Intelligent man-machine interaction method and system for dual-mode data processing

Abstract

The invention discloses an intelligent man-machine interaction method and system for dual-mode data processing, and relates to the technical field of data processing. Comprising the steps of obtaining an effective area for collecting user behavior data in an identification area; and dividing the user behavior data into first user behavior data and second user behavior data based on the user behavior data type. According to the method, the user behavior data are classified into two different types of behavior data, and the priority processing queue which is endowed with the priority authority to reach the behavior data standard is constructed, so that when the two different types of behavior data are processed, the behavior data with the high priority authority can be analyzed and processed firstly; according to the method, the behavior data with the high priority authority is acquired from the historical user behavior data, and the second user behavior data linked with the behavior data with the high priority authority is retrieved from the historical user behavior data based on the behavior data with the high priority authority, so that when data is interactively processed, another associated data can be quickly acquired through one data type, and the processing speed is guaranteed to a certain extent.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to an intelligent human-computer interaction method and system for dual-mode data processing. Background Art

[0002] Dual-mode data processing refers to the simultaneous processing of data from two different modes, such as voice and text, gestures and images, during the human-computer interaction process. Intelligent human-computer interaction refers to the direct interaction between humans and computers through computers. At this stage, with the development of intelligent human-computer interaction, intelligent human-computer interaction methods and systems for dual-mode data processing have been rapidly developed and applied.

[0003] After searching, the Chinese invention patent with publication number "CN114661248A" discloses a "data processing method and device". This application detects that the target checkpoint is created, and then reads the data processing status associated with the target checkpoint in the target memory according to the checkpoint identifier. Then, the data processing status and the task identifier corresponding to the data processing task are written into the target storage space, thereby realizing persistent storage of the data processing status through the storage space.

[0004] The Chinese invention patent with publication number "CN117784937A" discloses "an intelligent human-computer interaction system and method based on multimodal features". The application converts biometric features into text information through voice recognition and image recognition, thereby effectively protecting user privacy while providing intelligent services.

[0005] Since the intelligent human-computer interaction with dual-mode data processing needs to analyze two different types of data during operation, the above-mentioned disclosed method and similar methods do not set priorities during actual identification. As a result, when the intelligent human-computer interaction system with dual-mode data processing processes multi-task interaction data, there will be a great degree of simultaneous processing scenarios, which increases the pressure of data processing to a certain extent and is prone to processing bottlenecks, thereby causing a decrease in processing speed. Summary of the invention

[0006] The purpose of the present invention is to provide an intelligent human-computer interaction method and system for dual-mode data processing to solve the problems raised in the above background technology.

[0007] In the first aspect, to achieve the above-mentioned purpose, the present invention provides the following technical solutions: an intelligent human-computer interaction method for dual-mode data processing, applied to an intelligent human-computer interaction system for dual-mode data processing, the intelligent human-computer interaction system for dual-mode data processing comprising a plurality of collection nodes for identifying user behavior data, the plurality of collection nodes forming an identification area, the identification area being used to obtain user behavior data, the intelligent human-computer interaction method for dual-mode data processing comprising:

[0008] Obtaining a valid area within the identification area for collecting user behavior data;

[0009] Based on the type of user behavior data, dividing into first user behavior data and second user behavior data;

[0010] Constructing a priority processing queue, the priority processing queue acts on the first user behavior data and the second user behavior data, assigning priority authority based on the identification level of the first user behavior data and the second user data in the effective area, and outputting the first priority data and the second priority data;

[0011] An inductive model is constructed, which is used to read the first preferred data and detect the second user behavior data associated with the first preferred data type in the historical user behavior data, and finally integrate and output the interaction data of the intelligent human-computer interaction system acting on the dual-mode data processing.

