CN112711392B - Confidence coefficient calculation method for channels of multi-channel interactive system - Google Patents
Confidence coefficient calculation method for channels of multi-channel interactive system Download PDFInfo
- Publication number
- CN112711392B CN112711392B CN202011482079.8A CN202011482079A CN112711392B CN 112711392 B CN112711392 B CN 112711392B CN 202011482079 A CN202011482079 A CN 202011482079A CN 112711392 B CN112711392 B CN 112711392B
- Authority
- CN
- China
- Prior art keywords
- interactive
- data
- channel
- state
- credible
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/16—Sound input; Sound output
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/011—Arrangements for interaction with the human body, e.g. for user immersion in virtual reality
- G06F3/015—Input arrangements based on nervous system activity detection, e.g. brain waves [EEG] detection, electromyograms [EMG] detection, electrodermal response detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/017—Gesture based interaction, e.g. based on a set of recognized hand gestures
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Dermatology (AREA)
- Neurology (AREA)
- Neurosurgery (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- User Interface Of Digital Computer (AREA)
Abstract
The invention discloses a channel confidence calculation method of a multichannel interactive system, which comprises the following steps: in the process of fusing interactive data into interactive semantics by a multi-channel interactive system, storing a data credible state value of the interactive data into a state sliding queue with a fixed length; and calculating the channel confidence based on the interaction data confidence state values in the state sliding queue. The invention adaptively calculates the confidence coefficient of the interactive channel aiming at different multi-channel interactive scenes, and simultaneously solves the problems that the use environment of the interactive system is limited and the use process is complicated because an additional environment sensor needs to be introduced into the interactive system for data acquisition and surrounding natural environment perception in the common channel confidence coefficient calculation technology.
Description
Technical Field
The invention belongs to the field of human-computer interaction, and particularly relates to a channel confidence coefficient calculation method of a multi-channel interaction system.
Background
The multi-channel human-computer interaction is a human-computer interaction mode which considers various novel interaction modes and enables a user to communicate with a computer in various modes. Multi-channel human-computer interaction aims at accomplishing the interaction task with a computer in a parallel cooperation manner by comprehensively using a plurality of interaction function modes of people, namely an effect channel and a sense channel. The technology can improve the interaction experience of the user, and the user can interact with the computer more naturally and efficiently.
In the aspect of multi-channel man-machine interaction semantic fusion, when a plurality of channels of interaction data are processed, a solution method for the channel interaction data to have redundancy conflict is to select the interaction data of the channel which is relatively most reliable according to the reliability of the current interaction channel. The reliability of the interaction channel is measured by the confidence of the interaction channel, and most of the current channel confidence calculation methods are to introduce additional environment sensors, wherein the environment sensors are used for collecting the surrounding natural environment data in real time and calculating the channel confidence by using the collected environment data. However, the method for calculating the channel confidence by introducing an additional environment sensor may cause the usage environment of the interactive system to be limited due to the large number of devices required by the system, and the arrangement and usage process of the interactive system is cumbersome, and at the same time, additional system performance overhead and waste may be caused.
Disclosure of Invention
The invention aims to provide a channel confidence coefficient calculation method of a multi-channel interaction system.
The technical scheme for realizing the purpose of the invention is as follows: a channel confidence calculation method of a multi-channel interactive system comprises the following specific steps:
in the process of fusing interactive data into interactive semantics by a multi-channel interactive system, storing a data credible state value of the interactive data into a state sliding queue with a fixed length;
channel confidence is calculated based on the interaction data trusted state values in the state slide queue.
Preferably, the data trust status value of the interaction data comprises:
acquiring a credible state value of interactive data acquired by equipment;
fusing the credible state value of the interactive data corresponding to the failed interactive semantics in the fusion process;
and the multi-channel interactive scene module is used for reflecting the credible state value of the interactive data of the execution result of the interactive semantics after the successful fusion.
Preferably, in the process of fusing the interactive data into the interactive semantics by the multi-channel interactive system, the specific step of storing the data trusted state value of the interactive data into the fixed-length state sliding queue is as follows:
acquiring user interaction data by adopting data acquisition equipment and sending the user interaction data to a main control computer;
the main control computer judges whether the interactive data is credible according to the interactive word set and stores the credible state of the data into a state sliding queue;
and the multi-channel interactive data fusion module sends the fused interactive semantics to a multi-channel interactive scene module for execution, judges whether the interactive data contained in the interactive semantics is credible according to an interactive semantic execution result, and stores the credible state value of the interactive data into a state sliding queue.
Preferably, the interactive word set is composed of interactive operation instructions of the multi-channel interactive scene, object names and object attributes included in the multi-channel interactive scene, and button names included in the multi-channel interactive scene.
