Human-computer interaction method and system based on AI big model

Abstract

The present invention belongs to the field of human-computer interaction technology, and provides a human-computer interaction method and system based on an AI large model. The method includes: receiving interactive information input by a user, performing semantic analysis on the interactive information, and obtaining a user question set; performing a limited range assessment on each user question in the user question set, performing a safety analysis on user questions that exceed the limited range based on an accident case library, and obtaining a question safety value; for user questions whose question safety values ​​are higher than a safety threshold, obtaining relevant knowledge through the network, generating answer content based on the obtained relevant knowledge, and outputting the answer content to the user. The present invention can identify the user's out-of-range consultation situation, and perform a safety analysis on the corresponding question. Only when the safety standard is met can the corresponding answer content be obtained through the external network, thereby providing convenient interaction for users while ensuring the safety of production.

Description

Technical Field

[0001] The present invention relates to the field of human-computer interaction technology, and in particular to a human-computer interaction method and system based on an AI big model. Background Art

[0002] In order to improve production efficiency, many factories have deployed intelligent robots in production areas. These intelligent robots have built-in AI big models. By pre-training the AI ​​big models and embedding relevant industry knowledge, the intelligent robots can answer relevant difficult questions raised by staff, which is conducive to improving production efficiency. However, staff may ask intelligent robots questions beyond the scope of their work. For example, before being put into use, intelligent robots have built-in maintenance and repair knowledge about specific types of equipment, but the staff interacts with equipment operation information. This is an out-of-scope consultation, and it is difficult for the intelligent robot to provide a corresponding answer.

[0003] The above-mentioned out-of-scope consultation situation results in a poor user experience of the intelligent robot and cannot meet actual production needs. If the intelligent robot is pre-configured with a full system of industry knowledge, the implementation cost of the intelligent robot will become higher, and it will increase the difficulty of the intelligent robot in screening appropriate answers, thereby reducing the interaction efficiency.

[0004] Therefore, how to enable intelligent robots to provide quick and accurate answers when staff ask questions beyond the scope of their inquiry is a technical problem that urgently needs to be solved. Summary of the invention

[0005] In this regard, the present invention provides a human-computer interaction method, system, electronic device, computer storage medium and computer program product based on an AI big model to solve at least one of the above-mentioned technical problems.

[0006] The present invention provides a human-computer interaction method based on an AI big model, which is applied to an intelligent robot and is characterized in that it includes the following method steps: receiving interaction information input by a user, performing semantic analysis on the interaction information, and obtaining a user question set; wherein the user question set includes a number of user questions obtained by the semantic analysis; performing a limited range assessment on each of the user questions in the user question set, performing a safety analysis on the user questions that exceed the limited range based on an accident case library, and obtaining a problem safety value; wherein the accident case library is embedded in a storage device of the intelligent robot; for the user questions whose problem safety values ​​are higher than a safety threshold, obtaining relevant knowledge through a network, generating answer content based on the obtained relevant knowledge, and outputting the answer content to the user.

[0007] As an example, receiving interaction information input by a user includes: obtaining the user's identity information, retrieving the user's historical interaction records based on the identity information, and obtaining the user's average single interaction duration based on the historical interaction records; receiving the interaction information input by the user within the average single interaction duration, and if the user is still inputting when the average single interaction duration is reached, extending the average single interaction duration by a set percentage of the average single interaction duration, until an upper limit on the number of extensions is reached.

[0008] As an example, semantic analysis is performed on the interactive information to obtain a set of user questions, including: converting the interactive information into text segments, preprocessing the text segments, including checking for spelling and typos, and correcting errors; using a syntactic analysis technique combined with context to perform word segmentation, stem extraction, and sentence restoration on the preprocessed text segments to obtain a number of the user questions that constitute the user question set.