[0012] As a further preferred embodiment of the present technical solution, a method for obtaining the effective area includes:

[0013] The distance between two adjacent collection nodes is artificially set to a fixed value, a first collection node that collects user behavior data is obtained in the identification area, and all collection nodes except the first collection node are searched in the identification area based on the first collection node. When it is found that there is a collection node that does not obtain user behavior data, the collection node that does not obtain user behavior data is marked as a second collection node. There are two second collection nodes, and the number of collection nodes between the first collection node and the two second collection nodes is determined;

[0014] A classification determination method is set, and the classification determination method obtains the range of the effective area through the number of the collection nodes and the fixed value based on the position of the first collection node in the identification area.

[0015] As a further preferred embodiment of the present technical solution, the classification determination method includes:

[0016] When the first acquisition node is at the midpoint of the identification area, the product of twice the number of acquisition nodes and a fixed value is obtained, and the range of the effective area is determined.

[0017] As a further preferred embodiment of the present technical solution, the classification determination method includes:

[0018] When the first acquisition node is not at the midpoint of the identification area, a third acquisition node is acquired, the third acquisition node is between the first acquisition node and the second acquisition node, and at least one of the acquisition nodes is arranged between the third acquisition node and the first acquisition node, and two third acquisition nodes are arranged;

[0019] The number of collection nodes between the two third collection nodes and the first collection node and the second collection node is obtained respectively, and the range of the effective area is determined.

[0020] As a further preferred embodiment of the present technical solution, the priority processing queue includes:

[0021] Acquire user behavior data of users that can be triggered and identified in the identification area, and associate the user behavior data with a time axis to form evaluation benchmark data;

[0022] Get real-time user behavior data;

[0023] A comparison method is constructed, which outputs first priority data and second priority data based on real-time user behavior data and evaluation benchmark data.

[0024] As a further preferred embodiment of the present technical solution, the comparison method includes:

[0025] Construct a time-based user behavior data type axis;

[0026] Inputting the evaluation benchmark data and the real-time user behavior data into the user behavior data type coordinate axis respectively;

[0027] And form a judgment benchmark curve and a real-time user behavior curve;

[0028] Compare the arc point slopes based on the user behavior data type in the judgment reference curve and the real-time user behavior curve, extract the arc point types in the real-time user behavior curve whose arc point slopes are greater than or equal to the arc point slopes of the judgment reference curve, assign high priority permissions, and output the first priority data and the second priority data in sequence.

[0029] As a further preferred embodiment of the present technical solution, the inductive model includes:

[0030] Acquire historical user behavior data, based on the type of the first priority data, search for a user behavior data arrangement that matches the type of the first priority data in the historical user behavior data, read the type of the second priority data, and search for matching second user behavior data in the user behavior data arrangement;

[0031] The value of the second priority data is overwritten on the second user behavior data and combined with the first priority data to form interaction data for the intelligent human-computer interaction system acting on the dual-mode data processing.

[0032] As a further preferred embodiment of the present technical solution, the user behavior data includes: body movement data, voice data and facial image data, the user behavior data types include: body, voice and face, the first user behavior data and the second user behavior data are any one or a combination of two of the body movement data, voice data and facial image data, and the first user behavior data and the second user behavior data do not contain repeated user behavior data types.

[0033] In the second aspect, in order to improve the above-mentioned intelligent human-computer interaction method for dual-mode data processing, the present invention provides an intelligent human-computer interaction system for dual-mode data processing. It should be noted that the intelligent human-computer interaction system for dual-mode data processing disclosed in the present invention uses the above-mentioned intelligent human-computer interaction method for dual-mode data processing, and includes:

[0034] An identification area module, including a plurality of collection nodes for identifying user behavior data, is used to obtain a valid area for collecting user behavior data within the identification area;

[0035] A data preprocessing module, used to perform preliminary processing on the collected raw data, including data cleaning, denoising and format conversion, and identifying and dividing the first user behavior data and the second user behavior data;

[0036] A priority processing module, for constructing a priority processing queue according to a preset priority processing rule, prioritizing the first user behavior data and the second user behavior data, and assigning different priority permissions to different data based on the arc point slope of the data type by comparing the real-time user behavior data with the evaluation benchmark data;