Preferably, the specific step of calculating the channel confidence based on the data confidence state data in the state sliding queue is as follows:
calculating the number of the credible state values in the current state sliding queue as 1;
counting the number of continuous interactive data states by taking the tail of the state sliding queue as a reference value;
and calculating the channel confidence coefficient according to the number of the credible state values of 1, the continuous number of the interactive data states and the length of the state sliding queue.
Preferably, the number of consecutive states refers to: and taking the last state data of the state sliding queue as a reference value, and traversing and comparing from the reference value to the head direction of the queue until an interactive data state value different from the reference value is encountered, and stopping counting.
Preferably, the specific formula for calculating the channel confidence is as follows:
C＝a*C 1 +b*C 2
wherein a and b are coefficient parameters, a + b is 1,
C 2 ＝tanh(X 2 )，X 1 for the number of the credible state value in the current state sliding queue as 1, X 2 The number of consecutive states, L is the queue length.
Compared with the prior art, the invention has the following remarkable advantages: the credible state value of the interaction data for calculating the channel confidence coefficient is derived from three stages of the fusion interaction semantic process of the multi-channel interaction system, and other environment sensors except for interaction equipment are not required to be introduced into the multi-channel interaction system, so that the problems that the use environment of the multi-channel interaction system is limited by excessive introduction of sensor equipment and the use steps of the system are complicated due to excessive introduction of sensor equipment are avoided; the three stages are respectively positioned at the front stage, the middle stage and the later stage of the interactive semantic fusion process, the credibility state of the interactive primitive is fully confirmed, the channel confidence coefficient is calculated by using a piecewise function formula containing a tanh activation function, the change gradient of the channel confidence coefficient is more accurately reflected, and a more accurate channel confidence value is calculated in real time.
Drawings
Fig. 1 is a block diagram of a multi-channel interactive system on which the present invention is based.
FIG. 2 is a flow chart of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and specific embodiments.
As shown in fig. 1 and 2, a channel confidence calculation method for a multi-channel interactive system, for the multi-channel interactive system, includes: the system comprises data acquisition equipment, a main control computer, a multi-channel interactive data fusion algorithm module, a multi-channel interactive scene module and multi-channel interactive scene display equipment. The method comprises the following specific steps:
in the process of fusing interactive data into interactive semantics by a multi-channel interactive system, storing a data credible state value of the interactive data into a state sliding queue with a fixed length;
channel confidence is calculated based on the interaction data trusted state values in the state slide queue.
In one embodiment, in a process of fusing interactive data into interactive semantics by a multi-channel interactive system, a specific method for storing a data trusted state value of the interactive data into a fixed-length state sliding queue includes:
acquiring user interaction data by adopting data acquisition equipment and sending the user interaction data to a main control computer;
in some embodiments, the user interaction data comprises audio data, user arm electromyography data, and user palm gesture data;
the main control computer judges whether the interactive data is credible according to the interactive word set and stores the credible state of the data into a state sliding queue;
specifically, the multi-channel interactive data fusion algorithm module performs interactive semantic fusion by using interactive data, takes interactive data corresponding to interactive semantics which are failed in fusion in the fusion process as untrusted interactive data, sets the trusted state value of the untrusted interactive data to 0, takes interactive data corresponding to interactive semantics which are successfully fused in the fusion process as trusted interactive data, sets the trusted state value of the trusted interactive data to 1, and updates the trusted state value into a state sliding queue.
In some embodiments, the interactive word set is composed of all the interactive operation instructions of the multi-channel interactive scene, the object names and the object attributes contained in the multi-channel interactive scene, and the button names contained in the multi-channel interactive scene. And the main control computer determines whether the received interactive data is contained in the interactive word set as the basis for determining whether the interactive data is credible. If the interactive data is contained in the interactive word set, setting the credible state value of the interactive data to be 1; and if the interactive data is not contained in the interactive word set, setting the credible state value of the interactive data to be 0.
The multi-channel interactive data fusion module sends the fused interactive semantics to a multi-channel interactive scene module for execution, judges whether the interactive data contained in the interactive semantics is credible according to an interactive semantics execution result, and sets an interactive data credibility state value constituting the interactive semantics to 1 if the interactive semantics is successfully executed; and if the interactive semantics are not successfully executed, setting the interactive data credible state value of the interactive semantics which are not successfully executed to 0, and updating the interactive data credible state value into the state sliding queue.
In one embodiment, the specific steps of calculating the channel confidence based on the data confidence state data in the state sliding queue are as follows:
calculating the number of the credible state values in the current state sliding queue as 1, and recording as X 1 . And the number of data with the value of 1 in the state sliding queue is the number of the interactive data representing the credible state.
