Method, device and equipment for automatically identifying white-base map in commodity map and medium
By training the UNet network for product image segmentation and grayscale value judgment, the problem of low efficiency and high false positive rate in white background image recognition in traditional methods is solved, achieving fast and accurate white background image recognition, which is suitable for product image processing on e-commerce platforms.
Patent Information
- Application Number
- CN202511291228.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-10
- Publication Date
- 2026-01-13
AI Technical Summary
Existing white background image recognition methods are inefficient and have a high false positive rate. They cannot accurately identify white background images in product images under complex backgrounds, and cannot meet the needs of e-commerce platforms for fast and accurate processing of a large number of product images.
Image segmentation is performed using a trained UNet network, combined with masking techniques and grayscale value judgment. The segmented image is then used for image matting and white area ratio calculation, and a threshold is set to determine whether the product image is a white background image.
It achieves rapid recognition of white background images, reduces the false judgment rate, and improves recognition efficiency and accuracy. It is applicable to various types of product images and meets the batch processing needs of e-commerce platforms.
Smart Images
Figure CN121330346A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field, in particular to a white background automatic identification method, device and equipment and medium in commodity picture. BACKGROUND
[0002] In the fields of commodity display, commodity information retrieval and commodity data statistics of e-commerce platforms, accurately identifying the white background in commodity pictures is of great significance. Currently, traditional white background identification mainly relies on manual judgment, which is not only inefficient but also greatly affected by subjective factors, and is prone to misjudgment.
[0003] Existing automatic identification methods either have low recognition accuracy and cannot accurately distinguish white backgrounds in complex backgrounds, or have high requirements for image preprocessing and feature extraction, and poor adaptability. For example, in some commodity pictures, the main color of the commodity is similar to the white background, or there are many white elements in the background, which interfere with the traditional method, making it difficult to accurately identify whether it is a white background, resulting in low reliability of the identification result, and failing to meet the needs of e-commerce platforms and other scenarios for fast and accurate processing of a large number of commodity pictures. SUMMARY
[0004] The technical problem to be solved by the present application is to provide a white background automatic identification method, device and equipment and medium in commodity picture, by training a specific segmentation model, combining image segmentation, region calculation and other technologies, to realize the automatic identification of white background in commodity picture, improve the identification efficiency and accuracy, meet the needs of e-commerce platform commodity management, data statistics and other diversified scenarios, reduce the labor cost, and improve the work quality.
[0005] In a first aspect, the present application provides a white background automatic identification method in commodity picture, comprising the following steps:
[0006] Step 1, train the UNet network to obtain a trained UNet network, input the original commodity picture into the trained UNet network, and output a binary segmentation result image, wherein the white area represents the commodity main body and the black area represents the background;
[0007] Step 2, according to the segmentation result image, perform cutout operation on the original commodity picture, and through mask technology, retain the original image area corresponding to the white area in the segmentation result, remove the black area, and obtain a commodity main body picture containing only the commodity main body;
[0008] Step 3, convert the commodity main body picture obtained after cutout from RGB color space to gray space, set the pixels with gray value greater than or equal to 240 as white pixels, count the number of white pixels in the area, and calculate the ratio of the number of white pixels to the total number of pixels in the area, i.e. the ratio of the white area after segmentation;
[0009] Step 4, for the original commodity image, the white proportion of four set regions of the original commodity is calculated respectively; the images of the four set regions are converted to the gray space, the number of white pixels with a gray value greater than or equal to 240 in each region is counted, and the ratio of the number of white pixels in each corner region to the total number of pixels in the set region is calculated respectively to obtain the white bottom proportion of the four set regions;
[0010] Step 5, if the four white bottom proportions all exceed the first set threshold, and the white region proportion after segmentation exceeds the second set threshold, the original commodity image is a white bottom image; otherwise, the original commodity image is a non-white bottom image.
