CN117520892A - Method, device, equipment, vehicle and medium for processing vehicle data - Google Patents
Method, device, equipment, vehicle and medium for processing vehicle data Download PDFInfo
- Publication number
- CN117520892A CN117520892A CN202311493648.2A CN202311493648A CN117520892A CN 117520892 A CN117520892 A CN 117520892A CN 202311493648 A CN202311493648 A CN 202311493648A CN 117520892 A CN117520892 A CN 117520892A
- Authority
- CN
- China
- Prior art keywords
- cabin
- detection
- value
- person
- vehicle
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 74
- 238000012545 processing Methods 0.000 title claims abstract description 67
- 238000001514 detection method Methods 0.000 claims abstract description 432
- 230000015654 memory Effects 0.000 claims description 17
- 238000004891 communication Methods 0.000 claims description 8
- 238000012795 verification Methods 0.000 claims description 8
- 239000002775 capsule Substances 0.000 claims description 3
- 238000003672 processing method Methods 0.000 abstract description 8
- 230000005484 gravity Effects 0.000 description 9
- 238000007781 pre-processing Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 5
- 230000003321 amplification Effects 0.000 description 3
- 238000003199 nucleic acid amplification method Methods 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000007958 sleep Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 230000005236 sound signal Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V13/00—Manufacturing, calibrating, cleaning, or repairing instruments or devices covered by groups G01V1/00 – G01V11/00
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/217—Validation; Performance evaluation; Active pattern learning techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Biology (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Geophysics (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Manufacturing & Machinery (AREA)
- Fittings On The Vehicle Exterior For Carrying Loads, And Devices For Holding Or Mounting Articles (AREA)
Abstract
The embodiment of the invention relates to the technical field of auxiliary dialogue, and discloses a vehicle data processing method, device, equipment, a vehicle and a medium, wherein the method comprises the following steps: acquiring detection data of the first detection equipment for detecting the position of personnel in the cabin; determining a detection value of the position in the cabin according to the detection data; the detection value represents the confidence level of a person at the position in the cabin; acquiring a detection result indicating that the second detection device recognizes that a person exists at the position in the cabin; and checking the detection result according to the detection value, and comprehensively determining whether a person exists at the position in the cabin. The invention can combine various detection devices to detect the position of personnel in the cabin, and can improve the accuracy of personnel detection; in addition, the detection value of each cabin position can be determined in advance by checking the detection result of the detection value, so that the processing amount required during real-time processing can be reduced, and the real-time performance of vehicle data processing is improved.
Description
Technical Field
The embodiment of the invention relates to the technical field of auxiliary conversations, in particular to a vehicle data processing method, device, equipment, a vehicle and a medium.
Background
Currently, the position of personnel in a cabin needs to be determined in certain scenes; for example, it is necessary to determine whether a passenger is present in the rear seat of the vehicle, or the like. At present, a camera in a vehicle can be used for detecting the position of a person, but the environment in a cabin is sometimes complex, and the position of the person is likely to change to a certain extent, so that the interference in the position detection process is more, and the position identification precision is lower.
Disclosure of Invention
In view of the above problems, the embodiments of the present invention provide a method, an apparatus, a device, a vehicle, and a medium for processing vehicle data, which are used to solve the problem that the existing in-cabin personnel position recognition accuracy is low.
According to a first aspect of an embodiment of the present invention, there is provided a method for processing vehicle data, the method including:
acquiring detection data of the first detection equipment for detecting the position of personnel in the cabin;
determining a detection value of the position in the cabin according to the detection data; the detection value represents the confidence level of a person at the position in the cabin;
acquiring a detection result indicating that the second detection device recognizes that a person exists at the position in the cabin;
and checking the detection result according to the detection value, and comprehensively determining whether a person exists at the position in the cabin.
According to a second aspect of an embodiment of the present invention, there is provided a processing apparatus for vehicle data, including:
the detection data acquisition module is used for acquiring detection data of the first detection equipment for detecting the position of personnel in the cabin;
the detection value determining module is used for determining a detection value of the position in the cabin according to the detection data; the detection value represents the confidence level of a person at the position in the cabin;
the detection result acquisition module is used for acquiring a detection result indicating that the second detection equipment recognizes that a person exists at the position in the cabin;
and the verification module is used for verifying the detection result according to the detection value and comprehensively determining whether a person exists at the position in the cabin.
According to a third aspect of the embodiment of the present invention, there is provided a processing apparatus of vehicle data, including: the device comprises a memory and a processor, wherein the memory and the processor are in communication connection;
the memory is configured to store at least one executable instruction that causes the processor to perform the operations of the method for processing vehicle data according to the first aspect.
According to a fourth aspect of an embodiment of the present invention, there is provided a vehicle including a first detection apparatus, a second detection apparatus, and a processing apparatus for vehicle data according to the third aspect.
According to a fifth aspect of embodiments of the present invention, there is provided a computer readable storage medium having stored therein at least one executable instruction which when executed by a processor implements the method of processing vehicle data according to the first aspect.
According to the vehicle data processing method, device, equipment, vehicle and medium, the confidence degree of the person at each cabin position is determined based on the first detection equipment, and if the detection result indicating the person at the cabin position is obtained based on the second detection equipment, the detection result can be further checked based on the detection value capable of indicating the confidence degree of the cabin position, so that the confidence degree of the detection result is determined, and whether the cabin position is person or not is determined. The method can be combined with various detection devices to detect the positions of personnel in the cabin, so that the accuracy of personnel detection can be improved; and in addition, the detection value of each cabin position can be determined in advance in a mode of checking the detection result of the detection value, when the second detection equipment detects that a person exists at the cabin position, the detection result of the second detection equipment is verified quickly, the processing amount required during real-time processing can be reduced, and particularly for a processor with lower processing performance, the real-time performance of vehicle data processing can be improved.
The foregoing description is only an overview of the technical solutions of the embodiments of the present invention, and may be implemented according to the content of the specification, so that the technical means of the embodiments of the present invention can be more clearly understood, and the following specific embodiments of the present invention are given for clarity and understanding.
Drawings
The drawings are only for purposes of illustrating embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to designate like parts throughout the figures. In the drawings:
fig. 1 is a schematic diagram showing an arrangement of sound equipment in a vehicle according to the present invention;
fig. 2 is a schematic flow chart of a first embodiment of a method for processing vehicle data according to the present invention;
FIG. 3 illustrates a schematic diagram of partitioning an interior space of a vehicle provided by the present invention;
FIG. 4 is a flow chart of a second embodiment of a method for processing vehicle data according to the present invention;
FIG. 5 is a flow chart of a third embodiment of a method for processing vehicle data according to the present invention;
FIG. 6 is a schematic flow chart of a process for implementing vehicle data according to the present invention;
FIG. 7 is a schematic flow chart for judging whether a person exists at a position in a cabin according to the present invention;
fig. 8 is a schematic structural view showing an embodiment of a processing device for vehicle data provided by the present invention;
fig. 9 is a schematic structural view showing an embodiment of a processing apparatus for vehicle data provided by the present invention.
Detailed Description
Exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present invention are shown in the drawings, it should be understood that the present invention may be embodied in various forms and should not be limited to the embodiments set forth herein.
During driving, there is often a need to have conversations between multiple users in the vehicle, for example, a host driver needs to talk to others. Since there may be disturbances in navigation, multimedia sounds, etc. in the vehicle, this may affect the conversation process, resulting in a poor conversation in the vehicle. Especially when the main driver needs to talk with the rear-row personnel, the main driver needs to increase the speaking sound or return, which can influence the normal driving of the vehicle of the main driver and has potential safety hazard
Some current vehicles are provided with a plurality of groups of sound equipment. Fig. 1 shows a schematic layout of sound equipment in the vehicle. As shown in fig. 1, the vehicle 100 includes four sets of sound devices 110, and one set of sound devices 110 is provided at each seat in the vehicle 100; in fig. 1, the sound equipment 110 in the upper left corner is provided at the main driving seat, the sound equipment 110 in the upper right corner is provided at the sub driving seat, the sound equipment 110 in the lower left corner is provided at the rear left seat (rear left side seat), and the sound equipment 110 in the lower right corner is provided at the rear right seat (rear right side seat).
