Illumination parameter estimation method based on lightweight neural network model

Abstract

The invention relates to the technical field of mixed reality illumination estimation and reconstruction, in particular to an illumination parameter estimation method based on a lightweight neural network model, and the method comprises the steps: obtaining an RGB-L four-channel input image; constructing a feature extraction layer; constructing a lightweight illumination estimation network model, and performing training; and predicting illumination parameters. According to the method, illumination information is effectively separated, interference of irrelevant features is removed, and the quality of input data is optimized, so that the light-weight illumination estimation network model can accurately predict illumination parameters.

Description

Technical Field

[0001] The present invention relates to the technical field of mixed reality illumination estimation and reconstruction, and in particular to a method for estimating illumination parameters based on a lightweight neural network model. Background Art

[0002] Lighting information describes the distribution of light in a physical environment. It typically includes the direction of the light source, such as the sun's azimuth and altitude, light intensity, color temperature, and the reflectance characteristics of ambient light. These parameters collectively constitute the scene's lighting model, the foundation for visual interaction between virtual objects and the real environment. In virtual reality (VR) and augmented reality (AR) technologies, accurate restoration of lighting information directly determines the consistency of lighting between virtual content and the real scene, thereby impacting the realism and immersion of the user experience. In fields such as industrial inspection, military simulation, remote collaboration, and entertainment, VR systems must ensure that the lighting properties of virtual objects, such as equipment annotations, navigation guides, and dynamic special effects, match those of the real environment in order to dynamically integrate computer-generated virtual objects with the real world. For example, in outdoor AR equipment maintenance scenarios, if virtual tools fail to cast accurate shadows based on the lighting direction, operators may misjudge the tool's spatial position. In real-time compositing of special effects in film and television, lighting differences between virtual characters and the real-life footage can significantly reduce visual credibility. Therefore, restoring lighting information is a core challenge in achieving high-fidelity virtual-real fusion.

[0003] Early approaches to illumination estimation relied primarily on traditional computer vision and graphics techniques, typically based on hand-crafted feature extractors and physical models, such as those based on object reflectivity and shadows. These methods required specialized equipment or specialized measurement media to measure illumination, and were unable to provide timely feedback on changing illumination information within a scene. Deep learning-based illumination analysis methods for outdoor images typically employ sky parameter models as supporting tools, using ambient lighting parameters as labeled data. Deep learning networks are then used to train and optimize this data.

[0004] However, current deep learning methods often rely on large network architectures and require extensive computing resources for training and inference. For devices with weaker computing power, such as mobile VR devices or edge computing platforms, the scale and computational complexity of these network models can become a bottleneck, making them unable to meet the needs of real-time lighting prediction.

[0005] Therefore, it is necessary to provide a lighting parameter estimation method based on a lightweight neural network model to solve the above problems. Summary of the Invention

[0006] In view of the problems of insufficient accuracy and real-time performance in existing illumination parameter estimation methods based on deep learning, the present invention provides an illumination parameter estimation method based on a lightweight neural network model to solve the existing problems.

[0007] A first aspect of the present invention provides a method for estimating illumination parameters based on a lightweight neural network model, the method adopting the following technical solutions: Perform image segmentation on the outdoor scene panoramic image to obtain the sky area image; Separate and combine the multiple color channels of the sky area image to obtain an RGB-L four-channel input image; Construct a feature extraction layer, which includes a first Ghost module, a fully decoupled fully connected attention module, an SE attention module, and a second Ghost module. The feature extraction layer is used to use the first Ghost module and the fully decoupled fully connected attention module to respectively extract the features of the four-channel input image input by the input layer, and multiply the two features to obtain a first feature, perform a depth separation convolution on the first feature to obtain a convolved second feature, use the SE attention module to compress and excite the second feature to obtain a channel weight, multiply the channel weight with the feature of the corresponding channel of the second feature to obtain a third feature, use the second Ghost module to perform dimensionality reduction processing on the third feature to obtain a fourth feature, and fuse the fourth feature with the four-channel input image to obtain the final feature; A lightweight illumination estimation network model is constructed based on the feature extraction layer; a dataset is constructed based on the four-channel input image and the illumination parameter labels in the image, the four-channel input image and the illumination parameter labels in the image in the dataset are normalized respectively to obtain a target dataset, and the lightweight illumination estimation network model is trained based on the target dataset to obtain a target network model; The normalized four-channel input image corresponding to the panoramic image of the outdoor scene to be estimated is input into the target network model to predict the lighting parameters.