[0009] As an example, a limited range evaluation is performed on each of the user questions in the user question set, including: for each of the user questions in the user question set, a trial answer process is performed on each of the user questions based on a preset knowledge set to obtain a number of alternative answer contents, and a first quantity of each of the alternative answer contents is calculated; wherein, the preset knowledge set is embedded in the storage device of the intelligent robot; if the first quantity is greater than a quantity threshold, it is determined that the user question exceeds the limited range; otherwise, it is determined that the user question does not exceed the limited range; wherein, the quantity threshold is determined in the following manner: the user question is similarly matched with the interaction record data to obtain a second quantity of similar historical user questions, and the quantity threshold is calculated based on the second quantity.

[0010] As an example, a safety analysis is performed on the user questions that exceed the specified range based on the accident case library to obtain a problem safety value, including: performing semantic analysis on each of the user questions in the user question set to obtain the most relevant topic object; wherein the topic object refers to a certain type of equipment in the production area where the intelligent robot is deployed; extracting each accident case information related to the topic object from the accident case library, the accident case information including equipment type and accident cause, and the accident cause including erroneous operation information; calculating the positive correlation level between the user questions that exceed the specified range and each of the accident causes, and obtaining the problem safety value by matching the maximum positive correlation level with a preset comparison table.

[0011] As an example, the positive correlation level between the user problems beyond a limited range and each of the accident causes is calculated by an analysis model based on random forest and correlation coefficient.

[0012] The present invention also provides a human-computer interaction system based on an AI big model, which is applied to an intelligent robot, and the system includes a receiving module, an evaluation and safety analysis module, and a network acquisition module; the receiving module is used to receive interaction information input by a user, perform semantic analysis on the interaction information, and obtain a user question set; wherein the user question set includes several user questions obtained by semantic analysis; the evaluation and safety analysis module is used to perform a limited range evaluation on each user question in the user question set, perform a safety analysis on the user questions that exceed the limited range based on an accident case library, and obtain a problem safety value; wherein the accident case library is embedded in the storage device of the intelligent robot; the network acquisition module is used to obtain relevant knowledge through the network for the user questions whose problem safety values ​​are higher than the safety threshold, generate answer content based on the acquired relevant knowledge, and output the answer content to the user.

[0013] The present invention also provides an electronic device, which includes: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the computer program implements any of the methods described above when executed by the processor.

[0014] The present invention also provides a computer storage medium storing a computer program executable by a processor to implement any of the methods described above.

[0015] The present invention also provides a computer program product, which includes a computer program that can be executed by a processor to implement any of the methods described above.

[0016] The beneficial effect of the present invention is that the present invention can identify the user's out-of-range consultation situation and perform security analysis on the corresponding questions, and obtain the corresponding answer content through the external network only when the security standard is met, thereby providing convenient interaction for users while ensuring production safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 It is a flowchart of a human-computer interaction method based on an AI big model disclosed in an embodiment of the present invention.

[0019] Figure 2 It is a schematic diagram of a process for determining a problem safety value disclosed in an embodiment of the present invention.

[0020] Figure 3 It is a structural schematic diagram of a human-computer interaction system based on an AI big model disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0021] The following is a description of the implementation of the present application by specific specific embodiments. People familiar with the technology can easily understand other advantages and effects of the present application from the contents disclosed in this specification. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0022] In addition, the technical features involved in the different embodiments of the present application described below can be combined with each other as long as they do not conflict with each other.

[0023] like Figure 1 As shown, an embodiment of the present invention discloses a human-computer interaction method based on an AI big model, which is applied to an intelligent robot and includes the following method steps: receiving interaction information input by a user, performing semantic analysis on the interaction information, and obtaining a user question set; wherein the user question set includes several user questions obtained by the semantic analysis; performing a limited range assessment on each of the user questions in the user question set, performing a safety analysis on the user questions that exceed the limited range based on an accident case library, and obtaining a problem safety value; wherein the accident case library is embedded in the storage device of the intelligent robot; for the user questions whose problem safety values ​​are higher than the safety threshold, obtaining relevant knowledge through the network, generating answer content based on the obtained relevant knowledge, and outputting the answer content to the user.