[0011] In a second aspect, the present application provides a device for automatically identifying a white bottom image in a commodity image, comprising:
[0012] The network training module trains the UNet network to obtain a trained UNet network, and inputs the original commodity image into the trained UNet network to output a binary segmentation result image, wherein the white region represents the commodity main body and the black region represents the background.
[0013] The segmentation image module performs a cutout operation on the original commodity image according to the segmentation result image, and through a mask technology, the original image region corresponding to the white region in the segmentation result is retained and the black region is removed to obtain a commodity main body image containing only the commodity main body.
[0014] The main body region calculation module converts the commodity main body image obtained after the cutout from the RGB color space to the gray space, sets the pixels with a gray value greater than or equal to 240 as white pixels, counts the number of white pixels in the region, and calculates the ratio of the number of white pixels to the total number of pixels in the region, i.e. the white region proportion after segmentation.
[0015] The set region calculation module calculates the white proportion of four set regions of the original commodity image respectively; the images of the four set regions are converted to the gray space, the number of white pixels with a gray value greater than or equal to 240 in each region is counted, and the ratio of the number of white pixels in each corner region to the total number of pixels in the set region is calculated respectively to obtain the white bottom proportion of the four set regions.
[0016] The image white bottom identification module, if the four white bottom proportions all exceed the first set threshold, and the white region proportion after segmentation exceeds the second set threshold, the original commodity image is a white bottom image; otherwise, the original commodity image is a non-white bottom image.
[0017] In a third aspect, the present application provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method of the first aspect when executing the program.
[0018] In a fourth aspect, the present application provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the method of the first aspect.
[0019] The one or more technical solutions provided by the present application have at least the following technical effects or advantages:
[0020] 1. Efficiency
[0021] Compared with the traditional manual identification of white background pictures in commodity pictures, the present application uses a trained UNet network and automatic calculation steps to realize the rapid identification of white background pictures. The traditional manual identification of whether a commodity picture is a white background picture may take several seconds or even tens of seconds, while the processing time of a single image by the algorithm of the present application can be shortened to within 1 second, greatly improving the identification efficiency and meeting the needs of e-commerce platforms and other scenarios for batch processing of a large number of commodity pictures.
[0022] 2. Accuracy
[0023] The UNet network is trained by a large number of diversified training samples, which can accurately segment the main region in the commodity picture, laying a good foundation for subsequent white area ratio calculation. In the white area ratio calculation process, a scientific gray value judgment standard is adopted, combined with the calculation of the segmented region and the four corner regions, which can accurately judge whether the commodity picture is a white background picture, effectively reduce the misjudgment rate and improve the accuracy of identification.
[0024] 3. Adaptability
[0025] The algorithm of the present application is applicable to commodity pictures of various types, colors and backgrounds. Whether the commodity picture is of simple background or complex background, the white background picture in the commodity picture can be accurately identified. The user does not need to have professional image recognition knowledge. By inputting the commodity picture, the user can quickly obtain the identification result through the algorithm, meet the needs of commodity picture white background identification in different scenarios, and has strong universality and adaptability.
[0026] The above description is only a summary of the technical solutions of the present application. In order to more clearly understand the technical means of the present application, the content of the specification can be implemented, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific embodiments of the present application are described as follows. BRIEF DESCRIPTION OF DRAWINGS
[0027] The present application will be further described below with reference to the accompanying drawings and embodiments.
[0028] Figure 1 The flowchart in the method of the first embodiment of the present application is shown in the figure.
[0029] Figure 2This is a schematic diagram of the device in Embodiment 2 of the present invention. Detailed Implementation
[0030] The overall concept of the technical solution in this application is as follows:
[0031] UNet network: A semantic segmentation network based on deep learning with an encoder-decoder structure, which can effectively extract image features and is used for segmentation of the main body of product images in this invention.