[0037] The data summarization and integration module is used to take the first priority data, find the second user behavior data associated with it in the historical user behavior data, and perform numerical coverage and integration on the found second user behavior data, and combine it with the first priority data to generate the interaction data that ultimately acts on the system;

[0038] The interaction execution and feedback module is used to execute corresponding operations or instructions based on the generated interaction data, such as responding to the user's voice, body language and facial instructions.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] The intelligent human-computer interaction method and system for dual-mode data processing classifies user behavior data into two different types of behavior data, and constructs a priority processing queue that grants priority permissions based on behavior data standards. When processing two different types of behavior data, the behavior data with high priority permissions can be analyzed and processed first, and based on the behavior data with high priority permissions, the second user behavior data that is linked to the behavior data with high priority permissions is retrieved from historical user behavior data. When interactively processing data, one data type can be used to quickly obtain another related data, thereby ensuring the processing speed to a certain extent. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a diagram of the operation steps of the method of the present invention;

[0042] Figure 2 is an auxiliary illustration of step S230 of the present invention;

[0043] Figure 3 is an auxiliary illustration of step S331 of the present invention;

[0044] Figure 4 It is a system composition diagram of the present invention. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] It should be noted that in the use scenario of the science and technology exhibition hall, when the user faces the display robot equipped with the intelligent human-computer interaction system with dual-mode data processing, since the user may input actions, language and images at the same time when interacting with the display robot, the display robot is prone to processing bottlenecks when processing two or more interactions, resulting in a decrease in processing speed. Figure 1 It can be seen that in this embodiment, an intelligent human-computer interaction method for dual-mode data processing is proposed, including steps S100 to S400.

[0047] Step S100: constructing a recognition area.

[0048] Specifically, in the present embodiment, the recognition area is constructed by setting a number of collection nodes for identifying user behavior data. These collection nodes can be any one or a combination of two of various sensors or collection devices installed on the outside of the intelligent robot. Specifically, the collection device can be a miniature camera and a miniature collection microphone, and the various sensors can be visual sensors and sound source sensors. It should be added that, in the present embodiment, the range of the recognition area is the sum of the collection ranges of the sensors or collection devices.

[0049] It should be clear that since the behavioral actions performed by the user have a certain range, when the intelligent robot interacts with the user, all the collection nodes in the identification area run at a high frequency at the same time, which reduces the use cycle of the intelligent robot and increases the operating pressure of the collection nodes. For this reason, it is necessary to obtain the effective area in the identification area. Therefore, in this embodiment, step S200 is set to solve the problem arising in this scenario.

[0050] Step S200: Acquire a valid area within the identification area for collecting user behavior data.

[0051] It should be noted that the effective area is used to limit the identification interaction position based on the user interaction position to improve the interaction speed, where the user interaction position in this embodiment refers to the position where the intelligent robot receives the user behavior data input by the user to the intelligent robot.

[0052] Furthermore, it should be supplemented that, in this embodiment, the method for acquiring the effective area includes: step S210 to step S230.

[0053] Step S210: artificially setting the distance between two adjacent collection nodes to a fixed value.

[0054] Specifically, in this embodiment, artificially setting the distance between two adjacent collection nodes as a fixed value means that the collection range of the sensor or collection device in the identification area is set to the same, and is indirectly set to the same. In addition, it should be added that during actual operation, the fixed value can be set to 1 meter, wherein the fixed value can be adjusted according to actual application scenarios and requirements during actual operation. For example, when higher precision data is required, the fixed value can be set to a smaller value, such as 0.5 meters. On the contrary, when the data accuracy requirement is not high, or in order to reduce the amount of data processing, the fixed value can be set to a larger value, such as 1.5 meters.

[0055] Step S220: Acquire a first collection node that collects user behavior data in the identification area.

[0056] Specifically, in the present embodiment, the purpose of acquiring the first collection node is to determine the starting point of the user behavior data. In the identification area, when user behavior data is collected, the system will first identify which collection node first detects the data. This node is the first collection node. It should be added that when the present embodiment is actually running, the first collection node is randomly selected from all collection nodes that collect user behavior data in the identification area by a computer.