[0008] A further technical solution of the present invention is to segment the outdoor scene panoramic image to obtain the sky area image by the following steps: Annotate the sky area in the collected outdoor scene panoramic images and construct a sky area segmentation dataset; Build a sky area segmentation model based on U-Net; The segmentation dataset is input into the sky region segmentation model to obtain a sky region image.

[0009] A further technical solution of the present invention is to separate and combine the multiple color channels of the sky area image to obtain the RGB-L four-channel input image by the following steps: Map the sky area image from RGB color space to LAB color space to obtain the target image; The L channel of the target image is combined with the RGB three channels of the sky area image to obtain an RGB-L four-channel input image.

[0010] According to a further technical solution of the present invention, the illumination parameters include: solar azimuth, solar altitude, solar color, sky color and sky turbidity.

[0011] According to a further technical solution of the present invention, the normalized expression of the four-channel input image in the data set is:

[0012] Where, Represents the pixel value of a channel after normalization of the four-channel input image in the dataset; Represents the pixel value of a channel of the four-channel input image in the dataset, Represents the pixel mean of a channel of the four-channel input image in the dataset; Represents the pixel standard deviation of a channel of the four-channel input image in the dataset.

[0013] According to a further technical solution of the present invention, the normalized expression of the illumination parameter label in the four-channel input image in the data set is:

[0014] Where, Represents the label value after normalization of the illumination parameters in the four-channel input image in the dataset; Label values ​​of illumination parameters in the four-channel input images in the dataset; is the minimum label value, The maximum label value.

[0015] A second aspect of the present invention provides a lighting parameter estimation system based on a lightweight neural network model, the system comprising: The image processing module is used to segment the outdoor scene panoramic image to obtain the sky area image; separate and combine the multiple color channels of the sky area image to obtain the RGB-L four-channel input image; A feature extraction module is used to construct a feature extraction layer. The feature extraction layer includes a first Ghost module, a fully decoupled fully connected attention module, an SE attention module, and a second Ghost module. The feature extraction layer is used to use the first Ghost module and the fully decoupled fully connected attention module to respectively extract the features of the four-channel input image input by the input layer, and multiply the two features to obtain a first feature. The first feature is subjected to a depth-separated convolution to obtain a convolved second feature. The second feature is compressed and excited by the SE attention module to obtain a channel weight. The channel weight is multiplied by the feature of the corresponding channel of the second feature to obtain a third feature. The third feature is subjected to dimensionality reduction processing by the second Ghost module to obtain a fourth feature. The fourth feature is fused with the four-channel input image to obtain a final feature. A lightweight illumination estimation network module is used to construct a lightweight illumination estimation network model based on the feature extraction layer; a dataset is constructed based on the four-channel input image and the illumination parameter labels in the image, the four-channel input image and the illumination parameter labels in the image in the dataset are normalized to obtain a target dataset, and the lightweight illumination estimation network model is trained based on the target dataset to obtain a target network model; The parameter prediction module is used to input the normalized four-channel input image corresponding to the panoramic image of the outdoor scene to be estimated into the target network model to predict the lighting parameters.

[0016] The third aspect of the present invention provides an electronic device, comprising a memory, a processor, and a computer program stored on the memory and runnable on the processor. When the computer program is executed by the processor, the steps of a lighting parameter estimation method based on a lightweight neural network model of the present invention are implemented.