And, each group of sound devices 110 includes at least one in-vehicle sound collection apparatus and at least one in-vehicle sound playback apparatus; as shown in fig. 1, the sound equipment 110 in the upper left corner includes an in-vehicle sound collection device 111 and an in-vehicle sound playback device 112, and the sound equipment 110 in the upper left corner includes an in-vehicle sound collection device 113 and an in-vehicle sound playback device 114. The vehicle-mounted sound collecting device may be, for example, a microphone array, etc., and the vehicle-mounted sound playing device may be, for example, a speaker, an earphone, etc., which may be specifically determined based on actual requirements.
In order to accurately play sound to a person in the cabin using the in-vehicle sound playing device, it is necessary to detect the position of the person in the cabin relatively accurately. The embodiment of the invention provides a vehicle data processing method, which is used for detecting the position of personnel in a cabin based on two detection devices, checking the detection result determined by one detection device based on the confidence coefficient determined by the other detection device, realizing mutual verification and comprehensively determining the position of the personnel in the cabin.
FIG. 2 shows a flow chart of a first embodiment of a method of processing vehicle data of the present invention, which is applicable to a vehicle, such as the vehicle having multiple sets of sound devices shown in FIG. 1; in particular, the method may be applied to a controller of a vehicle, such as an intelligent cabin controller (CDC). As shown in fig. 2, the method includes the following steps 201 to 204.
Step 201: and acquiring detection data of the first detection equipment for detecting the position of personnel in the cabin.
In this embodiment, a plurality of devices capable of detecting the state in the vehicle generally exist in the vehicle cabin, and some of the devices can detect the position of personnel in the cabin, that is, detect where the personnel are in the cabin; for convenience of description, a part of the devices capable of realizing the position detection will be referred to as a first detection device. The first detection device can generate corresponding detection data when performing position detection. It is understood that the detection data belongs to a kind of vehicle data.
For example, an in-vehicle image acquisition device is arranged in the vehicle cabin, the in-vehicle image acquisition device can acquire an in-vehicle image, whether personnel exist in the vehicle or not and the position of the personnel in the vehicle can be identified by performing image processing on the in-vehicle image, and the in-vehicle image is detection data acquired by the first detection equipment. For example, the in-vehicle image capture device may be a camera of an OMS (Occupancy Monitoring System, an in-vehicle passenger monitoring system) based on which in-vehicle images may be captured.
Step 202: determining a detection value of the position in the cabin according to the detection data; the detection value indicates the confidence of a person at the position in the cabin.
In this embodiment, after the detection data is acquired, the position of the personnel in the cabin may be determined based on the detection data. Wherein, divide the vehicle cabin space into a plurality of cabin positions, regard cabin position as the unit, confirm whether there is the person in each cabin position department respectively, or stated, confirm that the cabin personnel is located which cabin position.
Specifically, a binary result indicating whether or not a person is present is not simply generated based on the detection data, but is expressed in the form of a confidence level as to whether or not a person is present at the corresponding in-cabin location; the greater the confidence, the higher the likelihood of someone at the location within the cabin. For example, if the confidence of the in-cabin location is close to 0, it indicates that there is a high probability that there is no person at that in-cabin location; if the confidence of the in-cabin location is close to 1, this indicates a high probability of someone at that in-cabin location.
Wherein the confidence of the person at the position in the cabin is represented by the detection value. For example, the detection value may be a confidence level, or may be a multiple value of the confidence level (for example, 2 times the confidence level), and the expression form of the detection value is not limited in this embodiment. Similar to the confidence, the higher the detection value of the in-cabin location, the more likely it is that the in-cabin location is.
For example, the number of zones may be determined according to the number of seats in the vehicle, thereby dividing the cabin space into a corresponding number of cabin locations. For example, for a 5-seat vehicle, five zones may be divided; for a 7-seat vehicle, seven zones may be divided. Fig. 3 shows a schematic diagram of the division of five areas.
As shown in fig. 3, five areas are divided: a main driving area 301, a sub driving area 302, a rear left area 303, a rear middle area 304, and a rear right area 305. The main driving area 301 corresponds to an area where a main driving seat is located, the auxiliary driving area 302 corresponds to an area where an auxiliary driving seat is located, the rear left area 303 corresponds to an area where a rear left-side seat is located, the rear middle area 304 corresponds to an area where a rear right-side seat is located, and the rear right area 305 corresponds to an area where a rear right-side seat is located. It will be appreciated that by determining whether personnel are present in each of the five zones, one can determine the confidence of the presence of personnel in the respective zone.
Step 203: a detection result is obtained indicating that the second detection device has identified a person at the position in the cabin.
In this embodiment, the position of the person in the cabin may also be detected based on the second detection device, so that it is also possible to detect where the person is in the cabin. The first detection device and the second detection device are different devices, and the first detection device and the second detection device are used for detecting the position of the person based on different modes.
In particular, the second detection device is specifically configured to detect whether a person is present at a certain or certain in-cabin location, and if a person is detected at a certain in-cabin location, a corresponding detection result may be generated, i.e. the detection result may indicate that a person is present at the in-cabin location. If no person is detected in the cabin, no subsequent processing may be performed; although it is also possible to generate a detection result indicating that there is no person at the position in the cabin at this time, the present embodiment mainly considers the former, that is, mainly considers a detection result indicating that there is a person at the position in the cabin. It will be appreciated that the test results also pertain to a vehicle data.
Step 204: and checking the detection result according to the detection value, and comprehensively determining whether a person exists at the position in the cabin.
In this embodiment, if it is determined that a person is located at a position in a certain cabin based on the second detection device, the detection result is checked according to the detection value at the position in the cabin, so that accuracy or reliability of the detection result can be determined. If the detection result is considered to be more reliable based on the detection value of the position in the cabin, the detection result obtained based on the second detection device is considered to be correct, namely, a person is present at the position in the cabin; if the reliability of the detection result is considered to be low based on the detection value of the cabin interior position, the second detection device can be considered to have false detection, namely the detection result obtained based on the second detection device is wrong, namely no person exists at the cabin interior position.
For example, if the detection result indicates that a person is present at a certain in-cabin position, and the detection value at the in-cabin position is high, that is, the confidence of the presence of a person at the in-cabin position is high, the detection result can be considered to be correct at this time. Conversely, if the detection result indicates that a person is present at a certain in-cabin location, and the detection value at the in-cabin location is low, that is, the confidence that a person is present at the in-cabin location is low, then the detection result may be considered to be erroneous.
In general, the detection accuracy of the first detection device is greater than that of the second detection device, so that the detection result of the second detection device with low accuracy can be verified by using the first detection device with high accuracy.
According to the vehicle data processing method provided by the embodiment, the confidence degree of the person at each cabin position is determined based on the first detection equipment, and if the detection result indicating the person at the cabin position is obtained based on the second detection equipment, the detection result can be further checked based on the detection value capable of indicating the confidence degree of the cabin position, so that the confidence degree of the detection result is determined, and whether the person is at the cabin position is determined. The method can be combined with various detection devices to detect the positions of personnel in the cabin, so that the accuracy of personnel detection can be improved; and in addition, the detection value of each cabin position can be determined in advance in a mode of checking the detection result of the detection value, when the second detection equipment detects that a person exists at the cabin position, the detection result of the second detection equipment is verified quickly, the processing amount required during real-time processing can be reduced, and particularly for a processor with lower processing performance, the real-time performance of vehicle data processing can be improved.
FIG. 4 shows a flow chart of a second embodiment of a method of processing vehicle data of the present invention, which is applicable to a vehicle, such as the vehicle having multiple sets of sound devices shown in FIG. 1; in particular, the method may be applied to a controller of a vehicle, for example, an intelligent cabin controller. As shown in fig. 4, the method includes the following steps 401 to 404.
Step 401: and acquiring detection data of the first detection equipment for detecting the position of personnel in the cabin.
In this embodiment, the first detection device may include: a seat state detection device and/or an in-vehicle image acquisition device. Accordingly, the detection data may include: the seat sensor state and the safety belt state acquired by the seat state detection device and/or the in-vehicle image acquired by the in-vehicle image acquisition device. The seat state detection device may include a seat sensor and a sensor for detecting a state of the seat belt.
For example, the seat state detection device includes a seat sensor (e.g., a gravity sensor, a pressure sensor, etc.) and a sensor for detecting a seat belt state (e.g., a seat belt buckle sensor), and the seat state detection device may acquire the seat sensor state and the seat belt state; for example, the seat sensor state may indicate the weight of an object on the seat, the presence of a person on the seat, or the like, and the seat belt state may indicate whether the seat belt buckle is buckled, or the like.