[0032] The algorithm is mainly implemented as follows:
[0033] Step 1: UNet Network Training and Product Image Segmentation
[0034] Training Sample Collection: Collect a large number of images containing various products as training samples, approximately 10,000 images. These samples should cover different product types (such as clothing, electronic products, food, household goods, etc.), product colors (dark, light, colorful, etc.), background types (complex backgrounds, single backgrounds, backgrounds containing white elements, etc.), and the position and size of the products in the images, ensuring the diversity of the training samples and comprehensively covering various product image situations that may occur in practical applications.
[0035] Network Training: The acquired training sample images and their corresponding labeled images (accurately labeling the main product area) are input into the UNet network for training. During training, the cross-entropy loss function is used as the optimization objective, and the Adam algorithm is selected as the optimization algorithm. The initial learning rate is set to 0.0008, and after every 8 training epochs, the learning rate decays to 0.75 times the original value. The training epochs are set to 50, and the batch size is 32. During training, the network parameters are continuously adjusted so that the UNet network can accurately learn the features of the main product, thereby achieving accurate segmentation of the main body in the product image.
[0036] Product image subject segmentation: The product image to be identified is input into the trained UNet network, and the network will output a binary segmentation result image, in which the white area represents the product subject and the black area represents the background, thus achieving the separation of the product image subject from the background.
[0037] Step 2: Cut out the main part of the product image
[0038] Based on the binarized segmentation result image obtained in step one, the original product image is processed by masking. Using a masking technique, the original image region corresponding to the white area (product subject) in the segmentation result is preserved, while the black area (background) is removed, resulting in an image region containing only the product subject.
[0039] Step 3: Calculate the proportion of white in the segmented area
[0040] The extracted product image is converted from RGB color space to grayscale space for more accurate white pixel identification. Pixels with a grayscale value greater than or equal to 240 are defined as white pixels (this threshold can be fine-tuned to adapt to different lighting conditions). The number of white pixels in this area is counted, and the ratio of the number of white pixels to the total number of pixels in the area is calculated to obtain the percentage of white area after segmentation.
[0041] Step 4: Calculate the percentage of white in the four corners of the image
[0042] For the original product image, the white percentage is calculated for the four corner regions of the image. The size of each corner region is 5% of the image's width and height. Specifically, the top-left corner region is a rectangle starting from the top-left corner of the image, with a width of 5% of the image width and a height of 5% of the image height; the top-right corner region is a rectangle starting from the top-right corner of the image, extending to the left with a width of 5% of the image width and downwards with a height of 5% of the image height; the bottom-left corner region is a rectangle starting from the bottom-left corner of the image, extending to the right with a width of 5% of the image width and upwards with a height of 5% of the image height; and the bottom-right corner region is a rectangle starting from the bottom-right corner of the image, extending to the left with a width of 5% of the image width and upwards with a height of 5% of the image height.
[0043] Similarly, convert the images of these four corner regions to grayscale space, count the number of white pixels with a grayscale value greater than or equal to 240 in each region, and calculate the ratio of the number of white pixels in each corner region to the total number of pixels in that corner region to obtain the white background percentage of the four corners.
[0044] Step 5: White background image judgment
[0045] The proportion of the white area after segmentation obtained in step three and the proportion of the white background in the four corners obtained in step four are judged. If the proportion of the white background in all four corners exceeds 80% and the proportion of the white area after segmentation exceeds 80%, the product image is judged to be a white background image; otherwise, it is judged to be a non-white background image.
[0046] Example 1
[0047] like Figure 1 As shown, this embodiment provides a method for automatically recognizing white background images in product images, including the following steps:
[0048] Step 1: Train the UNet network to obtain a trained UNet network. Input the original product image into the trained UNet network and output a binary segmentation result image, where the white area represents the product body and the black area represents the background.
[0049] Step 2: Based on the segmentation result image, perform a cutout operation on the original product image. Using masking technology, retain the original image area corresponding to the white area in the segmentation result and remove the black area to obtain a product image containing only the product body.