[0017] The fourth aspect of the present invention provides a storage medium, including a memory, a processor, and a computer program stored in the memory and runnable on the processor. When the computer program is executed by the processor, the steps of the lighting parameter estimation method based on a lightweight neural network model of the present invention are implemented.

[0018] A fifth aspect of the present invention provides a program product, comprising a computer program, which, when executed by the processor, implements the steps of a lighting parameter estimation method based on a lightweight neural network model of the present invention.

[0019] The beneficial effects of the present invention are: Through the preprocessing steps of image segmentation and separation and combination of multiple color channels of the image, the lighting information is effectively separated, the interference of irrelevant features is removed, and the quality of the input data is optimized. Then, by constructing a lightweight lighting estimation network model based on the Ghost module based on cheap feature extraction, the fully decoupled fully connected attention module, and the SE attention module, accurate prediction of lighting parameters is achieved based on the lightweight lighting estimation network model. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0021] Figure 1 Schematic diagram of a flow chart of a method for estimating illumination parameters based on a lightweight neural network model of the present invention; Figure 2 Schematic diagram of the process of obtaining an RGB-L four-channel input image in an embodiment of the present invention; Figure 3 Schematic diagram of the structure of the feature extraction layer of the lightweight illumination estimation network model in an embodiment of the present invention; Figure 4 Schematic diagram of the structure of a lightweight illumination estimation network model in an embodiment of the present invention; Figure 5 This is a comparison chart of the accuracy visualization results of the sun position prediction of the lightweight illumination estimation network model in an embodiment of the present invention and the existing model; Figure 6 FIG. 1 is a schematic diagram of an electronic device suitable for implementing an embodiment of the present invention. DETAILED DESCRIPTION

[0022] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0023] An embodiment of a method for estimating illumination parameters based on a lightweight neural network model of the present invention is as follows: Figure 1 As shown, including: S1. Get the RGB-L four-channel input image; Specifically, the outdoor scene panoramic image is segmented to obtain a sky area image; and the multiple color channels of the sky area image are separated and combined to obtain an RGB-L four-channel input image.

[0024] Exemplarily, in a specific embodiment, the steps of performing image segmentation on a panoramic image of an outdoor scene to obtain a sky area image are as follows: collecting panoramic images of outdoor scenes from the SUN360 public panoramic image dataset, annotating the sky area in image segmentation and annotation software, constructing a segmentation dataset of the sky area, and constructing a sky area segmentation model based on U-Net; inputting the segmentation dataset into the sky area segmentation model to obtain a sky area image, wherein the pixel values ​​outside the sky area of ​​the image are all 0.

[0025] For example, Figure 2 As shown, in a specific embodiment, the steps of separating and combining the multiple color channels of the sky area image to obtain the RGB-L four-channel input image are as follows: mapping the sky area image from the RGB color space to the LAB color space to obtain a target image; segmenting the L channel from the target image, and combining the L channel of the target image with the three RGB channels of the sky area image to obtain the RGB-L four-channel input image.

[0026] S2, construct feature extraction layer; Specifically, such as Figure 3 As shown, the feature extraction layer of this embodiment includes: a first Ghost module, a fully decoupled fully connected attention module, an SE attention module, and a second Ghost module. The feature extraction layer is used to use the first Ghost module and the fully decoupled fully connected attention module to respectively extract the features of the four-channel input image input by the input layer, and multiply the two features to obtain a first feature, perform deep separation convolution on the first feature to obtain a second feature after convolution, use the SE attention module to compress and excite the second feature to obtain a channel weight, multiply the channel weight with the feature of the corresponding channel of the second feature to obtain a third feature, use the second Ghost module to perform dimensionality reduction processing on the third feature to obtain a fourth feature, and fuse the fourth feature and the four-channel input image to obtain the final feature.

[0027] S3. Build a lightweight illumination estimation network model and train it; Specifically, a lightweight illumination estimation network model is constructed based on the feature extraction layer; a data set is constructed based on the four-channel input image and the illumination parameter labels in the image, the four-channel input image and the illumination parameter labels in the data set are normalized respectively to obtain a target data set, and the lightweight illumination estimation network model is trained based on the target data set to obtain a target network model.

