Obstruction detection method and device based on hash similarity and feature similarity

Abstract

The present application relates to a kind of based on hash similarity and feature similarity's occlusion detection method, device.Based on hash similarity and feature similarity's occlusion detection method, by the hash similarity and Embd feature similarity of template update time to the threshold judgment of the hash similarity and Embd feature similarity of template update time to the threshold judgment of the hash similarity and Embd feature similarity of template update time to the threshold judgment of the hash similarity and Embd feature similarity of template update time to the threshold judgment of the hash similarity and Embd feature similarity of template update time to the threshold judgment of the hash similarity and Embd feature similarity of template update time to the threshold judgment of the hash similarity and Embd feature similarity of template update time to the threshold judgment of the hash similarity and Embd feature similarity of template update time to the threshold judgment of the hash similarity and Embd feature similarity of template update time to the threshold judgment of the hash similarity and Embd feature similarity of template

Description

Technical Field

[0001] This invention relates to camera occlusion detection, and in particular to an occlusion detection method and apparatus based on hash similarity and feature similarity. Background Technology

[0002] With urban development, the number of elevators has increased year by year. While providing convenience, elevator safety has become increasingly important, requiring the detection and prohibition of dangerous behaviors. For example, in recent years, there have been numerous videos showing electric bicycles being illegally brought into elevators, resulting in fires and injuries. Currently, it is explicitly prohibited for users to push electric bicycles into elevators. Many elevators now have electric bicycle detection and control functions. If the camera inside the elevator detects a passenger pushing an electric bicycle, it will issue an alarm. However, some passengers obstruct the camera with umbrellas, hats, or other items after entering the elevator, preventing the camera from capturing normal images and affecting the electric bicycle control function. This hinders the detection and alerting of electric bicycles and other prohibited items and behaviors, posing a safety hazard. Such safety hazards concerning personal safety deserve serious attention. Therefore, timely detection of camera obstructions in elevators is crucial.

[0003] With the development of intelligent technology and neural network algorithms, image change detection methods have become increasingly mature. Therefore, appropriate methods can be selected to measure image features and the degree of image change, thereby determining whether the camera is obstructed in a short period of time and providing timely warnings.

[0004] In existing technologies, 1) there is a method for judging camera occlusion based on image color information. This involves acquiring the color information of each pixel unit in the image; counting the number of color types in the image based on the color information of each pixel unit; if the number of color types in the image is less than an occlusion threshold, then the camera capturing the image is determined to be occluded; if the number of color types in the image is greater than or equal to the occlusion threshold, then the camera capturing the image is determined not to be occluded. This method, using color information, is easily affected by factors such as lighting and environment, leading to false detections. Furthermore, the color information is relatively simple, requiring manual threshold adjustment, and is suitable for single scenarios. 2) Using the area ratio of connected regions to determine whether it exceeds a set threshold for camera occlusion detection. This method often results in false alarms due to large objects, leading to inaccurate detection results. 3) Using depth cameras or binocular cameras. These cameras can obtain corresponding position information to determine whether occlusion has occurred. However, this method has certain hardware requirements and is not applicable to existing cameras inside elevator cars. Summary of the Invention

[0005] The purpose and technical effects of this invention are as follows: To solve the above problems, this invention uses an occlusion detection method and device based on hash similarity and feature similarity to detect occlusion behavior of cameras.

[0006] An occlusion detection method based on hash similarity and feature similarity includes the following steps:

[0007] S1. Collect image data inside the elevator car using a camera;

[0008] S2. Process the image data. When there is a person in the current frame of the image data, add it to the image sequence to be detected, and finally obtain a sequence of n images to be detected arranged in time sequence.

[0009] S3. Take the first frame in the sequence of images to be detected as the initial template image. Then, according to the time series, calculate the difference between the acquisition time of each frame in the sequence of images to be detected and the current system time. If the difference is less than the template update interval threshold, perform the template update operation. The template update operation includes storing the current image as the template image and recording the current system time as the template update time. If the difference is greater than the template update interval threshold, do not perform the template image update operation.