[0024] The intelligent robot in the present invention is embedded with a preset knowledge set, and the knowledge contained therein corresponds to the corresponding production area of ​​the production plant. Based on the preset knowledge set, the intelligent robot can answer relevant questions to the staff in the production area to assist in the production of products, the operation and maintenance of production equipment, etc. At the same time, the intelligent robot in the present invention also performs a limited range evaluation on each user question contained in the interactive information input by the user. If it exceeds the limited range, that is, the intelligent robot cannot answer the user question based on the knowledge in the preset knowledge set, it is determined that the user has an out-of-range consultation. At this time, based on the accident case library, a safety analysis is performed on the user questions that exceed the limited range to obtain the problem safety value. For those user questions whose problem safety values ​​are higher than the safety threshold, the intelligent robot responds and obtains external relevant knowledge through the network, and generates the answer content accordingly, thereby realizing interaction with the user; and for those user questions whose problem safety values ​​are lower than the safety threshold, the intelligent robot determines that there is a safety risk if it responds to them, so it does not respond, or prompts the user that the question cannot be answered.

[0025] An intelligent robot typically includes a processing device, a storage device, an interactive interface, and an input device, wherein the storage device stores a preset knowledge set and an operating program code suitable for the processing device, and the operating program code constitutes an AI large model that supports the intelligent robot to interact with the user; the interactive interface is used to interact with the user, such as a display screen; the input device is such as a keyboard, mouse, microphone, camera, etc., and the user can input his or her required information to the intelligent robot through the input device. Among them, the interactive interface and the input device can be an integrated design, which is not specifically limited by the present invention. In addition, the intelligent robot should also include basic configurations such as a basic shell, power supply components, and mobile components, which will not be elaborated here.

[0026] As an example, the receiving of interaction information input by the user includes: obtaining the user's identity information, retrieving the user's historical interaction records based on the identity information, and obtaining the user's average single interaction duration based on the historical interaction records; receiving the interaction information input by the user within the average single interaction duration, and if the user is still inputting when the average single interaction duration is reached, extending the average single interaction duration by a set percentage of the average single interaction duration, until an upper limit on the number of extensions is reached.

[0027] In this embodiment, the intelligent robot is equipped with identity entry devices such as cameras and fingerprint identifiers. Every time a user needs to interact with the intelligent robot, he must first enter his own identity information through the above-mentioned identity entry device. The intelligent robot retrieves the user's historical interaction record based on the identity information. The historical interaction record records the input duration of each interaction of the user, so that the average single interaction duration of the user can be statistically analyzed. Therefore, the intelligent robot sets the input duration of the user's interactive information this time as the average single interaction duration, and continues to receive information within the average single interaction duration. In this way, even if the user pauses, thinks, etc., the information reception will not be terminated, which can ensure the integrity of the user's input information. Moreover, when the average single interaction duration is reached, if the user is still inputting information, it will be extended by a set number of times (for example, 5 times), and each extension is extended by a set percentage of the average single interaction duration, and the set percentage is, for example, 30%.

[0028] As an example, the semantic analysis of the interactive information to obtain a set of user questions includes: converting the interactive information into text segments, preprocessing the text segments, including checking for spelling and typos, and correcting errors; using a syntactic analysis technique combined with context to perform word segmentation, stem extraction, and sentence restoration on the preprocessed text segments to obtain a number of the user questions that constitute the set of user questions.

[0029] In this embodiment, voice interaction is a more commonly used interaction method, and its interaction speed is significantly faster than that of text interaction. Therefore, the intelligent robot in the present invention first determines the type of interactive information input by the user. If it is voice information, the voice information is first converted into a text segment, and the converted text segment is preprocessed including spelling and typo checking, error correction, etc. to ensure the accuracy of subsequent analysis; then, the preprocessed text segment is subjected to word segmentation, stem extraction, and sentence restoration using a syntactic analysis technique combined with the context, and a number of user questions are obtained to form the user question set. If the interactive information is text information, the syntactic analysis technique combined with the context is directly used to process it after preprocessing to obtain a user question set.