The detection data collected by the in-vehicle image collection device can be an original image which is currently shot, or the in-vehicle image collection device can also analyze based on the original image so as to identify personnel in the original image and the positions of the personnel in the vehicle.
Wherein, if the first detection device includes a seat state detection device, the detection data may include: the seat sensor state and the safety belt state are acquired by the seat state detection device; if the first detection device includes an in-vehicle image acquisition device, the detection data may include: the in-vehicle image acquisition device acquires an in-vehicle image; if the first detection apparatus includes a seat state detection device and an in-vehicle image acquisition device, the detection data may include: the seat sensor state and the safety belt state acquired by the seat state detection device and the in-vehicle image acquired by the in-vehicle image acquisition device.
In this embodiment, after a person gets on the vehicle, the position of the person in the vehicle generally does not change, so the position detection results of the seat state detection device and the in-vehicle image acquisition device generally also do not change, so in this embodiment, the seat state detection device and the in-vehicle image acquisition device are used as the first detection device, so that the person in each cabin position is conveniently detected in advance, so as to determine the confidence level of the person in each cabin position. It will be appreciated that the detection value for each intrabay location is generally constant.
The process of determining whether a person is in the cabin based on the detection data (e.g., the seat sensor state and the seat belt state) collected by the seat state detection device is referred to as "seat recognition", and the process of determining whether a person is in the cabin based on the detection data (e.g., the in-vehicle image) collected by the in-vehicle image acquisition device is referred to as "face recognition", for convenience of description.
Step 402: determining a detection value of the position in the cabin according to the detection data; the detection value indicates the confidence of a person at the position in the cabin.
Specifically, the above-described step 402 "determining the detection value of the in-cabin position from the detection data" may include the following steps 4021 to 4022.
Step 4021: and respectively determining whether a person exists at the position in the cabin according to each item of data in the detection data.
In the present embodiment, as described above, the detection data may include a plurality of items of data, such as a seat sensor state, a seat belt state, an in-vehicle image, and the like, and whether or not a person is present at a certain in-cabin position may be determined based on each item of data, respectively.
Specifically, for a certain in-cabin position (such as the main driving area 301 shown in fig. 3, etc.), whether a person is present at the in-cabin position may be determined based on a seat sensor state acquired by a seat sensor at the in-cabin position, whether a person is present at the in-cabin position may be determined based on a seat belt state at the in-cabin position, and whether a person is present at the in-cabin position may be determined by detecting whether a face is present at the in-cabin position based on an image of the in-cabin position in an in-vehicle image.
Step 4022: under the condition that the position in the cabin is determined to be someone, performing value-added processing on the detection value of the position in the cabin; the detection value of the position in the cabin is initially 0.
In this embodiment, a detection value of 0 is preset for each in-cabin position, and if it is determined that a person exists at a certain in-cabin position based on the detection data collected by the first detection device, value-added processing may be performed on the detection value at the in-cabin position, that is, the detection value is increased, so as to improve the confidence that a person exists at the in-cabin position. Wherein, because whether the position in the cabin is someone can be judged based on a plurality of items of data respectively, each item of data indicates that the position in the cabin is someone, a value-added process can be executed once; if the plurality of data items each indicate a person at the in-cabin location, then accumulation may be performed.
In some alternative embodiments, the step 4022 "performing the value-added processing on the detected value of the in-cabin position in the case where it is determined that the in-cabin position is a person" may specifically include the following steps A1 to A4.
Step A1: in the case where the seat sensor state indicates that a person is present at the in-cabin position, a first value is added to the detection value of the in-cabin position.
Step A2: in the case where the seat sensor state indicates that a person is present at the cabin interior position and then the seat belt state is attached, a second value is added to the detection value of the cabin interior position.
Step A3: when the state of the safety belt is tied and then the state of the seat sensor indicates that a person exists at the position in the cabin, a third numerical value is added to the detection value of the position in the cabin; the third value is less than or equal to the first value < the second value.
In this embodiment, a detection value of 0 is set for each cabin interior position, and when it is determined that a person may exist in the cabin interior position based on seat recognition or face recognition, a corresponding value is added to the detection value.
Specifically, if the seat sensor state indicates that there is a person, for example, the gravity sensor of the seat detects that there is a sufficiently heavy object on the seat, this indicates that there is a person on the seat, and a certain value, that is, the first value, may be added to the detected value. For example, the first value may be 1, 2, etc., and may be specific based on the actual situation.
If the seat sensor state indicates that a person is present and then the seat belt state is attached, it is considered that a person is sitting on the seat and then the seat belt is attached, and the detected value may be increased by a predetermined value, that is, a second value. The second value is greater than the first value because greater confidence is present in determining that a person is present in the seat; for example, the second value may be 2, 3, etc. Wherein, after the state of the seat sensor indicates that a person is present, a first numerical value can be added to the detection value; after detecting that the belt condition is on, the detected value may be increased by a related value, and the first value+the related value=the second value.
If the seat belt status is on, and then the seat sensor status indicates that there is a person, i.e., the seat belt buckle is on before the seat sensor is displayed, at this time, although there is a possibility that there is a person, the reliability at this time is low, so that a third value is added to the detection value at this time; and the third value is less than the second value. The third value may be the same as the first value or the third value may be less than the first value. That is, the third value is less than or equal to the first value < the second value.
Step A4: in the case where the in-vehicle image indicates that a person is present at the in-cabin position, a fourth value is added to the detected value of the in-cabin position.
In this embodiment, if a person is determined to be present at the in-cabin position based on face recognition, a certain value, that is, a fourth value may be added to the detection value of the in-cabin position. The magnitude relation between the fourth value and the first, second and third values may be specifically determined according to practical situations.
In general, the reliability of seat recognition is higher than that of face recognition, so the fourth value generally does not exceed the second value; for example, the fourth value is less than the second value, or the fourth value is equal to the first value. Or if the accuracy of the in-vehicle image acquisition device is higher, the reliability of face recognition is higher than that of seat recognition, and at this time, the fourth value may be larger than the second value.
It will be appreciated that the conditions in steps A1 to A4 described above are compatible, and that a larger number may be added to the detected value. For example, if the state of the seat sensor indicates that a person is present and the detection data of the in-vehicle image capturing device also indicates that a person is present at the in-cabin position, then the first value and the fourth value may be added to the detection value at the in-cabin position at this time, where the detection value at the in-cabin position is: first value + fourth value.
Alternatively, as described above, the detection value of the in-cabin position is initially 0, which can be updated based on the seat recognition and face recognition; if a door is closed, the position of the person in the cabin may change, and therefore, the detection value needs to be reset to 0 and the detection needs to be repeated. Specifically, before the above-described step 401 of acquiring the detection data of the position detection of the cabin interior personnel by the first detection device, the method may further include the following step B1 or step B2.
Step B1: in the case of detecting the door closing signal, the detection value of the in-cabin position is set to 0, and then the detection value of the in-cabin position is redetermined.
In this embodiment, the door closing operation of the user is taken as the judgment start, if the door closing signal is detected, the personnel seat in the cabin is considered to be basically fixed, at this time, the detection value of each cabin position can be set to 0, and the detection value is updated based on the steps A1 to A4 and the like, so as to determine the detection value of the cabin position. If the door is closed again, that is, if the door closing signal is detected again, the detection value is reset to 0 again, and the detection value of the position in the cabin is determined again.
Step B2: in the case where the seat sensor state changes from indicating that a person is present at the cabin interior position to no person and/or the seat belt state changes from being fastened to unfastened, the detection value of the cabin interior position is set to 0, and thereafter the detection value of the cabin interior position is newly determined.
In this embodiment, after a user gets on a vehicle, the state of the seat sensor is changed from no one to one under normal conditions, the state of the safety belt is changed from unfastened to fastened, and then the state of the seat sensor and the safety belt are kept unchanged, i.e. the updated detection value can be considered unchanged; conversely, if the seat sensor state changes from indicating that a person is present at the cabin interior position to no person and/or the seat belt state changes from being on to off, the position of the cabin interior person may be considered to change or will change, at which time the detection value needs to be set to 0 and the detection value of the cabin interior position is redetermined.
Step 403: a detection result is obtained indicating that the second detection device has identified a person at the position in the cabin.