[0010] S4. Calculate the hash value of each frame of the image to be detected and the template image in sequence, calculate the Hamming distance based on the hash value, and then determine the hash similarity between the template image and the image to be detected.

[0011] S5. Using a convolutional neural network, extract the Embd feature vectors of the template image and the image to be detected, and calculate the similarity difference between the Embd feature vectors of the template image and the Embd feature vectors of the image to be detected to obtain the similarity of the Embd features between the template image and the image to be detected.

[0012] S6. Compare the hash similarity with the hash similarity threshold, and compare the Embd feature similarity with the Embd feature similarity threshold. If both are greater than the threshold, it is determined that the similarity between the image to be detected and the template image is very large, and it is determined that there may be occlusion; otherwise, it is determined that the difference between the image to be detected and the template image is not large, and it is determined that there is no occlusion, that is, no occlusion occurred during the time period of camera acquisition.

[0013] S7. If it is determined that there may be occlusion, the difference between the template update time and the current system time is calculated. If the absolute value of the difference is less than the template update interval threshold, it means that the current image to be detected is occluded, and occlusion is determined, that is, occlusion occurs within the time period of camera acquisition.

[0014] S8. If the absolute value of the difference is greater than the template update interval threshold, the current image to be detected is used as the template image, and the remaining images in the image sequence to be detected continue to be occlusion detected according to steps S4-S8.

[0015] An occlusion detection device based on hash similarity and feature similarity, comprising:

[0016] The first acquisition module is used to acquire image data inside the elevator car via a camera.

[0017] The first processing module is used to process image data. When there is a person in the current frame of the image data, it is added to the image sequence to be detected, and finally a sequence of n images to be detected arranged in time sequence is obtained.

[0018] The second processing module is used to take the first frame in the sequence of images to be detected as the initial template image, and then calculate the difference between the acquisition time of each frame in the sequence and the current system time according to the time sequence. If the absolute value of the difference is less than the template update interval threshold, a template update operation is performed. The template update operation includes storing the current image as the template image and recording the current system time as the template update time. If the absolute value of the difference is greater than the template update interval threshold, no template image update operation is performed.

[0019] The third processing module is used to calculate the hash value of each frame of the image to be detected and the template image in sequence, calculate the Hamming distance based on the hash value, and then determine the hash similarity between the template image and the image to be detected.

[0020] The fourth processing module extracts the Embd feature vectors of the template image and the image to be detected through a convolutional neural network, and calculates the similarity difference between the Embd feature vectors of the template image and the Embd feature vectors of the image to be detected to obtain the similarity of the Embd features between the template image and the image to be detected.

[0021] The first judgment module is used to compare the hash similarity with the hash similarity threshold and the Embd feature similarity with the Embd feature similarity threshold. If both are greater than the threshold, it is determined that the similarity between the image to be detected and the template image is very large, and it is determined that there may be occlusion; otherwise, it is determined that the difference between the image to be detected and the template image is not large, and it is determined that there is no occlusion, that is, no occlusion occurred during the time period of camera acquisition.

[0022] The second judgment module is used to calculate the difference between the template update time and the current system time when it is determined that there may be occlusion. If the absolute value of the difference is less than the template update interval threshold, it means that the current image to be detected is occluded, and occlusion is judged to have occurred, that is, occlusion occurred within the time period of camera acquisition. If the absolute value of the difference is greater than the template update interval threshold, the current image to be detected is used as the template image, and the remaining images in the image sequence to be detected continue to be occluded according to steps S4-S8.

[0023] An electronic device, comprising:

[0024] One or more processors;

[0025] Memory, used to store one or more programs;

[0026] When the one or more programs are executed by the one or more processors, the one or more processors perform the methods described above.

[0027] A computer-readable storage medium having computer instructions stored thereon, which, when executed by a processor, implement the steps of the above-described method.

