characterized in that: include: Acquire a first infrared image and a first visible light image; Inputting the first infrared image into an infrared image quantum coding network for feature extraction to obtain an infrared feature image, and inputting the first visible light image into a visible light image quantum coding network for feature extraction to obtain a visible light feature image; Inputting the infrared feature image and the visible light feature image into a quantum feature fusion network, and fusing the infrared feature image and the visible light feature image using quantum entanglement characteristics to obtain a fused feature image; Calculating the mean and variance of the fused feature image, and constructing a feature vector based on the mean and variance of the fused feature image; The feature vector is input into a quantum decoding network to obtain a fused image.
2. The image fusion method according to claim 1, characterized in that: The quantum feature fusion network includes a controlled NOT gate and a controlled P gate; The step of inputting the infrared feature image and the visible light feature image into a quantum feature fusion network, and fusing the infrared feature image and the visible light feature image using quantum entanglement characteristics to obtain a fused feature image comprises: Using the control NOT gate to entangle corresponding bit information of the infrared characteristic image and the visible light characteristic image; Using the controlled P gate to perform quantum deposition on the entangled infrared characteristic image and the visible light characteristic image, respectively, so as to transfer each bit information in the entangled infrared characteristic image and the visible light characteristic image to the corresponding auxiliary bit, and obtain a fused quantum state characteristic image; Performing quantum measurement on the fused quantum state characteristic image to obtain a fused characteristic image.
3. The image fusion method according to claim 1, characterized in that: The acquiring of the first infrared image and the first visible light image comprises: Acquire a second infrared image and a second visible light image; Segmenting the second infrared image to obtain a plurality of infrared fragmented images, and segmenting the second visible light image to obtain a plurality of visible light fragmented images; Compressing the plurality of infrared slice images respectively, and splicing the compressed plurality of infrared slice images to obtain the first infrared image; The plurality of visible light slice images are compressed respectively, and the compressed plurality of visible light slice images are spliced to obtain the first visible light image.
4. The image fusion method according to claim 3, characterized in that: The compressing the plurality of infrared slice images respectively comprises: Inputting the plurality of infrared fragment images into the infrared image quantum coding network for feature coding to obtain a plurality of infrared fragment feature images; The compressing the plurality of visible light fragmented images respectively comprises: The multiple visible light slice images are input into the visible light image quantum coding network for feature coding to obtain multiple visible light slice feature images.
5. The image fusion method according to any one of claims 1 to 4, characterized in that: The infrared image quantum coding network includes a first quantum coding layer, a first quantum convolution layer, a first quantum pooling layer and a first quantum measurement layer connected in sequence, and the visible light image quantum coding network includes a second quantum coding layer, a second quantum convolution layer, a second quantum pooling layer and a second quantum measurement layer connected in sequence, the convolution kernel of the first quantum convolution layer is larger than the convolution kernel of the second quantum convolution layer, and the number of layers of the second quantum convolution layer is larger than the number of layers of the first quantum convolution layer.
6. The image fusion method according to claim 1, characterized in that: The quantum decoding network includes a plurality of quantum decoding modules; The constructing a feature vector based on the mean and variance of the fused feature image includes: Converting the mean and variance of the fused feature image into a one-dimensional latent vector; The step of inputting the feature vector into a quantum decoding network to obtain a fused image comprises: Inputting the one-dimensional potential vectors into the multiple quantum decoding modules for decoding processing respectively to obtain multiple fused sub-images, wherein different quantum decoding modules respectively decode different parts of the fused feature image to generate different parts of the fused image; The multiple fused sub-images are spliced together to obtain a fused image.
7. The image fusion method according to claim 1, characterized in that: The method further comprises: Calculating a reconstruction loss value according to the difference between the fused feature image and the fused image; calculating a perception loss value according to a difference between the fused image and the first infrared image; Calculating a content loss value according to a difference between the fused image and the first visible light image; According to the reconstruction loss value, the perception loss value and the content loss value, the structural parameters of the infrared image quantum coding network, the visible light image quantum coding network and the quantum decoding network are adjusted to optimize the infrared image quantum coding network, the visible light image quantum coding network and the quantum decoding network.
8. An image fusion device, characterized in that: The image fusion device comprises: A first acquisition module, used to acquire a first infrared image and a first visible light image; A first extraction module is used to input the first infrared image into an infrared image quantum coding network for feature extraction to obtain an infrared feature image, and input the first visible light image into a visible light image quantum coding network for feature extraction to obtain a visible light feature image; A first fusion module is used to input the infrared feature image and the visible light feature image into a quantum feature fusion network, and fuse the infrared feature image and the visible light feature image using quantum entanglement characteristics to obtain a fused feature image; A first construction module, used for calculating the mean and variance of the fused feature image, and constructing a feature vector based on the mean and variance of the fused feature image; The first processing module is used to input the feature vector into a quantum decoding network to obtain a fused image.
9. An electronic device, characterized in that: comprising a processor configured to: Acquire a first infrared image and a first visible light image; Inputting the first infrared image into an infrared image quantum coding network for feature extraction to obtain an infrared feature image, and inputting the first visible light image into a visible light image quantum coding network for feature extraction to obtain a visible light feature image; Inputting the infrared feature image and the visible light feature image into a quantum feature fusion network, and fusing the infrared feature image and the visible light feature image using quantum entanglement characteristics to obtain a fused feature image; Calculating the mean and variance of the fused feature image, and constructing a feature vector based on the mean and variance of the fused feature image; The feature vector is input into a quantum decoding network to obtain a fused image.
10. An electronic device, characterized in that: The method comprises a processor, a memory and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the image fusion method according to any one of claims 1 to 7.
11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the image fusion method according to any one of claims 1 to 7 are implemented.
12. A computer program product, characterized in that The method comprises computer instructions, which, when executed by a processor, implement the steps of the image fusion method according to any one of claims 1 to 7.
Citation Information
Patent Citations
Image generation method and device based on quantum generative adversarial network
CN116263972A
Temperature measurement method and device, equipment and storage medium
CN117073848A
Infrared and visible light image fusion method, device and system and storage medium
CN117152037A
Image multi-classification method based on multi-branch mixed quantum classical neural network
CN117237715A
Target object recognition method and device, electronic equipment and storage medium
CN117557781A