The detailed description can refer to step 203 shown in fig. 2, and will not be repeated here.
Step 404: and checking the detection result according to the detection value, and comprehensively determining whether a person exists at the position in the cabin.
The detailed description can refer to step 204 shown in fig. 2, and will not be repeated here.
According to the vehicle data processing method provided by the embodiment, the detection value of the position in the cabin can be predetermined based on seat recognition and face recognition, so that the subsequent verification based on the detection value is facilitated. By using each item of the detection data, whether a person exists at the position in the cabin can be respectively determined, and then the detection value which is initially 0 is updated, and the detection value of the position in the cabin can be simply determined by combining seat identification and face recognition.
FIG. 5 shows a flow chart of a third embodiment of a method of processing vehicle data of the present invention, which is applicable to a vehicle, such as the vehicle having multiple sets of sound devices shown in FIG. 1; in particular, the method may be applied to a controller of a vehicle, for example, an intelligent cabin controller. As shown in fig. 5, the method includes the following steps 501 to 504.
Step 501: and acquiring detection data of the first detection equipment for detecting the position of personnel in the cabin.
The detailed description will refer to step 201 shown in fig. 2, and will not be repeated here.
Step 502: determining a detection value of the position in the cabin according to the detection data; the detection value indicates the confidence of a person at the position in the cabin.
The detailed description may refer to step 202 shown in fig. 2 or step 402 shown in fig. 4, and will not be repeated here.
Step 503: and positioning according to the voice of the personnel in the cabin acquired by the vehicle-mounted sound acquisition device, determining the position of the personnel in the cabin, and generating a corresponding detection result.
In this embodiment, the second detection device includes: and the vehicle-mounted sound acquisition device. The vehicle-mounted sound collection device is a device capable of collecting sound as described above, such as the vehicle-mounted sound collection device 111 and the vehicle-mounted sound collection device 113 shown in fig. 1. The vehicle-mounted sound collection device can identify where a sound source exists, and then the position where the sound source exists is used as the position with personnel. For example, the vehicle-mounted sound collecting device is a microphone array, and the position of a sound source can be identified by utilizing the voice difference of personnel in the cabin collected by different microphones in the microphone array, so that the sound source is positioned, and the position of the personnel in the cabin making the sound is determined. Or, because a plurality of vehicle-mounted sound collecting devices exist in the vehicle, for example, each seat is provided with a corresponding vehicle-mounted sound collecting device, the person at which seat can sound can be determined based on the intensity of the sound collected by different vehicle-mounted sound collecting devices; for example, the vehicle-mounted sound collecting device located in the main driver and the vehicle-mounted sound collecting device located in the auxiliary driver collect sound signals with the same content, but the sound intensity collected by the vehicle-mounted sound collecting device in the main driver is obviously greater than that collected by the vehicle-mounted sound collecting device in the auxiliary driver, and at the moment, people in the seat of the main driver can be determined to exist and the people are speaking.
Correspondingly, the voice of the personnel in the cabin collected by the vehicle-mounted sound collecting device can comprise related data of the whole sound field in the vehicle, or can also only comprise sound data of the corresponding seat, and then sound source positioning can be realized based on the voice of the personnel in the cabin.
Wherein, because the personnel in the cabin do not always send out the voice, namely the vehicle-mounted sound collecting device cannot always collect the voice of the personnel in the cabin; in other words, only when the personnel in the cabin sound, the vehicle-mounted sound collecting device can collect the voice of the personnel in the cabin, and then the personnel position detection can be performed. Therefore, the detection value of the in-cabin position can be determined in advance, for example, the seat state detection device and the in-vehicle image acquisition device are taken as the first detection device, so that the detection value of the in-cabin position can be determined in advance; and when the personnel in the cabin make sound, the personnel position detection is carried out based on the voice of the personnel in the cabin acquired by the vehicle-mounted sound acquisition device, so that the identification of the personnel in the cabin is realized.
In this embodiment, for convenience of description, a process of judging whether a person is present at the position in the cabin based on the detection data of the in-vehicle sound collection device is referred to as "voice recognition".
The person in the cabin may make a plurality of sounds, i.e. there may be a plurality of speech recognition, so that each speech recognition may determine the position of the speaker, e.g. determine the position of the speaker, based on the previously determined detection value of the position in the cabin. The method determines the detection value of the position in the cabin based on seat recognition and face recognition, and further judges based on voice recognition, so that the accuracy of a recognition result can be ensured, and the method is applicable to the condition of multiple voice recognition.
For example, after a user gets on a vehicle, a detection value of each in-cabin position may be determined based on seat recognition and face recognition, where each seat corresponds to one in-cabin position, and the detection value of each seat may be determined. After the sound is collected, the voice recognition detection result can be combined to judge so as to determine the position of the speaker in the cabin.
Step 504: and checking the detection result according to the detection value, and comprehensively determining whether a person exists at the position in the cabin.
The detailed description can refer to step 204 shown in fig. 2, and will not be repeated here.
In some alternative embodiments, since the second detection device is the vehicle-mounted sound collection device, the position of the speaker in the cabin can be determined relatively accurately by checking the position of the person in the cabin identified by the vehicle-mounted sound collection device based on the detection value of the position in the cabin. Specifically, the method may further include the following steps C1 to C2.
Step C1: under the condition that people exist at the positions in at least two cabins are comprehensively determined, current voice data acquired by a vehicle-mounted sound acquisition device corresponding to the positions of speakers are acquired; the speaker is located in one of the personnel's cabin locations.
In the embodiment, after the personnel get on the vehicle, voices can be generated in the mutual conversation process, and after the vehicle-mounted sound acquisition device acquires the voices, the personnel in the cabin can be determined to be in conversation based on the verification of the detection value; it will be appreciated that at least two talents are required to be able to conduct a conversation and accordingly, it may be determined that there are at least two in-cabin locations. At this time, the personnel in the cabin corresponding to the collected current voice data are taken as speaking personnel, and the rest of the personnel in the cabin for conversation are taken as receiving personnel. It will be appreciated that the number of speakers and recipients may be one or more; typically, the speaker is one, and the recipient is one or more.
For example, if three persons are currently in the vehicle, a main driving a, a secondary driving B, and a person C sitting in the rear row; if only the primary driver A and the rear-row personnel C are detected to carry out conversation at present, the secondary driver B does not join in the conversation, and the primary driver A and the rear-row personnel C can be used as the conversation personnel; when voice data sent by a main driver A is acquired, the main driver A is a speaking person, and a rear person C is a receiving person; correspondingly, when voice data sent by a rear person C is collected, the main driver A is a receiving person, and the rear person C is a speaking person.
When in-vehicle dialogue, the vehicle-mounted sound collecting device can collect sound emitted by a speaker, so that corresponding voice data are generated; in this embodiment, the voice data collected at the current time is referred to as current voice data. The current voice data can also be used for generating a detection result which indicates which cabin position is occupied by a person, and checking the detection result based on the detection value so as to accurately determine which cabin position is occupied by the speaker.
Step C2: controlling a vehicle-mounted sound playing device corresponding to the position of the receiving person to play the current voice data; the receiving person is located at another personnel capsule location.
In this embodiment, after determining the speaker, the other personnel in the cabin participating in the conversation may be used as the receiving personnel, and further, based on the position of the receiving personnel in the cabin, the corresponding vehicle-mounted sound playing device may be determined, and further, based on the current voice data sent by the speaker, the current voice data sent by the speaker may be played by the vehicle-mounted sound playing device, so that the voice spoken by the speaker may be played to the receiving personnel, and the in-vehicle auxiliary conversation may be implemented. For example, the in-vehicle sound playing device may convert current voice data in the form of a digital signal into an analog signal and perform a playing process on the analog signal, so that a corresponding sound may be played to a receiving person.
For example, if the driver currently performs a dialogue with a person on the right side of the rear row, and as shown in fig. 1, the vehicle-mounted sound collection device 111 in the sound equipment 110 on the upper left corner collects sound, that is, the driver is the speaker, at this time, the person on the right side of the rear row may be taken as the receiving person, and the vehicle-mounted sound playing device corresponding to the position of the receiving person is the vehicle-mounted sound playing device 114 in the sound equipment 110 on the lower right corner, that is, the vehicle-mounted sound playing device 114 is taken as the vehicle-mounted sound playing device corresponding to the position of the receiving person, and controls the vehicle-mounted sound playing device 114 to play the current voice data.