[0028] The beneficial effects of this invention are as follows:

[0029] This application uses a threshold to determine the occlusion detection result by comparing the hash similarity and Embd feature similarity between the image to be detected and the template image, as well as the template update time. The detection is fast and accurate. Attached Figure Description

[0030] Figure 1 This is a flowchart of the method of the present invention;

[0031] Figure 2 This is a schematic diagram of the device module of the present invention;

[0032] The module comprises a first acquisition module 1, a first processing module 2, a second processing module 3, a third processing module 4, a fourth processing module 5, a first judgment module 6, and a second judgment module 7. Detailed Implementation

[0033] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0034] This application uses a threshold to determine the occlusion detection result by comparing the hash similarity and Embd feature similarity between the image to be detected and the template image, as well as the template update time. The detection is fast and accurate.

[0035] like Figure 1 A camera occlusion detection method based on hash similarity and feature similarity includes the following steps:

[0036] S1. Collect image data inside the elevator car using a camera;

[0037] S2. Process the image data. When there is a person in the current frame of the image data, add it to the image sequence to be detected, and finally obtain a sequence of n images to be detected arranged in time sequence; n = template update interval threshold * number of images that the algorithm can process per second. Specifically, using the HiSilicon Hi35xx computing chip, 10 images can be processed per second. If the template update threshold is set to 60 seconds, then the value of n is 600.

[0038] S3. Take the first frame in the sequence of images to be detected as the initial template image. Then, according to the time series, calculate the difference between the acquisition time of each frame in the sequence of images to be detected and the current system time. If the difference is less than the template update interval threshold, perform the template update operation. The template update operation includes storing the current image as the template image and recording the current system time as the template update time. If the difference is greater than the template update interval threshold, do not perform the template image update operation.

[0039] Specifically, when there are 600 images in the image sequence to be detected, the template is dynamically updated starting from the first frame, the template update time is recorded, and then the second frame image is used to compare with the template to determine whether occlusion has occurred, until the last image.

[0040] Template update interval threshold: This value ensures that occlusion detection is only effective within the update interval threshold time after someone is present. Based on actual testing, it is generally set to 60~120 seconds depending on the detection sensitivity.

[0041] S4. Calculate the hash value of each frame of the image to be detected and the template image in sequence, calculate the Hamming distance based on the hash value, and then determine the hash similarity between the template image and the image to be detected.

[0042] Specifically, if the current template is the first frame image, then the hash value is calculated for the first frame image and the second frame image, and the Hamming distance between the two images is calculated. The Hamming distance value is used as the hash similarity between the two images.

[0043] Hash value calculation: Scale the image to 10x10 pixels, convert the scaled image to a 64-level grayscale image, and calculate the average grayscale value of each of the 10x10 pixels. Compare the grayscale value of each pixel in the scaled image with the average grayscale value; set the pixel value greater than the average value to 1, and the pixel value less than the average value to 0, resulting in a 10x10 binary matrix. Combine each row to obtain a 100-bit image hash value.

[0044] Hamming distance calculation: Compare each bit of the hash value and count the number of unequal values, which is the Hamming distance.

[0045] Hash similarity calculation: i.e. Hamming distance value.

[0046] S5. Using a convolutional neural network, extract the Embd feature vectors of the template image and the image to be detected, and calculate the similarity difference between the Embd feature vectors of the template image and the Embd feature vectors of the image to be detected to obtain the similarity of the Embd features between the template image and the image to be detected.

[0047] The process of extracting Embed feature vectors is as follows:

[0048] 1) The input information is an image pixel matrix of 224*224*3. First, it is calculated by a convolutional layer with a kernel size of 3*3*64 and a stride of 1. Then, it is calculated by a convolutional layer with a kernel size of 3*3*64 and a stride of 1. The resulting data dimension is 224*224*64.

[0049] 2) The data with dimension 224*224*64 is processed by a pooling layer of size 2*2 with a stride of 2 to obtain a data dimension of 112*112*128; then processed by two convolutional layers of the same size 3*3*128 with a stride of 1 to obtain a data dimension of 112*112*128.