[0050] Step 3: Convert the product image obtained after cutout from RGB color space to grayscale space, and set pixels with a grayscale value greater than or equal to 240 as white pixels; count the number of white pixels in this area, and calculate the ratio of the number of white pixels to the total number of pixels in this area, which gives the proportion of white area after segmentation.
[0051] Step 4: For the original product image, calculate the white percentage for each of the four designated areas of the original product; convert the images of these four designated areas to grayscale space, count the number of white pixels with a grayscale value greater than or equal to 240 in each area, and calculate the ratio of the number of white pixels in each corner area to the total number of pixels in that designated area to obtain the white background percentage of the four designated areas.
[0052] Step 5: If the proportion of white background in all four areas exceeds the first set threshold, and the proportion of white area after segmentation exceeds the second set threshold, then the original product image is a white background image; otherwise, the original product image is a non-white background image.
[0053] In this embodiment, preferably, step 1 specifically comprises:
[0054] Collect images containing various products as training samples, and label the main regions of the training samples to obtain labeled samples;
[0055] The collected training sample images and their corresponding labeled samples are input into the UNet network for training. During training, the cross-entropy loss function is used as the optimization objective, and the Adam algorithm is selected as the optimization algorithm. The initial learning rate is set to 0.0008, and the learning rate is reduced to 0.75 times the original value every 8 training epochs. The training epochs are set to 50, and the batch size is 32. During the training process, the network parameters are continuously adjusted. When the expected result is obtained, the training is stopped, and the trained UNet network is obtained.
[0056] The original product image to be identified is input into the trained UNet network, which outputs a binary segmentation result image, where the white area represents the product and the black area represents the background.
[0057] In this embodiment, preferably, step 4 specifically involves: for the original product image, calculating the white percentage of the four corner regions of the image; the size of each corner region is a set percentage of the image width and height; converting the images of these four corner regions to grayscale space, counting the number of white pixels with a grayscale value greater than or equal to 240 in each region, and calculating the ratio of the number of white pixels in each corner region to the total number of pixels in that corner region to obtain the white background percentage of the four corners.
[0058] Based on the same inventive concept, this application also provides an apparatus corresponding to the method in Embodiment 1, as detailed in Embodiment 2.
[0059] Example 2
[0060] like Figure 2 As shown, this embodiment provides an automatic white background image recognition device for product images, including:
[0061] The network training module trains the UNet network to obtain a trained UNet network. The original product image is input into the trained UNet network, and a binary segmentation result image is output, in which the white area represents the product body and the black area represents the background.
[0062] The image segmentation module performs image cutout on the original product image based on the segmentation result image. Using masking technology, it retains the original image area corresponding to the white area in the segmentation result and removes the black area to obtain a product image containing only the product body.
[0063] The main body area calculation module converts the product main image obtained after image cutout from the RGB color space to the grayscale space, sets pixels with a grayscale value greater than or equal to 240 as white pixels, counts the number of white pixels in this area, and calculates the ratio of the number of white pixels to the total number of pixels in this area, which gives the proportion of the white area after segmentation.
[0064] The setting area calculation module calculates the white ratio of four setting areas of the original product image; converts the images of these four setting areas to grayscale space, counts the number of white pixels with a grayscale value greater than or equal to 240 in each area, and calculates the ratio of the number of white pixels in each corner area to the total number of pixels in the setting area to obtain the white background ratio of the four setting areas.
[0065] If the proportion of white background in all four images exceeds a first set threshold, and the proportion of white area after segmentation exceeds a second set threshold, then the original product image is a white background image; otherwise, the original product image is a non-white background image.