[0057] It should be noted that, in the present embodiment, the user behavior data includes: body movement data, voice data and facial image data, the user behavior data types include: body, voice and face, the first user behavior data and the second user behavior data are any one or a combination of two of the body movement data, voice data and facial image data, and the first user behavior data and the second user behavior data do not contain repeated user behavior data types.

[0058] Step S230: taking the first collection node as a reference, searching and identifying all collection nodes in the region except the first collection node, and when it is found that there are collection nodes that have not obtained user behavior data, marking the collection nodes that have not obtained user behavior data as second collection nodes.

[0059] Specifically, in this embodiment, two second acquisition nodes are provided for determining any one of the width or height of the identification area. In actual operation, this embodiment can also set four second acquisition nodes to further limit the scope of the identification area.

[0060] In addition, in this embodiment, the second acquisition node is obtained by a computer, and the number of acquisition nodes between the first acquisition node and the two second acquisition nodes is determined, and a classification determination method is set. The classification determination method is based on the position of the identification area where the first acquisition node is located, and obtains the range of the effective area through the number of acquisition nodes and a fixed value.

[0061] It should be noted that the reference Figure 2 It can be seen that in this embodiment, a computer is used to receive the amount of data collected by each collection node over time, and a computer graphics program in the prior art is used to construct a dynamic curve chart based on the amount of data collected and time. Figure 2 It can be seen that the slopes of the first collection node and the collection node that obtains the user behavior data are the same, while the slope of the second collection node is 0 because the second collection node does not recognize the user behavior data.

[0062] It should be supplemented that, in this embodiment, the steps of the classification determination method include: step S231 and step S232.

[0063] Step S231: It should be noted that when the first acquisition node is at the midpoint position in the identification area, the product of twice the number of acquisition nodes and a fixed value is obtained, and the range of the effective area is determined.

[0064] Specifically, in this embodiment, the number of acquisition nodes is 3, and the value of the set value is set to 0.25m. At this time, the range of the effective area is: 3X2X0.25=1.5m.

[0065] Step S232: When the first acquisition node is not at the midpoint of the identification area, obtain a third acquisition node, the third acquisition node is between the first acquisition node and the second acquisition node, and at least one acquisition node is set between the third acquisition node and the first acquisition node, and there are two third acquisition nodes, respectively obtain the number of acquisition nodes between the two third acquisition nodes and the first acquisition node and the second acquisition node, and determine the range of the effective area.

[0066] Specifically, in this embodiment, the number of acquisition nodes is 5 and 3 respectively, and the value of the set value is 0.15m. At this time, the distance between the first acquisition node and the second acquisition node is 0.3m, and the two third acquisition nodes are respectively located between the first acquisition node and the second acquisition node. The distance between each third acquisition node and the adjacent acquisition node is 0.1m. Therefore, the range of the effective area is 0.15m extending from the first acquisition node to the second acquisition node, and 0.15m extending from the second acquisition node to the first acquisition node, forming an area with a total length of 0.3m.

[0067] Step S300: construct a priority processing queue.

[0068] Specifically, in this embodiment, the priority processing queue includes steps S310 to S330.

[0069] Step S310: Obtain user behavior data of users that can be triggered for identification in the identification area, and associate the user behavior data with a time axis to form evaluation benchmark data.

[0070] It should be clear that in this embodiment, the evaluation benchmark data is the data of user interaction actions that the intelligent robot can recognize within the recognition area, and the duration of the data.

[0071] Step S320: Obtain real-time user behavior data.

[0072] It should be clear that in this embodiment, real-time user behavior data, historical user behavior data, and user behavior data respectively represent the user behavior data acquired by the collection node at the current stage, the user behavior data acquired by the collection node in the past time period, and the user behavior data of all time periods in a macro sense.

[0073] Step S330: construct a comparison method, which outputs first priority data and second priority data based on real-time user behavior data and evaluation benchmark data.