In this embodiment, when a plurality of persons in the vehicle need to perform a conversation, the vehicle-mounted sound collecting device of one group of sound devices can collect the sound emitted by the speaking person and control the vehicle-mounted sound playing device of the other group of sound devices to play the sound, so as to realize the conversation between the two parties. The vehicle-mounted sound playing device with the corresponding position is utilized, so that a receiving person can clearly hear the sound of a speaking person, and the interference of other sounds can be effectively reduced; for example, the in-vehicle sound playing device may directionally play, that is, directionally play, sound corresponding to the current voice data to the receiving person, so that the receiving person may receive the sound uttered by the speaking person more clearly.
And moreover, the speaker sends out the sound to the receiving personnel side through a plurality of sound devices in the vehicle, and compared with the traditional sound propagation in the vehicle space, the interference of the sound in the physical space between the speaker and the receiving personnel can be effectively reduced, the conversation quality can be ensured, and the conversation can be smoothly carried out. In addition, when the main driver and the rear-row personnel are in conversation, the main driver is not required to return, and potential safety hazards can be reduced.
In this embodiment, when an auxiliary dialogue is required, the method can be applied to at least the following various scenarios.
(1) Two persons in the vehicle, namely a main driver and a rear-row person, can carry out voice dialogue based on the method, and the main driver does not need to return to the head, so that the driving safety is ensured; in addition, if the rear-row personnel are persons with smaller speaking sounds such as children, the main driver can be ensured to hear the sounds of the rear-row personnel more clearly.
(2) More than two persons in the vehicle, and the owners in the vehicle can carry out voice dialogue based on the method; it will be appreciated that there are a plurality of recipients at this point.
(3) More than two persons in the vehicle, and the part of the persons in the vehicle can conduct voice dialogue based on the method. For example, three persons in the vehicle, namely a main driver, a rear left person (located on a rear left seat) and a rear right person (located on a rear right seat), can perform voice conversation based on the method, and cannot influence the rear right person. For example, when the right-back person sleeps, the main driver and the left-back person can normally communicate, the right-back person is not a receiving person, and the corresponding vehicle-mounted sound playing device does not play dialogue content, so that the influence on the right-back person can be reduced.
In some alternative embodiments, the denoised and/or amplified sound may be played to the recipient so that the recipient can clearly hear the sound uttered by the speaker. Specifically, the step C2 "controlling the in-vehicle sound playing device corresponding to the position of the receiving person to play the current voice data" may include: controlling a vehicle-mounted sound playing device corresponding to the position of the receiving person to play sound corresponding to the preprocessed current voice data; the preprocessing includes noise reduction processing and/or amplification processing.
In this embodiment, after determining the current voice data, the current voice data may be subjected to preprocessing such as noise reduction processing and/or amplification processing, and the vehicle-mounted sound playing device corresponding to the position of the receiving person plays the preprocessed sound, so as to reduce interference of other sounds, so that the receiving person can hear clear sound.
Wherein the preprocessing process may be performed by a controller for implementing the method; alternatively, the preprocessing may be performed by the in-vehicle sound playback apparatus; alternatively, the preprocessing process may be performed by other components having a sound processing function; alternatively, the controller performs a part of the preprocessing process (e.g., noise reduction processing), the in-vehicle sound playback apparatus performs another part of the preprocessing process (e.g., amplification processing), and the like. The implementation of the preprocessing is not limited in this embodiment.
For example, an AMP (amplifier) capable of amplifying sound is provided in the vehicle, and the controller may transmit current voice data to the AMP, and amplify the current voice data by the AMP. Taking a dialogue between the main driver and the rear-end personnel in the vehicle as shown in fig. 6, the process of implementing the vehicle data may include the following steps 601 to 610.
Step 601: the controller determines that the two parties of the conversation are the main driver and the rear-row personnel.
Wherein the in-cabin location of the in-cabin personnel participating in the conversation may be determined to determine who is the conversation personnel. For example, if only main driving personnel and rear-row personnel exist in the vehicle, the main driving personnel and the rear-row personnel can be used as two dialogue personnel; or after the personnel in the cabin pass through one or more rounds of conversations, the vehicle-mounted sound collecting device only collects the sound at the main driving seat and the rear seat, so that the main driving personnel and the rear seat can be determined to be two conversations.
Step 602: the main driving microphone collects sounds emitted by a main driver, namely main driving sounds.
Step 603: and the main driving microphone sends the acquired main driving analog signal to the controller.
The main driving microphone collects main driving sound, and can generate corresponding analog signals, namely main driving analog signals, and then the main driving analog signals can be sent to the controller. The controller is a device that can implement the processing method of the vehicle data, such as an intelligent cabin controller or the like.
Step 604: the controller converts the primary drive analog signal into a primary drive digital signal and sends the primary drive digital signal to the AMP.
Step 605: and the AMP amplifies the main driving digital signal and sends the generated main driving amplified signal to a rear-row loudspeaker.
The controller determines that the current dialogue is a main driver and a rear-row person, and the rear-row person is a receiving person when the main driver speaks, so that the rear-row loudspeaker is a vehicle-mounted sound playing device corresponding to the position of the receiving person. The controller may control the AMP to transmit the amplified main drive amplified signal to the rear-row speaker.
It can be understood that the above steps 602 to 605 are the process when the main driver speaks, and the main driver is the speaker and the rear person is the receiver.
Step 606: the rear-row microphone collects sounds made by rear-row personnel, namely rear-row sounds.
Step 607: the back-row microphone sends the collected back-row analog signals to the controller.
Step 608: the controller converts the back-end analog signal to a back-end digital signal and sends it to the AMP.
Step 609: the AMP amplifies the rear digital signal and sends the generated rear emission large signal to the main drive speaker.
It can be understood that the above steps 606 to 609 are the process when the person in the back row speaks, and the host person is the receiving person.
In the present embodiment, the description will be given taking an example in which the in-vehicle sound collection device is a microphone and the in-vehicle sound playback device is a speaker; as shown in fig. 6, the sound equipment on the main driving side includes a main driving microphone and a main driving speaker, and the sound equipment on the rear-row side includes a rear-row microphone and a rear-row speaker.
Step 610: after the session is completed, the controller stops conducting the in-vehicle session.
For example, if no sound is collected for a period of time, or no person is detected in the vehicle, the end of the session may be determined, at which point the controller may stop the auxiliary session.
According to the vehicle data processing method, when a person in a cabin needs to perform a conversation, the vehicle-mounted sound collecting device can be used for collecting sound emitted by a speaker and sending current sound data corresponding to the sound to the vehicle-mounted sound playing device on the side of a receiving person, so that the vehicle-mounted sound playing device can play the sound to the receiving person, the auxiliary conversation in the vehicle is realized, the conversation quality in the vehicle can be ensured, and the conversation in the vehicle can be smoothly performed. By determining the cabin position of the speaker, the speaker can be conveniently positioned, and then the sound equipment corresponding to the speaker and the receiving person can be determined.
In some alternative embodiments, the step 504 of verifying the detection result according to the detection value may specifically include the following steps D1 to D2.
Step D1: and in the case that the detection value indicates that a person exists at the position in the cabin, determining that the detection result is reliable.
In this embodiment, in order to make it possible to verify the detection result relatively easily, it is also possible to determine whether the detection value indicates that a person is present at the position in the cabin. Specifically, if the detection value is sufficiently large, it can be considered to indicate that a person is present at the position in the cabin. Wherein if the detected value indicates that a person is present at the position in the cabin, and the detected result also indicates that a person is present at the position in the cabin, the detected result is considered to be authentic, i.e., that a person is present at the position in the cabin.
For example, the second detection device is a vehicle-mounted sound collection device; if the detected value of a certain cabin position indicates that a person exists at the cabin position, the detected value of the cabin position can indicate that a speaking cabin person exists at the cabin position, namely, a speaking person.
Step D2: and if the detection value indicates that no person exists at the position in the cabin, determining whether the detection result is credible according to the size of the detection value.
In this embodiment, if the detection value indicates that no person is present at the position in the cabin, whether the detection result is reliable can be further determined according to the magnitude of the detection value.
In some alternative embodiments, the step D2 "determining whether the detection result is reliable according to the magnitude of the detection value" may specifically include step D21 and step D22.