[0028] For example, in a specific embodiment, the illumination parameters include the sun azimuth angle , solar altitude angle , sun color , the first controlling parameter of the sun's size , the second control parameter of the sun's size , sky turbidity t and sky color .

[0029] For example, Figure 4 As shown in a specific embodiment, the lightweight illumination estimation network model has 21 layers, the input layer uses standard convolution, and the next 16 layers are Figure 3 The feature extraction layer shown, the input image size is ,The last three layers are fully connected layers, which first reduce the dimension through the first two fully connected layers and then output the parameters through the linear layer.

[0030] For example, in a specific embodiment, the illumination parameter label data in the dataset is analyzed. The solar altitude angle and azimuth angle are both in the range of 0-1°, while other parameters, such as sky turbidity, are in the range of 13-26, which may cause gradient vanishing or exploding problems. The four-channel input image and the illumination parameter label are normalized. Specifically, the normalized expression of the four-channel input image in the dataset is:

[0031] Where, Represents the pixel value of a channel after normalization of the four-channel input image in the dataset; Represents the pixel value of a channel of the four-channel input image in the dataset, Represents the pixel mean of a channel of the four-channel input image in the dataset; Represents the pixel standard deviation of a channel of the four-channel input image in the dataset.

[0032] The normalized expression of the illumination parameter label in the four-channel input image in the dataset is:

[0033] Where, Represents the label value after normalization of the illumination parameters in the four-channel input image in the dataset; Label values ​​of illumination parameters in the four-channel input images in the dataset; is the minimum label value, The maximum label value.

[0034] Exemplarily, in a specific embodiment, the steps of training a lightweight illumination estimation network model based on a target data set to obtain a target network model are as follows: taking the normalized image and the normalized label as the target data set, inputting the target data set into the lightweight illumination estimation network model for training to obtain the optimal model weights, and thus obtaining the target network model. In the training of the target network model, the number of samples selected for each training, batchsize, is 16, the learning rate lr is 0.002, and the number of training rounds epoch is 200.

[0035] S4, predicting illumination parameters; Specifically, the normalized four-channel input image corresponding to the panoramic image of the outdoor scene to be estimated is input into the target network model to predict the illumination parameters.

[0036] The following tests the lightweight neural network model of this embodiment with specific parameters: First, the target dataset is divided into a target training set and a target test set. The target training set is used to train the lightweight neural network model to obtain the target network model. The target network model is evaluated using the four-channel input image in the target test set. The lightweight neural network model of the present invention is compared with the existing model (i.e., Jin et al. Comparison model, Jin X, Deng P, Li X, et al. Sun-sky model estimation from outdoor images[J]. Journalof Ambient Intelligence and Humanized Computing, 2022, 13(11): The parameter file size of the parameters of the proposed lightweight neural network model is shown in Table 1. The evaluation index of the solar position is the angular error between the true solar position and the predicted solar position, and the unit is the percentage of the number of pictures within the same angular error range in the test set pictures. When the cumulative error of the solar position angle is less than 10°, as shown in Table 2, the percentage of the number of pictures of the lightweight neural network model of the present invention in the test set pictures is 1.8 times that of the existing model. Table 2 also shows the percentage of the number of pictures under other cumulative error values ​​of the solar position angle, as shown in Table 3. The average cumulative error of the remaining parameters estimated by the lightweight neural network model of the present invention is about 1 / 5 of that of the existing model. The prediction visualization results of the solar position by the lightweight neural network model of the present invention and the existing model are shown in Figure 3. Figure 5 shown.

[0037] Table 1

[0038] Table 2

[0039] Table 3

[0040] Figure 6 Schematic diagram of an electronic device suitable for implementing an embodiment of the present invention is shown. Figure 6 The electronic device shown is only an example and should not limit the functions and scope of use of the embodiments of the present invention.