[0074] Specifically, in this embodiment, the comparison method includes steps S331 to S334.

[0075] Step S331: Construct a time-based user behavior data type coordinate axis.

[0076] Specifically, refer to Figure 3 It can be seen that in this embodiment, the time-based user behavior data type coordinate axis is constructed by using an existing computer mapping program, by setting the time axis as the X-axis and the user behavior data value as the Y-axis, thereby forming a two-dimensional coordinate system.

[0077] Step S332: inputting the judgment benchmark data and the real-time user behavior data into the user behavior data type coordinate axis respectively, and forming a judgment benchmark curve and a real-time user behavior curve.

[0078] Specifically, in the present embodiment, the discrimination reference curve and the real-time user behavior curve are two different curves, wherein the discrimination reference curve is a solid line segment, and the real-time user behavior curve is a wavy line. It should be supplemented that, since the user behavior data types include: body, voice and face, the first user behavior data and the second user behavior data are any one or a combination of two of body movement data, voice data and facial image data, and the first user behavior data and the second user behavior data do not contain repeated user behavior data types, and therefore there are multiple settings of the discrimination reference curve and the real-time user behavior curve obtained.

[0079] Step S333: Compare and judge the slopes of arc points based on the user behavior data type in the reference curve and the real-time user behavior curve.

[0080] Specifically, in this embodiment, the computer can obtain the similarity, that is, the change trend, between the discrimination benchmark and the user behavior data by acquiring the arc point slopes of the discrimination benchmark curve and the user behavior data type.

[0081] Step S334: extract arc point types in the real-time user behavior curve whose arc point slope is greater than or equal to the arc point slope of the judgment reference curve, assign high priority permissions, and output first priority data and second priority data in sequence.

[0082] It should be noted that, in this embodiment, when the computer reads the arc point type (condition) in which the slope of the arc point in the real-time user behavior curve is greater than or equal to the slope of the arc point of the judgment reference curve, at this time, any one of the first user behavior data or the second behavior data that meets the condition meets the behavior data standard, is assigned a priority processing queue with priority authority, and is output as the first priority data. At this time, the remaining other user behavior data is output as the second priority data.

[0083] It should be clear that in the scenario set in this embodiment, the intelligent robot has the ability of dual-mode data processing, that is, it is installed with two processors to process different types of user behavior data respectively. Therefore, when processing multi-task interactive data, simultaneous processing scenarios will occur to a great extent, which increases the pressure of data processing to a certain extent and is prone to processing bottlenecks, thereby causing a decrease in processing speed. Therefore, in this embodiment, step S400 is set to solve the problems arising in this scenario.

[0084] Step S400: construct an inductive model and output interaction data.

[0085] Specifically, in this embodiment, the inductive model is used to quickly retrieve the user data arrangement associated with the first priority data in the historical user behavior data of the intelligent robot based on the first priority data, and speed up the processing of the second priority data by reducing and narrowing the search scope and extracting historical user data.

[0086] It should be noted that, in this embodiment, the induction model includes steps S410 to S440.

[0087] Step S410: Obtain historical user behavior data.

[0088] Specifically, in this embodiment, historical user behavior data is acquired by a computer and stored separately in the real-time user behavior data, wherein the historical user behavior data is directly set in the storage unit of the intelligent robot in the scenario applied in this embodiment.

[0089] Step S420: Based on the type of the first priority data, retrieve the user behavior data arrangement that matches the type of the first priority data in the historical user behavior data.

[0090] Specifically, in the present embodiment, the type of the first priority data is based on the type of the user behavior data, that is, the types of the first priority data include: body, voice and face. It should be added that the user behavior data arrangement in the present embodiment specifically refers to an arrangement based on the body, voice and face in the user behavior data into a type similar to a (body, language) combination or a (body language, face) combination. In addition, it should be clear that retrieving the user behavior data arrangement matching the type of the first priority data in the historical user behavior data is achieved through an existing data retrieval program in the computer. Specifically, in the present embodiment, the computer determines the collection node by obtaining the source of the first priority data, and obtains the type of the first priority data by extracting the interaction interface between the collection node and the intelligent robot. Then, the computer retrieves the matching user behavior data arrangement in the user behavior data arrangement. For example, the type of the first priority data is voice, and the computer retrieves the voice and obtains the (body, language) combination, the (face, language) combination and the (body language, face) combination.