Step D21: and under the condition that the detection value is larger than the first threshold value, determining that the detection result is reliable.
Step D22: and under the condition that the detection value is smaller than the first threshold value, determining that the detection result is not credible.
In this embodiment, a threshold value of the detection value, that is, a first threshold value may be preset, if the detection result indicates that a person exists in the cabin, and the detection value of the position in the cabin is greater than the first threshold value, the confidence level of the person in the detection position may be considered to be higher, and if the second detection device further detects that the person exists in the position, the person in the position may be considered to be present, that is, the detection result is reliable. Conversely, if the detected value of the position in the cabin is smaller than the first threshold, the confidence that the person exists at the position in the cabin may be considered to be low, and if the second detection device detects that the person exists there, the second detection device may detect that the person exists at the position, and the detection result of the second detection device needs to be discarded at this time, that is, the detection result is not reliable.
For example, the second detection device is a vehicle-mounted sound collection device, and compared with the seat state detection device and the in-vehicle image collection device, the in-vehicle environment is complex, and the in-vehicle sound collection device is used for realizing positioning based on the in-cabin personnel sound collected by the vehicle-mounted sound collection device, so that interference is easy to exist, and the recognition is inaccurate; therefore, if no person is basically determined at a certain cabin location (i.e., the detection value is lower than the first threshold value) based on the seat state detection means and the in-vehicle image pickup means, the in-vehicle sound pickup means may be considered to recognize that there is a mistake when the in-vehicle sound pickup means considers that sound at the cabin location is picked up.
Alternatively, another threshold value of the detection value, that is, a second threshold value, may also be preset, and the second threshold value is greater than the first threshold value. It is determined whether the detection value indicates a person at the in-cabin location based on a second, greater threshold. Specifically, in the case where the detection value is greater than the second threshold value, determining that the detection value indicates that a person is present at the position in the cabin; in the event that the detection value is less than the second threshold, it is determined that the detection value indicates that no person is at the in-cabin location.
In this embodiment, if the detected value of the position in the cabin is smaller than the first threshold value, it may be basically considered that no person is located at the position in the cabin; if the detection value of the position in the cabin is larger than the first threshold value and smaller than the second threshold value, the position in the cabin can be considered to have a certain possibility, and at the moment, further judgment is needed by combining the second detection equipment; if the detected value of the position in the cabin is greater than the second threshold value, it is basically considered that a person is present at the position in the cabin.
The first threshold and the second threshold can be set based on actual demands, and whether the dialog personnel exist at the position in the cabin can be accurately and reasonably determined by setting the appropriate first threshold and the second threshold. In this embodiment, the first threshold may be the smaller of the second value and the fourth value, or the smaller of the first value, the third value, and the fourth value, and the second threshold may be the larger of the second value and the fourth value, or the sum of the first value and the fourth value, or the sum of the second value and the fourth value.
Specifically, in order to directly determine that a person is present at the position in the cabin without considering the speech recognition result, a sufficiently large second threshold value needs to be set, where n1, n2, n3, and n4 represent the first value, the second value, the third value, and the fourth value, and th1 and th2 represent the first threshold value and the second threshold value, respectively, and th2 may be the larger of the second value and the fourth value, that is, th2=max (n 2, n 4); if a person is considered to be present at the cabin interior position in one of the seat recognition and the face recognition that has higher reliability, the person can be directly considered to be present at the cabin interior position.
Alternatively, the second threshold th2 may be the sum of the second value and the fourth value, i.e. th2=n2+n4. At this time, only if the face recognition indicates a person, the seat sensor state indicates a person, and then the seat belt state is tied, that is, both the face recognition and the seat recognition indicate a person, the person at the cabin position is considered to be present at this time.
In addition, there is a contradiction in face recognition or seat recognition, i.e. one of them indicates that there is a person and the other indicates that there is no person, and at this time, further judgment needs to be performed in combination with voice recognition; therefore, the first threshold th1 may be the smaller of the second value and the fourth value, that is, th1=min (n 2, n 4); if the detection value is smaller than the first threshold th1, it is indicated that no person is present in both face recognition and seat recognition at this time, and if the result of the voice recognition indicates that a person is present there, the voice recognition is erroneous. If the detected value is between the first threshold th1 and the second threshold th2, it is considered that one of face recognition and seat recognition indicates a person, and further judgment is necessary in combination with voice recognition.
Alternatively, the first threshold th1 may be set to be the sum of the first value and the fourth value, or the sum of the second value and the fourth value, that is, th1=n1+n4, or th1=n3+n4, so that when a person is detected at the cabin position to some extent based on both the seat state detection device and the in-vehicle image acquisition device, further judgment is required in combination with voice recognition.
For example, a process of determining a detection value of a position in a cabin based on seat recognition and face recognition and comprehensively judging whether a person is present at the position in the cabin in combination with voice recognition is shown in fig. 7; as shown in fig. 7, the process of identifying the location of the dialog personnel may include the following steps 701 through 709.
Step 701: after the vehicle is closed, a detection value of 0 is initially set for each seat.
Wherein the body domain may generate a signal for closing the door of the vehicle to inform the controller that the door of the vehicle has been closed; here, this controller is exemplified by an intelligent cabin controller (CDC). Wherein each seat corresponds to a cabin location.
Step 702: if the seat is occupied, the detection value is increased by a first value.
Wherein the weight sensor of each seat may send the collected data to a seat controller, which thereby determines whether the seat is occupied. If the seat is unmanned, the detection value is kept unchanged; if the seat is occupied, the first value is increased for the detection value. For example, the first value is 1.
Step 703: if the seat has a face, the detection value is increased by a fourth value.
The method comprises the steps that face recognition is carried out based on an in-vehicle image acquired by an OMS camera, if a certain seat recognizes a face, a face recognition result can be determined to indicate that a person exists at the seat, and a fourth numerical value is added to a detection value of the seat at the moment; for example, the fourth value is also 1. The face detection may be continued for 5 minutes, for example, and if the face can be detected all the time, the seat is considered to be occupied.
Step 704: if the safety belt is tied, the detection value is increased by a corresponding value.
Wherein the corresponding value may be a difference between the second value and the first value such that the second value may be added to the detected value of the in-cabin position in case the seat sensor state indicates a person and thereafter the seat belt state is attached. For example, the corresponding value is 1; accordingly, the second value is 1+1=2.
It will be appreciated that if the gravity sensor indicates that a seat is occupied, the face recognition indicates that a face is present, and the seat belt is tethered, the detection values are: a second value + a fourth value; for example, the detection value is 3.
If the gravity sensor indicates that the seat is occupied and the face recognition indicates that the face is present but the seat belt is not fastened, the detection value is: a first value + a fourth value; for example, the detection value is 2.
If the gravity sensor indicates that the seat is unmanned, the face recognition indicates that the face exists, and the safety belt is fastened, the detection value is as follows: related value + fourth value; for example, the detection value is 2.
If the gravity sensor indicates that the seat is not people, the face recognition indicates that the face exists, and the safety belt is not tied, the detection value is as follows: a fourth value; for example, the detection value is 1.
If the gravity sensor indicates that the seat is occupied, the face recognition indicates that the face is not present, and the safety belt is not fastened, the detection value is: a first value; for example, the detection value is 1.
If the gravity sensor indicates that the seat is not occupied, the face recognition indicates that the face is not present, and the seat belt is not fastened, the detection value is still the initial value, and the detection value is 0.
Step 705: it is determined whether the detection value is greater than a second threshold th2. If the detected value is greater than the second threshold th2, continuing to step 706; if the detected value is smaller than the second threshold th2, step 707 is continued.
For example, the first threshold th1 may be a sum of the second value and the fourth value; for example, the first threshold th1 is 3.
Step 706: the identification results are consistent, and the detection results are reliable. I.e. the speech recognition result is trusted.
If the detected value is greater than the second threshold th2, it means that the person is determined to be at the cabin position based on the gravity sensor, the safety belt buckle, the face recognition and the voice recognition, so that the result of the voice recognition is reliable.
Step 707: it is determined whether the detection value is greater than a first threshold th1. If the detected value is greater than the first threshold th1, continuing to step 708; if the detected value is smaller than the first threshold th1, the process proceeds to step 709.
For example, the first threshold th1 may be a sum of a first value and a fourth value; for example, the first threshold th1 is 2.