[0050] 3) The data with dimension 112*112*128 is processed by a pooling layer of size 2*2 with a stride of 2 to obtain a data dimension of 56*56*256; then processed by three convolutional layers of the same size 3*3*256 with a stride of 1 to obtain a data dimension of 56*56*256.

[0051] 4) The data with dimension 56*56*256 is processed through a pooling layer of size 2*2 with a stride of 2 to obtain a data dimension of 28*28*512; then it is processed through three convolutional layers of the same size 3*3*512 with a stride of 1 to obtain a data dimension of 28*28*512.

[0052] 5) The data with dimension 28*28*512 is processed by a pooling layer of size 2*2 with a stride of 2 to obtain a data dimension of 14*14*512; then processed by three convolutional layers of the same size 3*3*512 with a stride of 1 to obtain a data dimension of 14*14*512.

[0053] 6) The data with dimension 14*14*512 is processed through a pooling layer of size 2*2 with a stride of 2 to obtain a data dimension of 7*7*512; then it is processed through three convolutional layers of the same size 3*3*512 with a stride of 1 to obtain a data dimension of 7*7*512.

[0054] 7) Data with dimension 7*7*512 is passed through two fully connected layers of dimension 1*1*4096 to obtain data with dimension 1*1*4096.

[0055] 8) Pass the data of dimension 1*1*4096 through a fully connected layer of dimension 1*1*1000 to obtain data of dimension 1*1*1000. This data is used as the Embd feature vector of the image.

[0056] Similarity difference calculation for Embd feature vectors: Calculate the difference between each corresponding value of the Embd feature vector of the template image and the Embd feature vector of the image to be detected, sum the squares of all differences to obtain the sum of squared differences, take the square root of the sum of squared differences, and then multiply by 100 to obtain the Embd feature similarity difference value.

[0057] Embd feature similarity calculation: that is, the difference value of Embd feature similarity.

[0058] S6. Compare the hash similarity with the hash similarity threshold, and compare the Embd feature similarity with the Embd feature similarity threshold. If both are greater than the threshold, it is determined that the similarity between the detected image and the template image is very large, and it is determined that there may be occlusion; otherwise, it is determined that the difference between the detected image and the template image is not large, and it is determined that there is no occlusion.

[0059] Specifically, based on calculations in S4 and S5, the hash similarity value h and feature similarity value f of the first and second frame images are obtained. The hash similarity threshold is 65, and the Embd feature similarity threshold is 70. h is compared with 65, and f is compared with 70. If both exceed the threshold, the second frame image is determined to be occluded; otherwise, the second frame image is determined not to be occluded. The first frame image is removed from the image sequence to be detected, and operations S3-S7 are repeated starting from the second frame image.

[0060] Hash similarity threshold: Based on actual performance testing, the default value of 65 yields the best results. A smaller value will result in less variation in image content being identified as occlusion, making occlusion detection more sensitive and leading to false positives; conversely, a larger value will result in less sensitive occlusion detection, potentially leading to missed detections.

[0061] Embed feature similarity threshold: Based on actual performance testing, the default value of 70 yields the best results. A smaller value will result in less variation in image content being identified as occlusion, making occlusion detection more sensitive but potentially leading to false positives; conversely, a larger value will result in less sensitive occlusion detection, potentially leading to missed detections.

[0062] S7. If it is determined that there may be occlusion, the difference between the template update time and the current system time is calculated. If the absolute value of the difference is less than the template update interval threshold, it means that the current image to be detected is occluded, and occlusion is determined, that is, occlusion occurs within the time period of camera acquisition.

[0063] S8. If the absolute value of the difference is greater than the template update interval threshold, the current image to be detected is used as the template image, and the remaining images in the image sequence to be detected continue to be occlusion detected according to steps S4-S8.