[0030] As an example, the limited range evaluation of each user question in the user question set includes: for each user question in the user question set, trial answering each user question based on a preset knowledge set to obtain a number of alternative answer contents, and calculating a first quantity of each alternative answer content; wherein the preset knowledge set is embedded in the storage device of the intelligent robot; if the first quantity is greater than a quantity threshold, it is determined that the user question exceeds the limited range; otherwise, it is determined that the user question does not exceed the limited range; wherein the quantity threshold is determined in the following manner: similarity matching is performed on the user question and the interaction record data to obtain a second quantity of similar historical user questions, and the quantity threshold is calculated based on the second quantity.

[0031] In this embodiment, when evaluating each user question within a limited scope, the intelligent robot in the present invention will first perform a trial answer process on the user question, that is, select relevant knowledge from the preset knowledge set, and then convert it into alternative answer content that conforms to natural language. When performing the answer process, the intelligent robot will generate multiple alternative answer contents that may meet the user's intention. During normal interaction, the intelligent robot will output the alternative answer content that best meets the user's intention to the user. Since the trial answer process does not involve outputting the alternative answer content, it is only necessary to calculate the first quantity of each alternative answer content.

[0032] The first number of alternative answer contents generated by the intelligent robot can be used to determine the above-mentioned quantity threshold. Specifically, historical user questions similar to the user question are matched in the interaction record data, and their second number is counted. After the intelligent robot answers similar questions for many times, its ability to grasp the user's true intention gradually increases, and the number of alternative answer contents determined correspondingly will gradually decrease, that is, low-correlation alternative answer contents will not be generated. In view of this actual situation, the present invention sets a quantity threshold calculated according to the second number. Obviously, the quantity threshold is negatively correlated with the second number. In other words, after the intelligent robot answers similar questions for more times, the upper limit of the number of alternative answer contents generated during the trial answer process will be lower, that is, the quantity threshold is set smaller; conversely, the upper limit of the number of alternative answer contents generated during the trial answer process will be higher, that is, the quantity threshold is set larger. Among them, the quantity threshold and the above-mentioned second number can be preset in a comparison table, and the corresponding quantity threshold can be determined by querying the comparison table.

[0033] Therefore, when the first number of alternative answers generated in the trial answering process is greater than the quantity threshold, it is determined that the intelligent robot cannot grasp the user's question more accurately, which indirectly indicates that the question has exceeded the intelligent robot's answering ability. At this time, it can be determined that the user's question exceeds the specified range; otherwise, it is determined that the user's question does not exceed the specified range.

[0034] The present invention determines whether the user's question exceeds the knowledge scope of the preset knowledge set by analyzing the intelligent robot's ability to accurately answer the user's question. The analysis accuracy is higher, and because there is no need to perform complex question matching analysis with the preset knowledge set, the determination speed is also faster.

[0035] As an example, Figure 2 As shown, the safety analysis of the user problems beyond the specified range is performed based on the accident case library to obtain the problem safety value, including: performing semantic analysis on each of the user problems in the user problem set to obtain the most relevant topic object; wherein the topic object refers to a certain type of equipment in the production area where the intelligent robot is deployed; extracting each accident case information related to the topic object from the accident case library, the accident case information including the equipment type and the cause of the accident, and the cause of the accident including the erroneous operation information; calculating the positive correlation level between the user problems beyond the specified range and each of the accident causes, and obtaining the problem safety value by matching the maximum positive correlation level with a preset comparison table.

[0036] In this embodiment, the intelligent robot of the present invention is also embedded with an accident case library, which records multiple accident records of various types of equipment. Each accident record includes the type of equipment where the accident occurred and the specific cause of the accident. The present invention sets the specific cause of the accident to be the user's incorrect operation, rather than the device's own cause.

[0037] First, semantic analysis is performed on all user questions in the user question set, and the topic objects involved in each question raised by the user can be determined. The topic object refers to a certain type of equipment in the production area where the intelligent robot is deployed. For example, the user mentioned three types of equipment in this interactive information, "Device A and device B always overheat. Is it because the gear of device C used for heat dissipation is too low? I plan to increase the workload of device C by two gears. Is it feasible?" After semantic analysis, it can be determined that the user should be asking about device C, not device A and device B.