Step 708: the seat is occupied. I.e. the speech recognition result is still authentic.
Step 709: no one is at the seat. I.e. the speech recognition result is not authentic.
Based on the process shown in fig. 7, the location of the dialog person can be recognized relatively accurately.
In this embodiment, if the result of the voice recognition is inconsistent with the result of the seat recognition and the face recognition, the current voice recognition result is considered to be unreliable, and the corresponding speaker cannot be determined based on the voice recognition result; or, the voice recognition result can be corrected to a certain extent according to the seat recognition and face recognition results so as to accurately determine the speaker.
For example, if a person or no person can be represented in some locations with high confidence based on seat recognition and face recognition, but the person location recognized by speech recognition contradicts this, the speech recognition may be considered to be unreliable. For example, if the speech recognition is currently speaking for the secondary driver, but the seat recognition and face recognition show that the front row is only the primary driver, i.e., the secondary driver is not the person, then the speaker can be identified as the primary driver. Alternatively, if the speech recognition is speaking in the middle of the back row, but the seat recognition and face recognition indicates that only the right side of the back row is occupied, the speaker is identified as right back.
Fig. 8 is a schematic structural view showing an embodiment of the processing device of vehicle data of the present invention. As shown in fig. 8, the vehicle data processing apparatus 800 includes:
a detection data acquisition module 801, configured to acquire detection data of position detection of personnel in the cabin by the first detection device;
a detection value determining module 802, configured to determine a detection value of the position in the cabin according to the detection data; the detection value represents the confidence level of a person at the position in the cabin;
a detection result obtaining module 803, configured to obtain a detection result indicating that the second detection device recognizes that the person is located at the position in the cabin;
and the verification module 804 is configured to verify the detection result according to the detection value, and comprehensively determine whether a person is present at the position in the cabin.
In some alternative embodiments, the second detection device includes: a vehicle-mounted sound collection device;
the detection result obtaining module 803 obtains a detection result indicating that the second detection device recognizes that the person is present at the position in the cabin, including:
and positioning according to the voice of the personnel in the cabin acquired by the vehicle-mounted sound acquisition device, determining the position of the personnel in the cabin, and generating a corresponding detection result.
In some optional embodiments, the verifying module 804 verifies the detection result according to the detection value, including:
determining that the detection result is authentic if the detection value indicates that a person is present at the in-cabin location;
and if the detection value indicates that no person exists at the position in the cabin, determining whether the detection result is credible according to the size of the detection value.
In some alternative embodiments, the verification module 804 determines whether the detection result is authentic according to the magnitude of the detection value, including:
under the condition that the detection value is larger than a first threshold value, determining that the detection result is credible;
and determining that the detection result is not trusted under the condition that the detection value is smaller than a first threshold value.
In some alternative embodiments, the verification module 804 is further configured to:
determining that the detection value indicates a person at the in-cabin location if the detection value is greater than a second threshold;
determining that the detection value represents no person at the in-cabin location if the detection value is less than a second threshold;
wherein the second threshold is greater than the first threshold.
In some alternative embodiments, the apparatus further comprises an auxiliary dialogue module for:
Under the condition that people exist at least two cabin positions, acquiring current voice data acquired by a vehicle-mounted sound acquisition device corresponding to the positions of speakers; the speaker is located in one of the manned cabin locations;
controlling a vehicle-mounted sound playing device corresponding to the position of the receiving person to play the current voice data; the receiving person is located at another personnel capsule location.
In some alternative embodiments, the first detection device includes: a seat state detection device and/or an in-vehicle image acquisition device;
the detection data includes: the seat sensor state and the safety belt state acquired by the seat state detection device and/or the in-vehicle image acquired by the in-vehicle image acquisition device;
the detection value determining module 802 determines a detection value of the position in the cabin according to the detection data, including:
respectively determining whether a person exists at the position in the cabin according to each item of data in the detection data;
performing value-added processing on the detection value of the cabin position under the condition that the cabin position is determined to be someone; the detection value of the position in the cabin is initially 0.
In some alternative embodiments, the detection value determining module 802 performs a value-added process on the detection value of the in-cabin location if it is determined that the in-cabin location is a person, including:
Increasing a first value for a detected value of the cabin interior position in a case where the seat sensor status indicates that a person is present at the cabin interior position;
increasing a second value for the detection value of the cabin interior position in the case where the seat sensor state indicates that a person is present at the cabin interior position and then the seat belt state is tethered;
increasing a third value for the detected value of the in-cabin position in the case where the seat belt state is tethered and thereafter the seat sensor state indicates a person at the in-cabin position; the third value is less than or equal to the first value < the second value;
and adding a fourth value to the detection value of the cabin interior position in the case that the vehicle interior image shows that the cabin interior position is a person.
In some alternative embodiments, the apparatus further comprises a zeroing module for: setting 0 to the detection value of the cabin interior position under the condition that a door closing signal is detected, and then re-determining the detection value of the cabin interior position;
alternatively, in the case where the seat sensor state changes from indicating that a person is present at the in-cabin position to no person, and/or the seat belt state changes from being on to off, the detection value of the in-cabin position is set to 0, and then the detection value of the in-cabin position is newly determined.
In the vehicle data processing apparatus provided in this embodiment, the confidence level of the person at each position in the cabin is determined based on the first detection device, and if a detection result indicating the person at the position in the cabin is obtained based on the second detection device, the detection result may be further checked based on a detection value capable of indicating the confidence level of the position in the cabin, so as to determine the confidence level of the detection result, thereby determining whether the person is at the position in the cabin. The method can be combined with various detection devices to detect the positions of personnel in the cabin, so that the accuracy of personnel detection can be improved; and in addition, the detection value of each cabin position can be determined in advance in a mode of checking the detection result of the detection value, when the second detection equipment detects that a person exists at the cabin position, the detection result of the second detection equipment is verified quickly, the processing amount required during real-time processing can be reduced, and particularly for a processor with lower processing performance, the real-time performance of vehicle data processing can be improved.
Fig. 9 shows a schematic structural diagram of an embodiment of the vehicle data processing apparatus of the present invention, and the embodiment of the present invention is not limited to the specific implementation of the vehicle data processing apparatus.
As shown in fig. 9, the processing device of vehicle data may include: a processor 902, a communication interface (Communications Interface), a memory 906, and a communication bus 908.
Wherein: processor 902, communication interface 904, and memory 906 communicate with each other via a communication bus 908. A communication interface 904 for communicating with network elements of other devices, such as clients or other servers. The processor 902 is configured to execute the program 910, and may specifically perform relevant steps in the above-described embodiment of a method for processing vehicle data.
In particular, the program 910 may include program code including computer-executable instructions.
The processor 902 may be a central processing unit, CPU, or a specific integrated circuit ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement embodiments of the present invention. The one or more processors included in the processing device of the vehicle data may be the same type of processor, such as one or more CPUs; but may also be different types of processors such as one or more CPUs and one or more ASICs.
A memory 906 for storing a program 910. Memory 906 may comprise high-speed RAM memory or may also include non-volatile memory (non-volatile memory), such as at least one disk memory.
The program 910 may be specifically invoked by the processor 902 to cause the vehicle data processing apparatus to implement the method shown in the above embodiment.
The embodiment of the invention also provides a vehicle which comprises a plurality of groups of sound equipment and processing equipment of vehicle data; the sound equipment comprises at least one vehicle-mounted sound collecting device and at least one vehicle-mounted sound playing device. For example, a schematic structural diagram of the vehicle may be shown in fig. 1, and the device for processing vehicle data may be the device shown in fig. 9, so as to implement the method for processing vehicle data provided in the foregoing method embodiment.
An embodiment of the present invention provides a computer-readable storage medium storing at least one executable instruction that, when executed by a processor, implements the method for processing vehicle data according to any one of the above embodiments.
The storage medium can be a magnetic disk, an optical disk, a read-only memory, a random access memory, a flash memory, a hard disk, a solid state disk or the like; further, the storage medium may also comprise a combination of memories of the kind described above. It will be appreciated that a computer, processor, microprocessor controller or programmable hardware includes a storage element that can store or receive software or computer code that, when accessed and executed by the computer, processor or hardware, implements the methods illustrated by the above embodiments.
The algorithms or displays presented herein are not inherently related to any particular computer, virtual system, or other apparatus. In addition, embodiments of the present invention are not directed to any particular programming language.