[0041] like Figure 6 As shown, the electronic device includes a central processing unit (CPU) 101, which can perform various appropriate actions and processes based on programs stored in a read-only memory (ROM) 102 or programs loaded from a storage unit 108 into a random access memory (RAM) 103. RAM 103 also stores various programs and data required for system operation. CPU 101, ROM 102, and RAM 103 are interconnected via a bus 104. An input / output (I / O) interface 105 is also connected to bus 104.

[0042] The following components are connected to the I / O interface 105: an input section 106 including a keyboard, mouse, and the like; an output section 107 including devices such as a cathode ray tube (CRT), a liquid crystal display (LCD), and speakers; a storage section 108 including devices such as a hard disk; and a communication section 109 including a network interface card such as a LAN (Local Area Network) card or a modem. The communication section 109 performs communication processing via a network such as the Internet. A drive 110 is also connected to the I / O interface 105 as needed. Removable media 111, such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory, is installed in the drive 110 as needed, so that computer programs read from the media can be installed in the storage section 108 as needed.

[0043] In particular, according to embodiments of the present invention, the processes described below with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program stored on a storage medium, the computer program containing program code for executing the methods illustrated in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 109 and / or installed from removable media 111. When executed by central processing unit (CPU) 101, the computer program performs the various functions defined in the system of the present application.

[0044] Specifically, the above-mentioned electronic device can be a computer, a tablet computer or a server device.

[0045] It should be noted that the storage medium described in the embodiments of the present invention may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to, an electrical connection having one or more conductors, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In the present invention, a computer-readable signal medium may include a data signal transmitted in baseband or as part of a carrier wave, which carries computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any storage medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code contained on the storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, or any suitable combination thereof.

[0046] It should be noted that, as another aspect, the present application also provides a storage medium, which can be included in an electronic device; or it can exist independently without being installed in the electronic device. The above storage medium carries one or more programs, and when the above one or more programs are executed by an electronic device, the electronic device implements the method described in the following embodiments. For example, the electronic device can implement the following Figure 1 The individual steps of the method are shown.

[0047] In one embodiment, the present application also provides a computer program product, including a computer program, which, when executed by a processor, implements Figure 1 The steps of the method shown.

[0048] Furthermore, the accompanying drawings are merely illustrative of the processes included in the method according to exemplary embodiments of the present invention and are not intended to be limiting. It should be understood that the processes illustrated in the accompanying drawings do not indicate or limit the temporal order of these processes. Furthermore, it should be understood that these processes may be executed synchronously or asynchronously, for example, in multiple modules.

[0049] Other embodiments of the present invention will readily occur to those skilled in the art after considering the specification and practicing the invention herein. This application is intended to cover any variations, uses, or adaptations of the present invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, with the true scope and spirit of the invention being indicated by the claims.

[0050] It should be understood that the present invention is not limited to the exact construction described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof, which is limited only by the appended claims.

Claims

1. A method for estimating illumination parameters based on a lightweight neural network model, characterized in that: include: Perform image segmentation on the outdoor scene panoramic image to obtain the sky area image; Separate and combine the multiple color channels of the sky area image to obtain an RGB-L four-channel input image; Construct a feature extraction layer, which includes a first Ghost module, a fully decoupled fully connected attention module, an SE attention module, and a second Ghost module. The feature extraction layer is used to use the first Ghost module and the fully decoupled fully connected attention module to respectively extract the features of the four-channel input image input by the input layer, and multiply the two features to obtain a first feature, perform a depth separation convolution on the first feature to obtain a convolved second feature, use the SE attention module to compress and excite the second feature to obtain a channel weight, multiply the channel weight with the feature of the corresponding channel of the second feature to obtain a third feature, use the second Ghost module to perform dimensionality reduction processing on the third feature to obtain a fourth feature, and fuse the fourth feature with the four-channel input image to obtain the final feature; A lightweight illumination estimation network model is constructed based on the feature extraction layer; a dataset is constructed based on the four-channel input image and the illumination parameter labels in the image, the four-channel input image and the illumination parameter labels in the image in the dataset are normalized respectively to obtain a target dataset, and the lightweight illumination estimation network model is trained based on the target dataset to obtain a target network model; The normalized four-channel input image corresponding to the panoramic image of the outdoor scene to be estimated is input into the target network model to predict the lighting parameters.