[0038] Next, the accident case information related to the C device is extracted from the accident case library, and the positive correlation level between the user questions that exceed the limited range and the causes of each accident is calculated. The positive correlation level here refers to the similarity between the operation intention involved in the user question and the erroneous operation information that caused the accident in the relevant accident case information. The higher the similarity, the higher the corresponding positive correlation level, that is, if the user takes this operation, there is a higher probability that the corresponding type of equipment will have an accident. At the same time, a comparison table is preset, which contains the correspondence between multiple sets of positive correlation levels and problem safety values. After determining the positive correlation level (the maximum value of all positive correlation levels), the corresponding problem safety value can be obtained by querying the preset comparison table.

[0039] It should be noted that when the safety value of the problem is lower than the safety threshold, the intelligent robot can output an answer such as "Do not take this action, please consult a senior engineer."

[0040] As an example, the positive correlation level between the user problems beyond a limited range and each of the accident causes is calculated by an analysis model based on random forest and correlation coefficient.

[0041] In this embodiment, the present invention selects to use the random forest algorithm and the correlation coefficient to construct the analysis model, and uses the analysis model to calculate the positive correlation level between the user problems beyond the specified range and each cause of the accident. The random forest algorithm can handle complex nonlinear relationships and can provide the importance ranking of features, while the correlation coefficient can be used to quantify the strength of the linear relationship between two variables, thereby obtaining the positive correlation level between the user problems beyond the specified range and each cause of the accident.

[0042] The correlation coefficient refers to the degree of correlation between the important features extracted from the cause of the accident and the features of the user problem beyond the specified range, and the Spearman or Kendall correlation coefficient may be used.

[0043] In addition, the above analysis model can also be constructed based on the GPT or BERT large model to be used to infer the positive correlation level between user problems that exceed the specified range and the causes of each accident. This type of large model has stronger generalization capabilities and does not require model construction. It only needs to be fine-tuned to a certain extent using small sample data, which makes it easier to implement.

[0044] like Figure 3As shown, the present invention also provides a human-computer interaction system based on an AI big model, which is applied to an intelligent robot, and the system includes a receiving module, an evaluation and safety analysis module, and a network acquisition module; the receiving module is used to receive interaction information input by a user, perform semantic analysis on the interaction information, and obtain a user question set; wherein the user question set includes several user questions obtained by semantic analysis; the evaluation and safety analysis module is used to perform a limited range evaluation on each of the user questions in the user question set, perform a safety analysis on the user questions that exceed the limited range based on an accident case library, and obtain a problem safety value; wherein the accident case library is embedded in the storage device of the intelligent robot; the network acquisition module is used to obtain relevant knowledge through the network for the user questions whose problem safety values ​​are higher than the safety threshold, generate answer content based on the acquired relevant knowledge, and output the answer content to the user.

[0045] The present invention also discloses an electronic device, which includes: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the computer program implements any of the methods described above when executed by the processor.

[0046] The present invention also discloses a computer storage medium, which stores a computer program that can be executed by a processor to implement any of the methods described above.

[0047] The present invention also discloses a computer program product, which includes a computer program that can be executed by a processor to implement any of the methods described above.

[0048] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0049] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A human-computer interaction method based on an AI large model, applied to an intelligent robot, characterized in that: The method comprises the following steps: receiving interactive information input by a user, performing semantic analysis on the interactive information, and obtaining a user question set; wherein the user question set includes a number of user questions obtained by semantic analysis; performing a limited range assessment on each user question in the user question set, and performing a safety analysis on the user questions exceeding the limited range based on an accident case library to obtain a question safety value; wherein the accident case library is embedded in the storage device of the intelligent robot; for the user questions whose question safety values ​​are higher than a safety threshold, obtaining relevant knowledge through a network, generating answer content based on the obtained relevant knowledge, and outputting the answer content to the user; Performing a limited range evaluation on each of the user questions in the user question set, including: for each of the user questions in the user question set, performing a trial answer process on each of the user questions based on a preset knowledge set, obtaining a number of alternative answer contents, and calculating a first quantity of each of the alternative answer contents; wherein the preset knowledge set is embedded in the storage device of the intelligent robot; if the first quantity is greater than a quantity threshold, then determining that the user question exceeds the limited range; otherwise, determining that the user question does not exceed the limited range; wherein the quantity threshold is determined in the following manner: performing similarity matching on the user question and the interaction record data, obtaining a second quantity of similar historical user questions, and calculating the quantity threshold based on the second quantity.