[0066] In this embodiment, preferably, the network training module specifically comprises:
[0067] Collect images containing various products as training samples, and label the main regions of the training samples to obtain labeled samples;
[0068] The collected training sample images and their corresponding labeled samples are input into the UNet network for training. During training, the cross-entropy loss function is used as the optimization objective, and the Adam algorithm is selected as the optimization algorithm. The initial learning rate is set to 0.0008, and the learning rate is reduced to 0.75 times the original value every 8 training epochs. The training epochs are set to 50, and the batch size is 32. During the training process, the network parameters are continuously adjusted. When the expected result is obtained, the training is stopped, and the trained UNet network is obtained.
[0069] The original product image to be identified is input into the trained UNet network, which outputs a binary segmentation result image, where the white area represents the product and the black area represents the background.
[0070] In this embodiment, preferably, the set area calculation module specifically performs the following steps: for the original product image, calculate the white percentage of the four corner areas of the image; the size of each corner area is a set percentage of the image width and height; convert the images of these four corner areas to grayscale space, count the number of white pixels with a grayscale value greater than or equal to 240 in each area, and calculate the ratio of the number of white pixels in each corner area to the total number of pixels in that corner area to obtain the white background percentage of the four corners.
[0071] Since the apparatus described in Embodiment 2 of the present invention is an apparatus used to implement the method of Embodiment 1 of the present invention, those skilled in the art can understand the specific structure and variations of the apparatus based on the method described in Embodiment 1 of the present invention, and therefore will not be described again here. All apparatuses used in the method of Embodiment 1 of the present invention fall within the scope of protection of the present invention.
[0072] Based on the same inventive concept, this application provides an electronic device embodiment corresponding to Embodiment 1, as detailed in Embodiment 3.
[0073] Example 3
[0074] This embodiment provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it can implement any of the implementation methods in Embodiment 1.
[0075] Since the electronic device described in this embodiment is the device used to implement the method in Embodiment 1 of this application, those skilled in the art can understand the specific implementation method and various variations of the electronic device in this embodiment based on the method described in Embodiment 1 of this application. Therefore, how the electronic device implements the method in the embodiment of this application will not be described in detail here. Any device used by those skilled in the art to implement the method in the embodiment of this application falls within the scope of protection of this application.
[0076] Based on the same inventive concept, this application provides a storage medium corresponding to Embodiment 1, as detailed in Embodiment 4.
[0077] Example 4
[0078] This embodiment provides a computer-readable storage medium storing a computer program thereon. When the computer program is executed by a processor, it can implement any of the implementation methods in Embodiment 1.
[0079] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0080] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0081] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.
[0082] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0083] While specific embodiments of the present invention have been described above, those skilled in the art should understand that the specific embodiments described are merely illustrative and not intended to limit the scope of the present invention. Equivalent modifications and variations made by those skilled in the art in accordance with the spirit of the present invention should be covered within the scope of protection of the claims of the present invention.
Claims
1. A method for automatically recognizing white background images in product images, characterized in that: Includes the following steps: Step 1: Train the UNet network to obtain a trained UNet network. Input the original product image into the trained UNet network and output a binary segmentation result image, where the white area represents the product body and the black area represents the background. Step 2: Based on the segmentation result image, perform a cutout operation on the original product image. Using masking technology, retain the original image area corresponding to the white area in the segmentation result and remove the black area to obtain a product image containing only the product body. Step 3: Convert the product image obtained after cutout from RGB color space to grayscale space, and set pixels with a grayscale value greater than or equal to 240 as white pixels; count the number of white pixels in this area, and calculate the ratio of the number of white pixels to the total number of pixels in this area, which gives the proportion of white area after segmentation. Step 4: For the original product image, calculate the white percentage for each of the four designated areas of the original product. Convert the images of these four designated areas to grayscale space, count the number of white pixels with a grayscale value greater than or equal to 240 in each area, and calculate the ratio of the number of white pixels in each corner area to the total number of pixels in that designated area to obtain the white background ratio of the four designated areas. Step 5: If the proportion of white background in all four areas exceeds the first set threshold, and the proportion of white area after segmentation exceeds the second set threshold, then the original product image is a white background image; otherwise, the original product image is a non-white background image.