2. The method for estimating illumination parameters based on a lightweight neural network model according to claim 1, wherein: The steps for segmenting the outdoor scene panoramic image to obtain the sky area image are as follows: Annotate the sky area in the collected outdoor scene panoramic images and construct a sky area segmentation dataset; Build a sky area segmentation model based on U-Net; The segmentation dataset is input into the sky region segmentation model to obtain a sky region image.

3. The method for estimating illumination parameters based on a lightweight neural network model according to claim 1, wherein: The steps for separating and combining the multiple color channels of the sky area image to obtain the RGB-L four-channel input image are as follows: Map the sky area image from RGB color space to LAB color space to obtain the target image; The L channel of the target image is combined with the RGB three channels of the sky area image to obtain an RGB-L four-channel input image.

4. The method for estimating illumination parameters based on a lightweight neural network model according to claim 1, wherein: Lighting parameters include: sun azimuth, sun altitude, sun color, sky color, and sky turbidity.

5. The method for estimating illumination parameters based on a lightweight neural network model according to claim 1, wherein: The normalized expression of the four-channel input image in the dataset is: Where, Represents the pixel value of a channel after normalization of the four-channel input image in the dataset; Represents the pixel value of a channel of the four-channel input image in the dataset, Represents the pixel mean of a channel of the four-channel input image in the dataset; Represents the pixel standard deviation of a channel of the four-channel input image in the dataset.

6. The method for estimating illumination parameters based on a lightweight neural network model according to claim 1, wherein: The normalized expression of the illumination parameter label in the four-channel input image in the dataset is: Where, Represents the label value after normalization of the illumination parameters in the four-channel input image in the dataset; Label values ​​of illumination parameters in the four-channel input images in the dataset; is the minimum label value, The maximum label value.

7. A light parameter estimation system based on a lightweight neural network model, characterized in that: include: An image processing module is used to segment the outdoor scene panoramic image to obtain a sky area image; Separate and combine the multiple color channels of the sky area image to obtain an RGB-L four-channel input image; A feature extraction module is used to construct a feature extraction layer. The feature extraction layer includes a first Ghost module, a fully decoupled fully connected attention module, an SE attention module, and a second Ghost module. The feature extraction layer is used to use the first Ghost module and the fully decoupled fully connected attention module to respectively extract the features of the four-channel input image input by the input layer, and multiply the two features to obtain a first feature. The first feature is subjected to a depth-separated convolution to obtain a convolved second feature. The second feature is compressed and excited by the SE attention module to obtain a channel weight. The channel weight is multiplied by the feature of the corresponding channel of the second feature to obtain a third feature. The third feature is subjected to dimensionality reduction processing by the second Ghost module to obtain a fourth feature. The fourth feature is fused with the four-channel input image to obtain a final feature. A lightweight illumination estimation network module is used to construct a lightweight illumination estimation network model based on the feature extraction layer; a dataset is constructed based on the four-channel input image and the illumination parameter labels in the image, the four-channel input image and the illumination parameter labels in the image in the dataset are normalized to obtain a target dataset, and the lightweight illumination estimation network model is trained based on the target dataset to obtain a target network model; The parameter prediction module is used to input the normalized four-channel input image corresponding to the panoramic image of the outdoor scene to be estimated into the target network model to predict the lighting parameters.

8. An electronic device, characterized in that: The method comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program implements the steps of the method according to any one of claims 1 to 6 when executed by the processor.

9. A storage medium, characterized in that: The method comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program implements the steps of the method according to any one of claims 1 to 6 when executed by the processor.

10. A program product comprising a computer program, characterized in that When the computer program is executed by the processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images