[0091] Step S430: Read the type of the second priority data, and search for matching second user behavior data in the user behavior data array.

[0092] Specifically, in this embodiment, after the user behavior data arrangement is obtained, the second user behavior data is further selected by extracting the type of the second priority data.

[0093] Step S440: Overwrite the second user behavior data with the value of the second priority data and combine with the first priority data to form interaction data for the intelligent human-computer interaction system acting on the dual-mode data processing.

[0094] It should also be added that reference Figure 4 It can be seen that in this embodiment, in order to improve the intelligent human-computer interaction method for dual-mode data processing proposed above, an intelligent human-computer interaction system for dual-mode data processing is also proposed, and includes:

[0095] The recognition area module includes a plurality of collection nodes for recognizing human behavior data, and is used to obtain a valid area in the recognition area for collecting user behavior data;

[0096] A data preprocessing module, used to perform preliminary processing on the collected raw data, including data cleaning, denoising and format conversion, and identifying and dividing the first user behavior data and the second user behavior data;

[0097] A priority processing module, for constructing a priority processing queue according to a preset priority processing rule, prioritizing the first user behavior data and the second user behavior data, and assigning different priority permissions to different data based on the arc point slope of the data type by comparing the real-time user behavior data with the evaluation benchmark data;

[0098] The data summarization and integration module is used to take the first priority data, find the second user behavior data associated with it in the historical user behavior data, and perform numerical coverage and integration on the found second user behavior data, and combine it with the first priority data to generate the interaction data that ultimately acts on the system;

[0099] The interaction execution and feedback module is used to execute corresponding operations or instructions based on the generated interaction data, such as responding to the user's voice, body language and facial instructions.

[0100] It should be noted that when this embodiment is used in the scenario provided in this application, the recognition area module is a camera, microphone or various sensors in the intelligent robot, including visual sensors and sound source sensors, which are used to capture the user's body movements, voice commands, facial expressions and other behavioral data in real time.

[0101] The data preprocessing module, priority processing module, data summarization and integration module, and interactive execution and feedback module are the processing chips inside the intelligent robot.

[0102] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is limited by the attached embodiments and their equivalents.

Claims

1. An intelligent human-computer interaction method for dual-mode data processing, applied to an intelligent human-computer interaction system for dual-mode data processing, the intelligent human-computer interaction system for dual-mode data processing comprising a plurality of acquisition nodes for identifying user behavior data, the plurality of acquisition nodes forming an identification area, the identification area being used to obtain user behavior data, characterized in that: The intelligent human-computer interaction method for dual-mode data processing includes: Obtaining a valid area within the identification area for collecting user behavior data; Based on the type of user behavior data, dividing into first user behavior data and second user behavior data; Constructing a priority processing queue, the priority processing queue acts on the first user behavior data and the second user behavior data, assigning priority authority based on the identification level of the first user behavior data and the second user data in the effective area, and outputting the first priority data and the second priority data; An inductive model is constructed, which is used to read the first preferred data and detect the second user behavior data associated with the first preferred data type in the historical user behavior data, and finally integrate and output the interaction data of the intelligent human-computer interaction system acting on the dual-mode data processing.

2. The intelligent human-computer interaction method for dual-mode data processing according to claim 1, characterized in that: Methods for obtaining the effective area include: The distance between two adjacent collection nodes is artificially set to a fixed value, a first collection node that collects user behavior data is obtained in the identification area, and all collection nodes except the first collection node are searched in the identification area based on the first collection node. When it is found that there is a collection node that does not obtain user behavior data, the collection node that does not obtain user behavior data is marked as a second collection node. There are two second collection nodes, and the number of collection nodes between the first collection node and the two second collection nodes is determined; A classification determination method is set, and the classification determination method obtains the range of the effective area through the number of collection nodes and a fixed value based on the position of the first collection node in the identification area.