In the description provided herein, numerous specific details are set forth. It will be appreciated, however, that embodiments of the invention may be practiced without such specific details. Similarly, in the above description of exemplary embodiments of the invention, various features of embodiments of the invention are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of one or more of the various inventive aspects. Wherein the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of this invention.
Those skilled in the art will appreciate that the modules in the apparatus of the embodiments may be adaptively changed and disposed in one or more apparatuses different from the embodiments. The modules or units or components of the embodiments may be combined into one module or unit or component and, furthermore, they may be divided into a plurality of sub-modules or sub-units or sub-components. Except that at least some of such features and/or processes or elements are mutually exclusive.
It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The use of the words first, second, third, etc. do not denote any order. These words may be interpreted as names. The steps in the above embodiments should not be construed as limiting the order of execution unless specifically stated.
Claims (13)
1. A method of processing vehicle data, the method comprising:
acquiring detection data of the first detection equipment for detecting the position of personnel in the cabin;
Determining a detection value of the position in the cabin according to the detection data; the detection value represents the confidence level of a person at the position in the cabin;
acquiring a detection result indicating that the second detection device recognizes that a person exists at the position in the cabin;
and checking the detection result according to the detection value, and comprehensively determining whether a person exists at the position in the cabin.
2. The method of claim 1, wherein the second detection device comprises: a vehicle-mounted sound collection device;
the acquiring a detection result indicating that the second detection device recognizes that a person is present at the position in the cabin includes:
and positioning according to the voice of the personnel in the cabin acquired by the vehicle-mounted sound acquisition device, determining the position of the personnel in the cabin, and generating a corresponding detection result.
3. The method according to claim 2, wherein verifying the detection result according to the detection value comprises:
determining that the detection result is authentic if the detection value indicates that a person is present at the in-cabin location;
and if the detection value indicates that no person exists at the position in the cabin, determining whether the detection result is credible according to the size of the detection value.
4. A method according to claim 3, wherein said determining whether the detection result is authentic according to the magnitude of the detection value comprises:
under the condition that the detection value is larger than a first threshold value, determining that the detection result is credible;
and determining that the detection result is not trusted under the condition that the detection value is smaller than a first threshold value.
5. The method as recited in claim 4, further comprising:
determining that the detection value indicates a person at the in-cabin location if the detection value is greater than a second threshold;
determining that the detection value represents no person at the in-cabin location if the detection value is less than a second threshold;
wherein the second threshold is greater than the first threshold.
6. The method as recited in claim 2, further comprising:
under the condition that people exist at least two cabin positions, acquiring current voice data acquired by a vehicle-mounted sound acquisition device corresponding to the positions of speakers; the speaker is located in one of the manned cabin locations;
controlling a vehicle-mounted sound playing device corresponding to the position of the receiving person to play the current voice data; the receiving person is located at another personnel capsule location.
7. The method according to any one of claims 1 to 6, wherein the first detection device comprises: a seat state detection device and/or an in-vehicle image acquisition device;
the detection data includes: the seat sensor state and the safety belt state acquired by the seat state detection device and/or the in-vehicle image acquired by the in-vehicle image acquisition device;
the determining the detection value of the position in the cabin according to the detection data comprises the following steps:
respectively determining whether a person exists at the position in the cabin according to each item of data in the detection data;
performing value-added processing on the detection value of the cabin position under the condition that the cabin position is determined to be someone; the detection value of the position in the cabin is initially 0.
8. The method of claim 7, wherein the value-added processing of the detected value of the in-cabin location if it is determined that there is a person at the in-cabin location comprises:
increasing a first value for a detected value of the cabin interior position in a case where the seat sensor status indicates that a person is present at the cabin interior position;
increasing a second value for the detection value of the cabin interior position in the case where the seat sensor state indicates that a person is present at the cabin interior position and then the seat belt state is tethered;
Increasing a third value for the detected value of the in-cabin position in the case where the seat belt state is tethered and thereafter the seat sensor state indicates a person at the in-cabin position; the third value is less than or equal to the first value < the second value;
and adding a fourth value to the detection value of the cabin interior position in the case that the vehicle interior image shows that the cabin interior position is a person.
9. The method as recited in claim 7, further comprising:
setting 0 to the detection value of the cabin interior position under the condition that a door closing signal is detected, and then re-determining the detection value of the cabin interior position;
alternatively, in the case where the seat sensor state changes from indicating that a person is present at the in-cabin position to no person, and/or the seat belt state changes from being on to off, the detection value of the in-cabin position is set to 0, and then the detection value of the in-cabin position is newly determined.
10. A processing apparatus of vehicle data, characterized in that the apparatus comprises:
the detection data acquisition module is used for acquiring detection data of the first detection equipment for detecting the position of personnel in the cabin;
the detection value determining module is used for determining a detection value of the position in the cabin according to the detection data; the detection value represents the confidence level of a person at the position in the cabin;
The detection result acquisition module is used for acquiring a detection result indicating that the second detection equipment recognizes that a person exists at the position in the cabin;
and the verification module is used for verifying the detection result according to the detection value and comprehensively determining whether a person exists at the position in the cabin.
11. A processing apparatus of vehicle data, characterized by comprising: the device comprises a memory and a processor, wherein the memory and the processor are in communication connection;
the memory is configured to store at least one executable instruction that causes the processor to perform the method of processing vehicle data according to any one of claims 1 to 9.
12. A vehicle comprising a first detection device, a second detection device, and the processing device for vehicle data according to claim 11.
13. A computer-readable storage medium, characterized in that at least one executable instruction is stored in the storage medium, which when being executed by a processor implements the method of processing vehicle data according to any one of claims 1 to 9.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311493648.2A CN117520892A (en) | 2023-11-09 | 2023-11-09 | Method, device, equipment, vehicle and medium for processing vehicle data |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311493648.2A CN117520892A (en) | 2023-11-09 | 2023-11-09 | Method, device, equipment, vehicle and medium for processing vehicle data |
Publications (1)
Publication Number | Publication Date |
---|---|
CN117520892A true CN117520892A (en) | 2024-02-06 |
Family
ID=89750799
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311493648.2A Pending CN117520892A (en) | 2023-11-09 | 2023-11-09 | Method, device, equipment, vehicle and medium for processing vehicle data |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117520892A (en) |
-
2023
- 2023-11-09 CN CN202311493648.2A patent/CN117520892A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US9743213B2 (en) | Enhanced auditory experience in shared acoustic space | |
CN101064975B (en) | Vehicle communication system | |
EP3231166B1 (en) | Enhanced conversational communications in shared acoustic space | |
CN109119060A (en) | A kind of reduction method and system applied to automobile | |
US20170026764A1 (en) | Automatic car audio volume control to aid passenger conversation | |
US20190037363A1 (en) | Vehicle based acoustic zoning system for smartphones | |
US10805730B2 (en) | Sound input/output device for vehicle | |
US10020785B2 (en) | Automatic vehicle occupant audio control | |
CN110402584A (en) | Interior call control apparatus, cabincommuni cation system and interior call control method | |
US11061236B2 (en) | Head-mounted display and control method thereof | |
WO2023020620A1 (en) | Audio-based processing method and apparatus | |
US11673512B2 (en) | Audio processing method and system for a seat headrest audio system | |
CN110865788B (en) | Vehicle communication system and method of operating a vehicle communication system | |
CN113270095B (en) | Voice processing method, device, storage medium and electronic equipment | |
CN117520892A (en) | Method, device, equipment, vehicle and medium for processing vehicle data | |
US10917717B2 (en) | Multi-channel microphone signal gain equalization based on evaluation of cross talk components | |
CN115831141B (en) | Noise reduction method and device for vehicle-mounted voice, vehicle and storage medium | |
US20180157459A1 (en) | Ear monitoring audio | |
JP2005354223A (en) | Sound source information processing apparatus, sound source information processing method, and sound source information processing program | |
WO2018173112A1 (en) | Sound output control device, sound output control system, and sound output control method | |
CN108886551B (en) | Method and apparatus for controlling an intercom system | |
US20240121555A1 (en) | Zoned Audio Duck For In Car Conversation | |
WO2013098983A1 (en) | Sound control device, sound control method, sound control program, and recording medium on which sound control program is recorded | |
JP2019137167A (en) | Voice output control device, and voice output control program | |
WO2021156945A1 (en) | Sound separation device and sound separation method |
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 |