[0064] like Figure 2 A camera occlusion detection device based on hash similarity and feature similarity, comprising:

[0065] The first acquisition module 1 is used to acquire image data inside the elevator car through a camera;

[0066] The first processing module 2 is used to process image data. When there is a person in the current frame of the image data, it is added to the image sequence to be detected, and finally a sequence of n images to be detected arranged in time sequence is obtained.

[0067] The second processing module 3 is used to take the first frame in the sequence of images to be detected as the initial template image, and then calculate the difference between the acquisition time of each frame in the sequence of images to be detected and the current system time according to the time sequence. If the absolute value of the difference is less than the template update interval threshold, a template update operation is performed. The template update operation includes storing the current image as the template image and recording the current system time as the template update time. If the absolute value of the difference is greater than the template update interval threshold, no template image update operation is performed.

[0068] The third processing module 4 is used to calculate the hash value of each frame of the image to be detected and the template image in sequence, calculate the Hamming distance based on the hash value, and then determine the hash similarity between the template image and the image to be detected.

[0069] The fourth processing module 5 extracts the Embd feature vectors of the template image and the image to be detected through a convolutional neural network, and calculates the similarity difference between the Embd feature vectors of the template image and the Embd feature vectors of the image to be detected to obtain the similarity of the Embd features between the template image and the image to be detected.

[0070] The first judgment module 6 is used to compare the hash similarity with the hash similarity threshold and the Embd feature similarity with the Embd feature similarity threshold. If both are greater than the threshold, it is determined that the similarity between the image to be detected and the template image is very large, and it is determined that there may be occlusion; otherwise, it is determined that the difference between the image to be detected and the template image is not large, and it is determined that there is no occlusion, that is, no occlusion occurred during the time period of camera acquisition.

[0071] The second judgment module 7 is used to calculate the difference between the template update time and the current system time when it is determined that there may be occlusion. If the absolute value of the difference is less than the template update interval threshold, it means that the current image to be detected is occluded, and the occlusion is determined to have occurred within the time period of the camera acquisition. If the absolute value of the difference is greater than the template update interval threshold, the current image to be detected is used as the template image, and the remaining images in the image sequence to be detected continue to be occluded according to steps S4-S8.

[0072] An electronic device, comprising:

[0073] One or more processors;

[0074] Memory, used to store one or more programs;

[0075] When the one or more programs are executed by the one or more processors, the one or more processors perform the methods described above.

[0076] In the embodiments provided in this application, it should be understood that the disclosed methods and systems can also be implemented in other ways. The method and system embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of methods and systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0077] In addition, the functional modules in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0078] On the other hand, a computer-readable storage medium stores computer instructions thereon, which, when executed by a processor, implement the steps of the above-described method. When the computer program is executed by the processor, it implements the method as described in any of the first aspects above. If the function is implemented as a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0079] This application achieves fast and accurate occlusion detection by thresholding hash values ​​and features in images. For example, in the detection of electric scooters entering elevator cars, if a passenger pushes an electric scooter into the elevator and triggers an alarm, the scooter may be obstructed by the camera, preventing detection and allowing the passenger to evade the elevator control. Therefore, it is necessary to detect camera occlusion and promptly alert the system when the camera is obstructed. This invention detects camera occlusion by acquiring a single image, eliminating the need for video or manual filtering, saving time and effort, and providing timely and effective feedback on camera occlusion status. The purpose of this invention is to promptly and effectively detect whether cameras inside the elevator car are obstructed, thereby preventing certain prohibited passenger behaviors from going undetected and improving the detection effect of safe elevator use and the internal status of the elevator car.