3. The intelligent human-computer interaction method for dual-mode data processing according to claim 2, characterized in that: The classification determination method comprises: When the first acquisition node is at the midpoint of the identification area, the product of twice the number of acquisition nodes and a fixed value is obtained, and the range of the effective area is determined.

4. The intelligent human-computer interaction method for dual-mode data processing according to claim 2, characterized in that: The classification determination method comprises: When the first acquisition node is not at the midpoint of the identification area, a third acquisition node is acquired, the third acquisition node is between the first acquisition node and the second acquisition node, and at least one of the acquisition nodes is arranged between the third acquisition node and the first acquisition node, and two third acquisition nodes are arranged; The number of collection nodes between the two third collection nodes and the first collection node and the second collection node is obtained respectively, and the range of the effective area is determined.

5. The intelligent human-computer interaction method for dual-mode data processing according to claim 1, characterized in that: The priority processing queue includes: Acquire user behavior data of users that can be triggered and identified in the identification area, and associate the user behavior data with a time axis to form evaluation benchmark data; Get real-time user behavior data; A comparison method is constructed, which outputs first priority data and second priority data based on real-time user behavior data and evaluation benchmark data.

6. The intelligent human-computer interaction method for dual-mode data processing according to claim 5, characterized in that: The comparison method comprises: Construct a time-based user behavior data type axis; Input the evaluation benchmark data and real-time user behavior data into the user behavior data type coordinate axis respectively; And form a judgment benchmark curve and a real-time user behavior curve; Compare the arc point slopes based on the user behavior data type in the judgment reference curve and the real-time user behavior curve, extract the arc point types in the real-time user behavior curve whose arc point slopes are greater than or equal to the arc point slopes of the judgment reference curve, assign high priority permissions, and output the first priority data and the second priority data in sequence.

7. The intelligent human-computer interaction method for dual-mode data processing according to claim 1, characterized in that: The inductive model includes: Acquire historical user behavior data, based on the type of the first priority data, search for a user behavior data arrangement that matches the type of the first priority data in the historical user behavior data, read the type of the second priority data, and search for matching second user behavior data in the user behavior data arrangement; The value of the second priority data is overwritten on the second user behavior data and combined with the first priority data to form interaction data for the intelligent human-computer interaction system acting on the dual-mode data processing.

8. The intelligent human-computer interaction method for dual-mode data processing according to claim 1, characterized in that: The user behavior data includes: body movement data, voice data and facial image data, the user behavior data type includes: body, voice and face, the first user behavior data and the second user behavior data are any one or a combination of two of the body movement data, voice data and facial image data, and the first user behavior data and the second user behavior data do not contain repeated user behavior data types.

9. An intelligent human-computer interaction system for dual-mode data processing, using an intelligent human-computer interaction method for dual-mode data processing as claimed in any one of claims 1 to 8, characterized in that: include: An identification area module, including a plurality of collection nodes for identifying user behavior data, is used to obtain a valid area for collecting user behavior data within the identification area; A data preprocessing module, used to perform preliminary processing on the collected raw data, including data cleaning, denoising and format conversion, and identifying and dividing the first user behavior data and the second user behavior data; A priority processing module, for constructing a priority processing queue according to a preset priority processing rule, prioritizing the first user behavior data and the second user behavior data, and assigning different priority permissions to different data based on the arc point slope of the data type by comparing the real-time user behavior data with the evaluation benchmark data; The data summarization and integration module is used to take the first priority data, find the second user behavior data associated with it in the historical user behavior data, and perform numerical coverage and integration on the found second user behavior data, and combine it with the first priority data to generate the interaction data that ultimately acts on the system; The interaction execution and feedback module is used to execute corresponding operations or instructions based on the generated interaction data, such as responding to the user's voice, body language and facial instructions.

Images