Counting the number of continuous interactive data states by taking the tail of the state sliding queue as a reference value, and recording as X 2 。
The number of state successes refers to: taking the last state data of the state sliding queue as a reference value, traversing and comparing from the reference value to the head direction of the queue until an interactive data state value different from the reference value is met, stopping counting, and recording the continuous number of recorded states as X 2 。
According to the number X of the credible states of the interactive data 1 Number of successive states X 2 And calculating the channel confidence coefficient C according to the state sliding queue length L.
The channel confidence C is composed of two parts, the first part C 1 By number X of trusted states of the interactive data 1 And queue length L, as follows:
second part C 2 Number X of successive states 2 Calculated substituted into the tanh activation function:
C 2 ＝tanh(X 2 )
wherein, tan h (X) 2 ) The function is:
the channel confidence C is calculated by the formula:
C＝C 1 +C 2
wherein, C 1 Increasing and decreasing along with the increase and decrease of the number of times of adopting the interactive data, and the method is used for the confidence coefficient of a linear reaction channel; c 2 The channel confidence degree condition at the latest moment is partially reflected by the continuous number of states, and in order to control the channel confidence degree value to change in a smaller range and be easy to observe, the channel confidence degree value is respectively in C 1 、C 2 Two parts are preceded by a coefficient parameter, expressed as:
C＝a*C 1 +b*C 2
in the formula:
a+b＝1
derived from the above formula:
the invention adaptively calculates the confidence coefficient of the interactive channel aiming at different multi-channel interactive scenes, and simultaneously solves the problems that the use environment of the interactive system is limited and the use process is complicated because an additional environment sensor needs to be introduced into the interactive system for data acquisition and surrounding natural environment perception in the common channel confidence coefficient calculation technology.
Examples
The embodiment is based on a multi-channel interactive system, and the multi-channel interactive system comprises a voice data acquisition device, a myoelectric data acquisition device, an inertial data acquisition device, a multi-channel interactive data fusion algorithm module, a main control computer, a multi-channel interactive scene module and a multi-channel interactive scene display device. The multi-channel interactive data fusion algorithm module is arranged on a main control computer, the voice data acquisition equipment, the myoelectric gesture data acquisition equipment, the inertia gesture data acquisition equipment and the multi-channel interactive scene display equipment are all connected with the main control computer, and the multi-channel interactive scene module is arranged on the multi-channel interactive scene display equipment.
In this embodiment, the confidence level calculation of the channel is performed by using the trusted state value of the interactive data acquired by the data acquisition device, the trusted state value of the interactive data included in the interactive semantics that are failed in fusion in the fusion process, and the trusted state value of the interactive data that is reflected by the execution result of the interactive semantics that is successfully fused by the electronic sand table interaction module.
The interactive data collected by the data collecting device comprises: the voice data acquisition equipment records the audio data, identifies the audio data, and transmits the obtained interactive data to the main control computer. The myoelectric data acquisition equipment acquires myoelectric data of the arm of the user, recognizes user interaction gestures, and transmits the acquired interaction data to the main control computer. The inertial data acquisition equipment acquires palm posture data of the user, identifies the palm posture of the user and transmits the acquired interactive data to the main control computer.
Finally, the formula for performing confidence calculation of the channel is:
Claims (4)
1. a channel confidence calculation method of a multi-channel interactive system is characterized by comprising the following specific steps:
in the process of integrating interactive data into interactive semantics by a multi-channel interactive system, storing a data credible state value of the interactive data into a state sliding queue with a fixed length, wherein the data credible state value of the interactive data comprises:
acquiring a credible state value of interactive data acquired by equipment;
fusing the credible state value of the interactive data corresponding to the failed interactive semantics in the fusion process;
the multi-channel interactive scene module is used for reflecting the credible state value of the interactive data of the execution result of the interactive semantics after the successful fusion;
calculating a channel confidence based on the interactive data credible state value in the state sliding queue, and specifically comprising the following steps:
calculating the number of the credible state values in the current state sliding queue as 1;
counting the number of continuous interactive data states by taking the tail of the state sliding queue as a reference value;
calculating the channel confidence according to the number of the credible state values as 1, the continuous number of the interactive data states and the length of the state sliding queue, wherein the specific formula is as follows:
C＝a*C 1 +b*C 2
wherein a and b are coefficient parameters, a + b is 1,
C 2 ＝tanh(X 2 )，X 1 for the number of the credible state value in the current state sliding queue as 1, X 2 The number of consecutive states, L is the queue length.