[0080] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. An occlusion detection method based on hash similarity and feature similarity, characterized in that, Includes the following steps: S1. Collect image data inside the elevator car using a camera; S2. Process the image data. When there is a person in the current frame of the image data, add it to the image sequence to be detected, and finally obtain a sequence of n images to be detected arranged in time sequence. S3. Take the first frame in the sequence of images to be detected as the initial template image, and then take the difference between the acquisition time of each frame in the sequence of images to be detected and the current system time according to the time series. If the difference is less than the template update interval threshold, then perform the template update operation. The template update operation includes storing the current image as the template image and recording the current system time as the template update time; If the difference is greater than the template update interval threshold, the template image update operation will not be performed; S4. Calculate the hash value of each frame of the image to be detected and the template image in sequence, calculate the Hamming distance based on the hash value, and then determine the hash similarity between the template image and the image to be detected. S5. Using a convolutional neural network, extract the Embd feature vectors of the template image and the image to be detected, and calculate the similarity difference between the Embd feature vectors of the template image and the Embd feature vectors of the image to be detected to obtain the similarity of the Embd features between the template image and the image to be detected. S6. Compare the hash similarity with the hash similarity threshold, and compare the Embd feature similarity with the Embd feature similarity threshold. If both are greater than the threshold, it is determined that the similarity between the image to be detected and the template image is very large, and it is determined that there may be occlusion; otherwise, it is determined that the difference between the image to be detected and the template image is not large, and it is determined that there is no occlusion, that is, no occlusion occurred during the time period of camera acquisition. S7. If it is determined that there may be occlusion, the difference between the template update time and the current system time is calculated. If the absolute value of the difference is less than the template update interval threshold, it means that the current image to be detected is occluded, and occlusion is determined, that is, occlusion occurs within the time period of camera acquisition. S8. If the absolute value of the difference is greater than the template update interval threshold, the current image to be detected is used as the template image, and the remaining images in the image sequence to be detected continue to be occlusion detected according to steps S4-S8.

2. An occlusion detection device based on hash similarity and Embd feature similarity, characterized in that, include: The first acquisition module is used to acquire image data inside the elevator car via a camera. The first processing module is used to process image data. When there is a person in the current frame of the image data, it is added to the image sequence to be detected, and finally a sequence of n images to be detected arranged in time sequence is obtained. The second processing module is used to take the first frame in the sequence of images to be detected as the initial template image, and then calculate the difference between the acquisition time of each frame in the sequence of images to be detected and the current system time according to the time series. If the absolute value of the difference is less than the template update interval threshold, the template update operation is performed. The template update operation includes storing the current image as the template image and recording the current system time as the template update time. If the absolute value of the difference is greater than the template update interval threshold, the template image update operation will not be performed; The third processing module is used to calculate the hash value of each frame of the image to be detected and the template image in sequence, calculate the Hamming distance based on the hash value, and then determine the hash similarity between the template image and the image to be detected. The fourth processing module extracts the Embd feature vectors of the template image and the image to be detected through a convolutional neural network, and calculates the similarity difference between the Embd feature vectors of the template image and the Embd feature vectors of the image to be detected to obtain the similarity of the Embd features between the template image and the image to be detected. The first judgment module is used to compare the hash similarity with the hash similarity threshold and the Embd feature similarity with the Embd feature similarity threshold. If both are greater than the threshold, it is determined that the similarity between the image to be detected and the template image is very large, and it is determined that there may be occlusion; otherwise, it is determined that the difference between the image to be detected and the template image is not large, and it is determined that there is no occlusion, that is, no occlusion occurred during the time period of camera acquisition. The second judgment module is used to calculate the difference between the template update time and the current system time when it is judged that there may be occlusion. If the absolute value of the difference is less than the template update interval threshold, it means that the current image to be detected is occluded, and the occlusion is judged to have occurred, that is, the occlusion occurred within the time period of the camera acquisition. If the absolute value of the difference is greater than the template update interval threshold, the current image to be detected is used as the template image, and the remaining images in the image sequence to be detected continue to be occlusion detected according to steps S4-S8.

3. An electronic device, characterized in that, include: One or more processors; Memory, used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in claim 1.

4. A computer-readable storage medium storing computer instructions thereon, characterized in that, When executed by the processor, this instruction implements the steps of the method as described in claim 1.

Images