2. According to claim 1, a human-computer interaction method based on an AI large model is characterized in that: Receiving interaction information input by a user includes: obtaining the user's identity information, retrieving the user's historical interaction record based on the identity information, and obtaining the user's average single interaction time based on the historical interaction record; receiving the interaction information input by the user within the average single interaction time, and if the user is still inputting when the average single interaction time is reached, extending the average single interaction time by a set percentage of the average single interaction time until an upper limit on the number of extensions is reached.

3. The human-computer interaction method based on AI big model according to claim 2 is characterized in that: Performing semantic analysis on the interactive information to obtain a user question set includes: converting the interactive information into text segments, preprocessing the text segments, including checking for spelling and typos, and correcting errors; performing word segmentation, stem extraction, and sentence restoration on the preprocessed text segments using a syntactic analysis technique combined with context to obtain a number of the user questions that constitute the user question set.

4. The human-computer interaction method based on AI big model according to claim 1, characterized in that: Based on the accident case library, a safety analysis is performed on the user problems that exceed the limited range to obtain a problem safety value, including: performing a semantic analysis on each of the user problems in the user problem set to obtain the most relevant topic object; wherein the topic object refers to a certain type of equipment in the production area where the intelligent robot is deployed; extracting each accident case information related to the topic object from the accident case library, the accident case information includes equipment type and accident cause, and the accident cause includes erroneous operation information; calculating the positive correlation level between the user problems that exceed the limited range and each of the accident causes, and obtaining the problem safety value by matching the maximum positive correlation level with a preset comparison table.

5. The human-computer interaction method based on AI big model according to claim 4 is characterized in that: The positive correlation level between the user problems beyond a limited range and each of the accident causes is calculated by an analysis model based on random forest and correlation coefficient.

6. A human-computer interaction system based on AI big model, applied to intelligent robots, characterized in that: The system includes a receiving module, an evaluation and safety analysis module, and a network acquisition module; the receiving module is used to receive interactive information input by a user, perform semantic analysis on the interactive information, and obtain a user question set; wherein the user question set includes a number of user questions obtained by semantic analysis; the evaluation and safety analysis module is used to perform a limited range evaluation on each user question in the user question set, perform safety analysis on the user questions that exceed the limited range based on an accident case library, and obtain a problem safety value; wherein the accident case library is embedded in the storage device of the intelligent robot; the network acquisition module is used to obtain relevant knowledge through the network for the user questions whose problem safety values ​​are higher than the safety threshold, generate answer content based on the obtained relevant knowledge, and output the answer content to the user; Performing a limited range evaluation on each of the user questions in the user question set, including: for each of the user questions in the user question set, performing a trial answer process on each of the user questions based on a preset knowledge set, obtaining a number of alternative answer contents, and calculating a first quantity of each of the alternative answer contents; wherein the preset knowledge set is embedded in the storage device of the intelligent robot; if the first quantity is greater than a quantity threshold, then determining that the user question exceeds the limited range; otherwise, determining that the user question does not exceed the limited range; wherein the quantity threshold is determined in the following manner: performing similarity matching on the user question and the interaction record data, obtaining a second quantity of similar historical user questions, and calculating the quantity threshold based on the second quantity.

7. An electronic device, characterized in that: The electronic device comprises: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the computer program implements the method according to any one of claims 1 to 5 when executed by the processor.

8. A computer storage medium, characterized in that: The computer storage medium stores a computer program that can be executed by a processor to implement the method according to any one of claims 1 to 5.

9. A computer program product, characterized in that: The computer program product comprises a computer program executable by a processor to implement the method according to any one of claims 1 to 5.

Images