2. The method for automatic recognition of white background images in product images according to claim 1, characterized in that: Step 1 specifically involves: Collect images containing various products as training samples, and label the main regions of the training samples to obtain labeled samples; The collected training sample images and their corresponding labeled samples are input into the UNet network for training; During training, the cross-entropy loss function was used as the optimization objective, and the Adam algorithm was selected as the optimization algorithm. The initial learning rate was set to 0.0008, and after every 8 training epochs, the learning rate was reduced to 0.75 times the original value. The training epochs were set to 50, and the batch size was 32. During the training process, the network parameters were continuously adjusted, and training was stopped when the expected result was obtained, resulting in the trained UNet network. The original product image to be identified is input into the trained UNet network, which outputs a binary segmentation result image, where the white area represents the product and the black area represents the background.
3. The method for automatic recognition of white background images in product images according to claim 1, characterized in that: Step 4 specifically involves: for the original product image, calculating the white percentage of the four corner regions of the image; the size of each corner region is a set percentage of the image's width and height; converting the images of these four corner regions to grayscale space, counting the number of white pixels with a grayscale value greater than or equal to 240 in each region, and calculating the ratio of the number of white pixels in each corner region to the total number of pixels in that corner region to obtain the white background percentage of the four corners.
4. An automatic white background image recognition device for product images, characterized in that: include: The network training module trains the UNet network to obtain a trained UNet network. The original product image is input into the trained UNet network, and a binary segmentation result image is output, in which the white area represents the product body and the black area represents the background. The image segmentation module performs image cutout on the original product image based on the segmentation result image. Using masking technology, it retains the original image area corresponding to the white area in the segmentation result and removes the black area to obtain a product image containing only the product body. The main body area calculation module converts the product main image obtained after image cutout from the RGB color space to the grayscale space, sets pixels with a grayscale value greater than or equal to 240 as white pixels, counts the number of white pixels in this area, and calculates the ratio of the number of white pixels to the total number of pixels in this area, which gives the proportion of the white area after segmentation. The defined area calculation module calculates the white percentage for each of the four defined areas of the original product image. Convert the images of these four designated areas to grayscale space, count the number of white pixels with a grayscale value greater than or equal to 240 in each area, and calculate the ratio of the number of white pixels in each corner area to the total number of pixels in that designated area to obtain the white background ratio of the four designated areas. If the proportion of white background in all four images exceeds a first set threshold, and the proportion of white area after segmentation exceeds a second set threshold, then the original product image is a white background image. No, then the original product image is a non-white background image.
5. The automatic white background image recognition device for product images according to claim 4, characterized in that: The network training module is specifically as follows: Collect images containing various products as training samples, and label the main regions of the training samples to obtain labeled samples; The collected training sample images and their corresponding labeled samples are input into the UNet network for training; During training, the cross-entropy loss function was used as the optimization objective, and the Adam algorithm was selected as the optimization algorithm. The initial learning rate was set to 0.0008, and after every 8 training epochs, the learning rate was reduced to 0.75 times the original value. The training epochs were set to 50, and the batch size was 32. During the training process, the network parameters were continuously adjusted, and training was stopped when the expected result was obtained, resulting in the trained UNet network. The original product image to be identified is input into the trained UNet network, which outputs a binary segmentation result image, where the white area represents the product and the black area represents the background.
6. The automatic white background image recognition device for product images according to claim 4, characterized in that: The specific steps of the set area calculation module are as follows: For the original product image, the white ratio of the four corner areas of the image is calculated; the size of each corner area is a set percentage of the image width and height; the images of these four corner areas are converted to grayscale space, the number of white pixels with a grayscale value greater than or equal to 240 in each area is counted, and the ratio of the number of white pixels in each corner area to the total number of pixels in that corner area is calculated to obtain the white background ratio of the four corners.
7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method as described in any one of claims 1 to 3.
8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1 to 3.