2. The channel confidence calculation method of the multi-channel interactive system according to claim 1, wherein in the process of fusing the interactive data into the interactive semantics by the multi-channel interactive system, the specific step of storing the data confidence state value of the interactive data into the fixed-length state sliding queue is:
acquiring user interaction data by adopting data acquisition equipment and sending the user interaction data to a main control computer;
the main control computer judges whether the interactive data is credible according to the interactive word set and stores the credible state of the data into a state sliding queue;
and the multi-channel interactive data fusion module sends the fused interactive semantics to a multi-channel interactive scene module for execution, judges whether the interactive data contained in the interactive semantics is credible according to an interactive semantic execution result, and stores the credible state value of the interactive data into a state sliding queue.
3. The channel confidence computation method of a multi-channel interactive system according to claim 2, wherein the interactive word set is composed of the interactive operation commands of the multi-channel interactive scene, the object names and object attributes included in the multi-channel interactive scene, and the button names included in the multi-channel interactive scene.
4. The channel confidence calculation method of the multi-channel interactive system according to claim 1, wherein the number of the continuous states is: and traversing and comparing the last state data of the state sliding queue from the reference value to the head direction of the queue by taking the last state data of the state sliding queue as the reference value until an interactive data state value different from the reference value is met, and stopping counting.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011482079.8A CN112711392B (en) | 2020-12-15 | 2020-12-15 | Confidence coefficient calculation method for channels of multi-channel interactive system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011482079.8A CN112711392B (en) | 2020-12-15 | 2020-12-15 | Confidence coefficient calculation method for channels of multi-channel interactive system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112711392A CN112711392A (en) | 2021-04-27 |
| CN112711392B true CN112711392B (en) | 2022-09-27 |
Family
ID=75542162
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011482079.8A Active CN112711392B (en) | 2020-12-15 | 2020-12-15 | Confidence coefficient calculation method for channels of multi-channel interactive system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112711392B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107122109A (en) * | 2017-05-31 | 2017-09-01 | 吉林大学 | A kind of multi-channel adaptive operating method towards three-dimensional pen-based interaction platform |
| CN111665941A (en) * | 2020-06-07 | 2020-09-15 | 济南大学 | Multi-mode semantic fusion man-machine interaction system and method for virtual experiment |
-
2020
- 2020-12-15 CN CN202011482079.8A patent/CN112711392B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107122109A (en) * | 2017-05-31 | 2017-09-01 | 吉林大学 | A kind of multi-channel adaptive operating method towards three-dimensional pen-based interaction platform |
| CN111665941A (en) * | 2020-06-07 | 2020-09-15 | 济南大学 | Multi-mode semantic fusion man-machine interaction system and method for virtual experiment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112711392A (en) | 2021-04-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6635049B2 (en) | Information processing apparatus, information processing method and program | |
| EP1408443B1 (en) | Method and apparatus for analysing gestures produced by a human, e.g. for commanding apparatus by gesture recognition | |
| US20180210552A1 (en) | Haptic conversion system using segmenting and combining | |
| CN111566730A (en) | Voice command processing in low power devices | |
| EP2570901A1 (en) | Method and mobile terminal for automatically recognizing gesture | |
| EP3441874B1 (en) | Scene sound effect control method, and electronic device | |
| CN109669663B (en) | Method and device for acquiring range amplitude, electronic equipment and storage medium | |
| CN112382285B (en) | Voice control method, voice control device, electronic equipment and storage medium | |
| EP3639748A1 (en) | System and method for monitoring pathological breathing patterns | |
| US10770077B2 (en) | Electronic device and method | |
| US20220152476A1 (en) | Method and device for processing information in game, storage medium and electronic device | |
| US20210248359A1 (en) | Dynamic Motion Detection Method and Apparatus, and Storage Medium | |
| EP3139377A1 (en) | Guidance device, guidance method, program, and information storage medium | |
| CN108172221A (en) | The method and apparatus of manipulation aircraft based on intelligent terminal | |
| CN111158487A (en) | Man-machine interaction method for interacting with intelligent terminal by using wireless earphone | |
| JP2016181018A (en) | Information processing system and information processing method | |
| WO2017000772A1 (en) | Front-end audio processing system | |
| CN112711392B (en) | Confidence coefficient calculation method for channels of multi-channel interactive system | |
| CN104345880B (en) | The method and electronic equipment of a kind of information processing | |
| CN114613362A (en) | Device control method and apparatus, electronic device, and medium | |
| CN107112018A (en) | Information processor, information processing method and program | |
| WO2021184732A1 (en) | Audio packet loss repairing method, device and system based on neural network | |
| CN113380229A (en) | Voice response speed determination method, related device and computer program product | |
| WO2024077873A1 (en) | Virtual-scene scaling method and apparatus, and storage medium and electronic device | |
| JPWO2020071200A1 (en) | Information processing equipment, information processing methods, and programs |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |