CN114676277A

CN114676277A - Parameter result prediction method and system based on database

Info

Publication number: CN114676277A
Application number: CN202210348752.1A
Authority: CN
Inventors: 李晓光
Original assignee: Wuxi Fifth Peoples Hospital
Current assignee: Wuxi Fifth Peoples Hospital
Priority date: 2022-04-01
Filing date: 2022-04-01
Publication date: 2022-06-28

Abstract

The invention provides a database-based parameter result prediction method and a database-based parameter result prediction system, wherein the method comprises the following steps: building a behavior prediction analysis database; acquiring images of the target vehicle within a preset circumference range in multiple angles, and extracting key elements from the obtained multiple-angle image set to obtain a multi-view element analysis result; generating a preset behavior track parameter sequence to obtain a preset driving behavior parameter of the target vehicle; and matching the parameter association degree of the preset behavior track parameter sequence and the pre-driving behavior parameters to generate a first target behavior parameter, and driving and monitoring the target vehicle. The technical problems that the prediction is inaccurate based on the database parameter result and the prediction result of the vehicle database is not matched with the actual condition in the prior art are solved, the parameter result prediction accuracy based on the database is improved, and the technical effect of improving the matching degree of the prediction result of the vehicle database and the actual condition is improved.

Description

Parameter result prediction method and system based on database

Technical Field

The invention relates to the technical field of unmanned driving, in particular to a parameter result prediction method and system based on a database.

Background

The automobile industry is in a revolution, and today with the increasing importance of intelligent transportation, the automatic driving direction in the automobile field is considered as the most potential application direction, mainly comprising the digital and intelligent development of automobiles.

Currently, the conditions of the vehicle and the surrounding environment are generally sensed in real time by using a sensor, and then planning decision is made through an intelligent system. At present, parameters of the surrounding environment are collected, and after the parameters are analyzed, road information is judged according to an analysis result, so that the accuracy of intelligent prediction is improved.

In the prior art, because the condition of mixed running of people and vehicles on a road is serious and has high randomness, the behavior information of the people and the vehicles around the road can not be accurately judged according to the database, the condition that the driving is collided with the people and the vehicles around the road according to the prediction result to cause traffic accidents can occur, and the technical problem that the prediction is not accurate based on the parameter result of the database, the reliable prediction result can not be provided, the track information can not be accurately predicted, and the prediction result of the vehicle database is not matched with the actual condition can be caused.

Disclosure of Invention

The application provides a parameter result prediction method and system based on a database, which are used for solving the technical problems that in the prior art, the prediction based on the database parameter result is inaccurate, the track information cannot be accurately predicted, and the prediction result of a vehicle database is not matched with the actual situation.

In view of the foregoing, the present application provides a method and system for predicting a parameter result based on a database.

According to a first aspect of the application, a parameter result prediction method based on a database is provided, and the method comprises the steps of building a behavior prediction analysis database, wherein a multi-view prediction analysis model is embedded in the behavior prediction analysis database; the method comprises the steps that an image acquisition device based on a target vehicle acquires images of the target vehicle within a preset circumference range in a multi-angle mode to obtain a multi-angle image set; extracting key elements from the multi-angle image set to obtain a key element image set; uploading the key element image set as input information to the multi-view prediction analysis model to perform interpretation analysis of different views, and obtaining multi-view element analysis results; uploading the multi-view element analysis result to a probability grading track matching model for data processing, and generating a preset behavior track parameter sequence; obtaining a pre-driving behavior parameter of the target vehicle; performing parameter association degree matching on the preset behavior track parameter sequence and the pre-driving behavior parameters to generate first target behavior parameters corresponding to a first target parameter set; and sending the first target behavior parameters to an intelligent driving end of the target vehicle, and monitoring the driving of the target vehicle.

In a second aspect of the present application, there is provided a database-based parameter outcome prediction system, the system comprising: the system comprises a first building unit, a second building unit and a third building unit, wherein the first building unit is used for building a behavior prediction analysis database, and a multi-view prediction analysis model is embedded in the behavior prediction analysis database; the system comprises a first obtaining unit, a second obtaining unit and a third obtaining unit, wherein the first obtaining unit is used for carrying out multi-angle collection on images of a target vehicle within a preset circumference range based on an image collecting device of the target vehicle to obtain a multi-angle image set; a second obtaining unit, configured to perform key element extraction on the multi-angle image set to obtain a key element image set; a third obtaining unit, configured to upload the key element image set as input information to the multi-view predictive analysis model to perform interpretation analysis on different views, so as to obtain a multi-view element analysis result; the first generation unit is used for uploading the multi-view element analysis result to a probability grading track matching model for data processing to generate a preset behavior track parameter sequence; a fourth obtaining unit configured to obtain a pre-driving behavior parameter of the target vehicle; the first matching unit is used for performing parameter association degree matching on the preset behavior track parameter sequence and the pre-driving behavior parameters to generate first target behavior parameters corresponding to a first target parameter set; and the first monitoring unit is used for sending the first target behavior parameters to an intelligent driving end of the target vehicle and monitoring the driving of the target vehicle.

In a third aspect of the present application, there is provided a database-based parameter result prediction apparatus, including: a processor coupled to a memory for storing a program that, when executed by the processor, causes a system to perform the steps of the method according to the first aspect.

In a fourth aspect of the present application, a computer-readable storage medium is provided, on which a computer program is stored, which computer program, when being executed by a processor, carries out the steps of the method according to the first aspect.

One or more technical solutions provided in the present application have at least the following technical effects or advantages:

the method provided by the embodiment of the application provides a parameter result prediction method based on a database, and the method comprises the steps of building a behavior prediction analysis database, wherein a multi-view prediction analysis model is embedded in the behavior prediction analysis database; the method comprises the steps that an image acquisition device based on a target vehicle acquires images of the target vehicle within a preset circumference range in a multi-angle mode to obtain a multi-angle image set; extracting key elements from the multi-angle image set to obtain a key element image set; uploading the key element image set as input information to the multi-view prediction analysis model to perform interpretation analysis of different views, and obtaining multi-view element analysis results; uploading the multi-view element analysis result to a probability grading track matching model for data processing, and generating a preset behavior track parameter sequence; obtaining a pre-driving behavior parameter of the target vehicle; performing parameter association degree matching on the preset behavior track parameter sequence and the pre-driving behavior parameters to generate first target behavior parameters corresponding to a first target parameter set; and sending the first target behavior parameters to an intelligent driving end of the target vehicle, and monitoring the driving of the target vehicle. The method and the device have the advantages that the prediction accuracy of the parameter result based on the database is improved, the reliable driving environment prediction is provided for driving, the accuracy of predicting the tracks of surrounding pedestrians or surrounding vehicles is improved, and the technical effect of matching degree of the prediction result of the vehicle database and the actual situation is improved.

The foregoing description is only an overview of the technical solutions of the present application, and the present application can be implemented according to the content of the description in order to make the technical means of the present application more clearly understood, and the following detailed description of the present application is given in order to make the above and other objects, features, and advantages of the present application more clearly understandable.

Drawings

Fig. 1 is a schematic flow chart of a database-based parameter result prediction method provided in the present application;

FIG. 2 is a schematic diagram illustrating a process of extracting key elements from the multi-angle image set in the database-based parameter result prediction method according to the present application;

fig. 3 is a schematic flow chart illustrating a multi-view element analysis result obtained in a database-based parameter result prediction method according to the present application;

FIG. 4 is a schematic diagram of a database-based parameter result prediction system according to the present application;

FIG. 5 is a schematic diagram of an exemplary electronic device of the present application;

description of reference numerals: the system comprises a first building unit 11, a first obtaining unit 12, a second obtaining unit 13, a third obtaining unit 14, a first generating unit 15, a fourth obtaining unit 16, a first matching unit 17, a first monitoring unit 18, an electronic device 300, a memory 301, a processor 302, a communication interface 303 and a bus architecture 304.

Detailed Description

The application provides a parameter result prediction method and system based on a database, and aims to solve the technical problems that in the prior art, the prediction is inaccurate based on the database parameter result, the track information cannot be accurately predicted, and the prediction result of a vehicle database is not matched with the actual situation.

Summary of the application

Against the rapid development of economy and science and technology in China, rapid revolution is carried out in each field, and at present, the direction of automatic driving in the automobile field is considered to be the most potential application direction in the increasingly important intelligent transportation field, and mainly comprises the digital and intelligent development of automobiles. Currently, the conditions of the vehicle and the surrounding environment are generally sensed in real time by using a sensor, and then planning decision is made through an intelligent system. At present, parameters of the surrounding environment are collected, and after the parameters are analyzed, road information is judged according to an analysis result, so that the accuracy of intelligent prediction is improved.

In the prior art, due to the fact that the situation of people and vehicles mixed in a road is serious and has high randomness, the behavior information of the pedestrians and the surrounding vehicles cannot be accurately judged according to the database, the situation that driving is carried out according to the prediction result, the pedestrians and the surrounding vehicles collide can occur, and traffic accidents are caused, the technical problem that the prediction is not accurate based on the database parameter result, reliable prediction results cannot be provided, track information can be accurately predicted, and the prediction result of the vehicle database is not matched with the actual situation is caused.

In view of the above technical problems, the technical solution provided by the present application has the following general idea:

the method provided by the embodiment of the application builds a behavior prediction analysis database, wherein a multi-view prediction analysis model is embedded in the behavior prediction analysis database; the method comprises the steps that an image acquisition device based on a target vehicle acquires images of the target vehicle within a preset circumference range in a multi-angle mode to obtain a multi-angle image set; extracting key elements from the multi-angle image set to obtain a key element image set; uploading the key element image set as input information to the multi-view prediction analysis model to perform interpretation analysis of different views, and obtaining multi-view element analysis results; uploading the multi-view element analysis result to a probability grading track matching model for data processing, and generating a preset behavior track parameter sequence; obtaining a pre-driving behavior parameter of the target vehicle; performing parameter association degree matching on the preset behavior track parameter sequence and the pre-driving behavior parameters to generate first target behavior parameters corresponding to a first target parameter set; and sending the first target behavior parameter to an intelligent driving end of the target vehicle, and monitoring the driving of the target vehicle.

Having described the basic principles of the present application, the technical solutions in the present application will be described clearly and completely with reference to the accompanying drawings, and it should be understood that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments of the present application, and the present application is not limited to the exemplary embodiments described herein. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present application without making any creative effort, shall fall within the protection scope of the present application. It should be further noted that, for the convenience of description, only some but not all of the elements relevant to the present application are shown in the drawings.

Example one

As shown in fig. 1, the present application provides a database-based parameter result prediction method, including:

step S100: building a behavior prediction analysis database, wherein a multi-view prediction analysis model is embedded in the behavior prediction analysis database;

specifically, the behavior prediction analysis database is a database for establishing action track analysis of moving objects around a target vehicle, wherein a multi-view prediction analysis model is embedded in the behavior analysis database and used for analyzing the surrounding environment of the target vehicle from multiple angles, and the motion states of different objects are obtained by analyzing the behaviors of different objects, so that accurate motion state information is provided for the driving of the target vehicle, and the technical effect of improving the prediction accuracy of parameter results based on the database is achieved.

Step S200: the method comprises the steps that an image acquisition device based on a target vehicle acquires images of the target vehicle within a preset circumference range in a multi-angle mode to obtain a multi-angle image set;

specifically, the target vehicle-based image acquisition device is used for acquiring images of a target vehicle in a preset circumferential range in real time at multiple angles, wherein the preset circumferential range is a range within a circle with a certain length as a radius and taking the target vehicle as a center of the circle, the multiple-angle acquisition is used for acquiring environmental information around the target vehicle from different angles such as the driving state of a driver, the moving intention of pedestrians outside the target vehicle, and the driving intentions of other vehicles outside the target vehicle, and the multi-angle image collection is an image information collection in an acquisition time period of the target vehicle, the pedestrians around the target vehicle, the vehicles around the target vehicle and the like.

For example, the preset circumference range may be 50m, and may be specifically set according to actual conditions, mainly by considering the density of the flow of people and the flow of vehicles in the surrounding environment.

Step S300: extracting key elements from the multi-angle image set to obtain a key element image set;

specifically, the key elements are related factors that affect the driving state of the target vehicle, and for example, the key elements are: the method comprises the steps that key elements of pedestrians, vehicles, head postures of pedestrians, people in vehicles and the like are extracted from a multi-angle image set, parameter information relevant to the driving state of a target vehicle is obtained, and further the key element image set is obtained, wherein the key element image set summarizes information influencing the driving state of the target vehicle together, so that the environment around the target vehicle can be selectively analyzed, and a foundation is laid for subsequent parameter result analysis based on a database.

Step S400: uploading the key element image set as input information to the multi-view prediction analysis model to perform interpretation analysis of different views, and obtaining multi-view element analysis results;

specifically, the multi-view predictive analysis model is used for analyzing the key element image set, and analyzing each key element in the key element image set from a plurality of different angles, wherein the multi-view element analysis result is the action state of each key element. The state of relevant factors influencing the action state of the target vehicle can be systematically analyzed through the model, so that the influence degree of the action state of the target vehicle is further analyzed.

Step S500: uploading the multi-view element analysis result to a probability grading track matching model for data processing, and generating a preset behavior track parameter sequence;

specifically, the probability classification trajectory matching model may predict the motion state trajectory of the next step according to the real-time state change of the element by processing the analysis result of the multi-view element, and further generate a preset behavior trajectory parameter sequence of each element, where the preset behavior trajectory parameter sequence is parameter information of a trajectory generated by the motion state of each predicted element after data processing is performed on the analysis result of each element by the probability classification trajectory matching model. By obtaining the predicted track parameter sequence of the elements which have influence on the driving state of the target vehicle, an accurate basis is provided for further finding out the parameter information which conflicts/influences with the driving action of the target vehicle, and the technical effect of improving the accuracy of parameter result prediction based on a database is achieved.

Step S600: obtaining a pre-driving behavior parameter of the target vehicle;

specifically, the pre-driving behavior parameter of the target vehicle is a relevant parameter characterizing a future driving state of the target vehicle.

For example, the pre-driving behavior parameters may be: a travel destination, a travel speed, etc. of the target driving parameter. A data basis is provided for subsequent comparison with a preset behavior trajectory of an element having an influence on the driving state of the target vehicle.

Step S700: performing parameter association degree matching on the preset behavior track parameter sequence and the pre-driving behavior parameters to generate first target behavior parameters corresponding to a first target parameter set;

specifically, the matching of the parameter association degree between the preset behavior track parameter sequence and the preset driving behavior parameter refers to matching the behavior track parameter sequence of the multi-view element and the preset driving behavior parameter of the target vehicle to obtain an element parameter coinciding with the preset driving behavior parameter, where the first target set is a parameter set corresponding to the preset behavior track parameter sequence coinciding with the preset driving behavior parameter, and the first target behavior parameter is the first target behavior parameter corresponding to the first target parameter set, so that a parameter that may affect the preset driving behavior of the target vehicle can be obtained, and these parameters can be accurately monitored to achieve the technical effect of improving the accuracy of parameter result prediction based on a database.

Step S800: and sending the first target behavior parameters to an intelligent driving end of the target vehicle, and monitoring the driving of the target vehicle.

Specifically, the intelligent driving end of the target vehicle is a platform connected with a database and used for driving control over the target vehicle, wherein the driving monitoring of the target vehicle refers to monitoring of the target vehicle in the driving process through the intelligent driving end after the first target behavior parameter is obtained.

Further, as shown in fig. 2, in the extracting key elements from the multi-angle image set, step S300 in this embodiment of the present application further includes:

step S310: carrying out multi-angle splitting on the multi-angle image set to obtain a first angle image, a second angle image and a third angle image;

step S320: acquiring in-vehicle audio information of the target vehicle based on the audio acquisition device of the target vehicle to obtain first audio data;

step S330: generating a first angle element set by extracting key elements from the first angle image and the first audio data;

step S340: extracting and analyzing the movement track of the target user set in the second angle image to obtain a second angle element set;

step S350: and extracting and analyzing the driving states of the target vehicle set in the third angle image to obtain a third angle element set.

Specifically, the multi-angle splitting of the multi-angle image set refers to splitting of the multi-angle image set according to different angles, and the multi-angle image set can be divided, so that analysis is more targeted. Wherein the first angle image is an image relating to the driver's own driving intention, the second angle image is an image relating to the pedestrian movement intention outside the vehicle of the target vehicle, and the third angle image is an image relating to the driving intention of the other vehicle outside the vehicle of the target vehicle.

For example, the first angle image may be an image showing a traveling destination, a traveling speed, a head posture of the driver, and the like of the target vehicle, which reflect the target vehicle. The second angle image may be an image showing the head posture, the limb movement, and the movement track of the pedestrian outside the vehicle. The third angle image may be an image related to a driver state of another vehicle outside the vehicle, a vehicle traveling direction, and a vehicle traveling speed, which displays the target vehicle. Therefore, regular classification of the images in the multi-angle image set can be realized, and the efficiency and the accurate technical effect of subsequent analysis are improved.

Further, the first audio data is related audio information about a driving state in the vehicle of the target vehicle, the first angle element set is a set obtained by combining the first angle image and the first audio data and extracting key elements, and can reflect the driving state of the target vehicle, the second angle element set is a set obtained by extracting and analyzing a movement trajectory of the target user set in the second angle image, and the third angle element set is a set obtained by extracting and analyzing a driving state of the target vehicle set in the third angle image.

Further, as shown in fig. 3, in the obtained multi-perspective element analysis result, step S400 in this embodiment of the present application further includes:

step S410: the multi-view prediction analysis model corresponds to the multi-angle image set one by one, and comprises a first view prediction analysis model, a second view prediction analysis model and a third view prediction analysis model;

step S420: uploading the second angle element set to the second view angle prediction analysis model, and performing prediction analysis on the target moving direction in the second angle element set to obtain a first analysis result;

step S430: performing predictive analysis on the target moving speed in the second angle element set to obtain a second analysis result;

step S440: and performing data union processing on the first analysis result and the second analysis result to generate a second visual angle element analysis result.

Specifically, the multi-view predictive analysis model is a model for analyzing and predicting the multi-angle image set, wherein the first view predictive analysis model is a model for analyzing the driving state of the target vehicle corresponding to the first angle image set, the second view predictive analysis model is a model for analyzing the behavior state of the pedestrian outside the vehicle corresponding to the second angle image set, and the third view predictive analysis model is a model for analyzing the driving state of the other vehicle outside the target vehicle.

Further, the first analysis result is a result of analyzing the target moving direction in the second angle element set, and the moving direction of the pedestrian outside the vehicle can be predicted, and the second analysis result is a result of analyzing the target moving speed in the second angle element set, and the moving speed of the pedestrian outside the vehicle can be predicted. The merging of the first analysis result and the second analysis result refers to processing data by integrating the moving direction and the moving speed of the pedestrian outside the vehicle, and the second perspective element analysis result is an analysis result of the behavior state of the pedestrian outside the vehicle after integrating the moving direction and the moving speed of the pedestrian outside the vehicle. Because people and vehicles are mixed on the road, the change speed of the motion trail of the pedestrians is high, and the technical effect of accurately acquiring the action state of the pedestrians outside the vehicle can be realized by integrating the moving direction and the moving speed.

Further, the multi-view element analysis result includes a first view element analysis result, a second view element analysis result, and a third view element analysis result, where in the generating of the preset behavior trajectory parameter sequence, step S500 in this embodiment of the present application further includes:

step S510: based on the probability grading track matching model, carrying out track matching training on the first visual angle element analysis result, the second visual angle element analysis result and the third visual angle element analysis result in sequence to generate a first track probability value, a second track probability value and a third track probability value;

step S520: obtaining a first preset behavior track parameter corresponding to the first track probability value, a second preset behavior track parameter corresponding to the second track probability value and a third preset behavior track parameter corresponding to the third track probability value by performing parameter conversion on the track probability value;

step S530: and generating the preset behavior track parameter sequence according to the first preset behavior track parameter, the second preset behavior track parameter and the third preset behavior track parameter.

Specifically, the multi-view element analysis result includes a first view element analysis result, a second view element analysis result, and a third view element analysis result, where the first view element analysis result is an analysis result of the driving state of the target vehicle by the multi-view predictive analysis model, the second view element analysis result is an analysis result of the behavior state of the pedestrian outside the vehicle by the multi-view predictive analysis model by integrating two aspects of the moving direction and the moving speed of the pedestrian outside the vehicle, and the third view element analysis result is an analysis result of the driving state of the other vehicle outside the vehicle of the target vehicle by the multi-view predictive analysis model. After the multi-view element analysis result is obtained, a prediction analysis result is obtained on the motion states of the target vehicle, pedestrians outside the vehicle and other vehicles outside the target vehicle, the obtained result is further input into the probability grading track matching model, tracks predicted by the three view elements are analyzed, probability values of the three view elements acting according to the predicted tracks are obtained, further, parameter conversion is carried out on the track probability values, track parameter information of the three view element predicted action tracks is obtained, the preset action track parameter sequence is generated, a foundation is laid for comparison of the next step and the pre-driving action tracks of the target vehicle, and the technical effect of analyzing and predicting the parameters of a database from multiple angles is achieved. The probability grading track matching model is a mathematical logic model constructed on the basis of a neural network model, can be analyzed by using the characteristic of continuous convergence of mathematical data, and then outputs converged information based on machine learning, namely a simulation result output track probability value.

The step of performing parameter conversion on the track probability value refers to generating corresponding preset behavior track parameters on a predicted behavior track according to the track probability value, converting abstract prediction data information into specific visual parameter information, and preparing for rendering a subsequent preset behavior track parameter sequence to the space coordinate, wherein the preset behavior track parameter sequence refers to combining the first preset behavior track parameter, the second preset behavior track parameter and the third preset behavior track parameter to form all track parameter sequences which can affect the driving state of the target vehicle. The method and the device have the advantages that the prediction accuracy of the parameter result based on the database is improved, the reliable driving environment prediction is provided for driving, the accuracy of predicting the tracks of surrounding pedestrians or surrounding vehicles is improved, and the technical effect of the matching degree of the prediction result of the vehicle database and the actual situation is improved.

Further, step S510 in the embodiment of the present application further includes:

step S511: inputting the first view element analysis result into the probability grading track matching model as input information;

step S512: the probability grading track matching model is obtained by training a plurality of groups of training data, and each group of training data in the plurality of groups of training data comprises: the first visual angle element analysis result and identification information used for identifying a preset track probability value;

step S513: and training the probability grading track matching model to a convergence state, and acquiring output information, wherein the output information comprises the first track probability value corresponding to the first visual angle element analysis result.

Specifically, the probabilistic hierarchical trajectory matching model is a neural network model in machine learning, and a neural network is a complex neural network system formed by widely connecting a large number of simple processing units (called neurons), reflects many basic features of human brain functions, and is a highly complex nonlinear dynamical learning system. Neural network models are described based on mathematical models of neurons. Artificial neural networks are a description of the first-order characteristics of the human brain system. Briefly, it is a mathematical model. And inputting the first visual angle element analysis result into a neural network model through training of a large amount of training data, and outputting the first track probability value corresponding to the first visual angle element analysis result.

Further, the training process is essentially a supervised learning process, each group of supervised data comprises the first visual angle element analysis result and identification information for identifying a preset track probability value, the first visual angle element analysis result is input into a neural network model, the model is continuously self-corrected and adjusted according to the identification information for identifying the preset track probability value, and the supervised learning is ended and the next group of data is supervised learning is carried out until the obtained output result is consistent with the identification information; and when the output information of the neural network model reaches a preset accuracy rate, finishing the supervised learning process. Through supervised learning of the model, the model can process the input information more accurately, so that a more accurate and suitable first track probability value is obtained, and a foundation is laid for improving the prediction accuracy of the parameter result based on the database.

Further, in the step S700 of performing parameter association degree matching on the preset behavior trajectory parameter sequence and the preset driving behavior parameter, in the embodiment of the present application, the step S further includes:

step S710: carrying out distributed display of space coordinates on the pre-driving behavior parameters to obtain a pre-driving behavior parameter curve;

step S720: rendering the preset behavior track parameter sequence to the space coordinate, and performing curve fitting on the pre-driving behavior parameter curve to obtain a first curve fitting result;

step S730: according to the first curve fitting result, obtaining the first target parameter set coincident with the pre-driving behavior parameters;

step S740: and obtaining the first target behavior parameter corresponding to the first target parameter set.

Specifically, the spatial coordinates are established in the surrounding environment of the target vehicle, the pre-driving behavior parameters are displayed in the spatial coordinates in a distributed manner, a pre-driving behavior parameter curve is obtained, further, the preset behavior trajectory parameter sequence is rendered to the spatial coordinates, and meanwhile, curve fitting is performed on the pre-driving behavior parameter curve, so that the first curve fitting result is obtained. The first curve fitting result refers to a part, coincident with the pre-driving behavior parameter curve, in the preset behavior track parameter sequence in the space coordinate, the first target parameter is combined into a parameter set of the coincident part, and the first target behavior parameter corresponding to the first target parameter set is further obtained, wherein the first target behavior parameter is used for being sent to an intelligent driving end of the target vehicle to monitor driving of the target vehicle. Therefore, the parameters which can influence the driving state of the target vehicle are found, and the driving of the target vehicle is accurately and specifically monitored.

Step S440 in the embodiment of the present application further includes:

step S441: obtaining head pose information of a target user in the second angle element set;

step S442: judging whether the head posture information and the target moving direction are consistent or not;

step S443: if the head posture information is not consistent with the target moving direction, determining a first correction parameter based on the head posture information;

step S444: and according to the first correction parameter, performing result correction on the second visual angle element analysis result.

Specifically, in the process of predicting the action track of the pedestrian, the head posture of the pedestrian is a very important pre-determination factor, the head posture is not necessarily consistent with the moving direction of the pedestrian, and when the head posture is not consistent with the moving direction of the pedestrian, the direction of the head posture may be the direction in which the pedestrian walks next or the direction in which the pedestrian observes the road condition, so when the action track of the pedestrian is predicted in the moving direction of the pedestrian, the action track of the pedestrian is further determined by considering the influence of the head posture.

Further, the head posture information may be head-facing direction information of a pedestrian, may be the same direction as a moving direction of the pedestrian, or may be a direction opposite to the moving direction of the pedestrian, and the first correction parameter may be a parameter that, when the head posture information does not coincide with the target moving direction, determines, based on the head posture information, that the predicted trajectory information of the pedestrian is generated, and that a result of the second perspective element analysis result is corrected. Therefore, the randomness of the movement of the pedestrian can be overcome, and the technical effect of accurately judging the action track of the pedestrian is realized.

Through the above description of the embodiments, those skilled in the art will clearly understand that the present application can be implemented by software plus necessary general-purpose hardware, and certainly can also be implemented by special-purpose hardware including special-purpose integrated circuits, special-purpose CPUs, special-purpose memories, special-purpose components and the like. Generally, functions performed by computer programs can be easily implemented by corresponding hardware, and specific hardware structures for implementing the same functions may be various, such as analog circuits, digital circuits, or dedicated circuits. However, for the present application, the implementation of a software program is more preferable. Based on such understanding, the technical solutions of the present application may be substantially embodied in the form of a software product, which is stored in a readable storage medium, such as a floppy disk, a usb disk, a removable hard disk, a ROM, a RAM, a magnetic disk, or an optical disk of a computer, and includes several instructions for causing a computer device to execute the method according to the embodiments of the present application.

To sum up, the parameter result prediction method based on the database provided by the embodiment of the application has the following technical effects:

1. because a behavior prediction analysis database is built, a multi-view prediction analysis model is embedded in the behavior prediction analysis database; the method comprises the steps that an image acquisition device based on a target vehicle acquires images of the target vehicle within a preset circumference range in a multi-angle mode to obtain a multi-angle image set; then extracting key elements from the multi-angle image set to obtain a key element image set; uploading the key element image set as input information to the multi-view prediction analysis model to perform interpretation analysis of different views, and obtaining multi-view element analysis results; then uploading the multi-view element analysis result to a probability grading track matching model for data processing, and generating a preset behavior track parameter sequence; obtaining a pre-driving behavior parameter of the target vehicle; further, performing parameter association degree matching on the preset behavior track parameter sequence and the pre-driving behavior parameters to generate first target behavior parameters corresponding to a first target parameter set; and sending the first target behavior parameters to an intelligent driving end of the target vehicle, and monitoring the driving of the target vehicle. The technical effects of improving the prediction accuracy of the parameter result based on the database, providing reliable driving environment prediction for driving, improving the accuracy of predicting the track of surrounding pedestrians or surrounding vehicles and improving the matching degree of the prediction result of the vehicle database and the actual situation are achieved.

2. By carrying out multi-angle splitting on the multi-angle image set and splitting the multi-angle image set according to different angles, the multi-angle image set can be analyzed in a more targeted manner after being divided, so that regular classification of the images in the multi-angle image set can be realized, and the efficiency and the accurate technical effect of subsequent analysis are improved.

3. By carrying out data union processing on the first analysis result and the second analysis result and considering the moving direction and the moving speed, the technical effect of accurately predicting the action condition of the pedestrian outside the vehicle can be realized.

Example two

Based on the same inventive concept as the database-based parameter result prediction method in the foregoing embodiment, as shown in fig. 4, the present application provides a database-based parameter result prediction system, wherein the system includes:

the system comprises a first building unit 11, a second building unit 11 and a behavior prediction analysis database, wherein the behavior prediction analysis database is embedded with a multi-view prediction analysis model;

a first obtaining unit 12, where the first obtaining unit 12 is configured to perform multi-angle collection on an image of a target vehicle within a preset circumference range based on an image collecting device of the target vehicle, so as to obtain a multi-angle image set;

a second obtaining unit 13, where the second obtaining unit 13 is configured to perform key element extraction on the multi-angle image set to obtain a key element image set;

a third obtaining unit 14, where the third obtaining unit 14 is configured to upload the key element image set as input information to the multi-view predictive analysis model for performing interpretation analysis on different views, and obtain a multi-view element analysis result;

the first generating unit 15 is configured to upload the multi-view element analysis result to a probability classification trajectory matching model for data processing, so as to generate a preset behavior trajectory parameter sequence;

a fourth obtaining unit 16, wherein the fourth obtaining unit 16 is used for obtaining the pre-driving behavior parameters of the target vehicle;

the first matching unit 17 is configured to perform parameter association degree matching on the preset behavior trajectory parameter sequence and the pre-driving behavior parameters, and generate first target behavior parameters corresponding to a first target parameter set;

the first monitoring unit 18 is configured to send the first target behavior parameter to an intelligent driving end of the target vehicle, and monitor driving of the target vehicle.

Further, the system further comprises:

a fifth obtaining unit, configured to split the multi-angle image set at multiple angles to obtain a first angle image, a second angle image, and a third angle image;

a sixth obtaining unit, configured to obtain first audio data by collecting in-vehicle audio information of the target vehicle based on an audio collecting device of the target vehicle;

a second generation unit configured to generate a first angle element set by performing key element extraction on the first angle image and the first audio data;

a seventh obtaining unit, configured to obtain a second angle element set by extracting and analyzing a movement trajectory of the target user set in the second angle image;

an eighth obtaining unit, configured to obtain a third angle element set by extracting and analyzing driving states of the target vehicle set in the third angle image.

Further, the system further comprises:

a ninth obtaining unit, configured to enable the multi-view prediction analysis model to correspond to the multi-angle image set one by one, and enable the multi-view prediction analysis model to include a first view prediction analysis model, a second view prediction analysis model, and a third view prediction analysis model; uploading the second angle element set to the second view angle prediction analysis model, and performing prediction analysis on the target moving direction in the second angle element set to obtain a first analysis result;

a tenth obtaining unit, configured to perform predictive analysis on the target movement speed in the second angle element set to obtain a second analysis result;

and the third generating unit is used for carrying out data union processing on the first analysis result and the second analysis result to generate a second visual angle element analysis result.

Further, the system further comprises: the first training unit is used for carrying out track matching training on the first visual angle element analysis result, the second visual angle element analysis result and the third visual angle element analysis result in sequence based on the probability grading track matching model to generate a first track probability value, a second track probability value and a third track probability value;

an eleventh obtaining unit, configured to obtain a first preset behavior track parameter corresponding to the first track probability value, a second preset behavior track parameter corresponding to the second track probability value, and a third preset behavior track parameter corresponding to the third track probability value by performing parameter conversion on a track probability value;

a fourth generating unit, configured to generate the preset behavior trajectory parameter sequence according to the first preset behavior trajectory parameter, the second preset behavior trajectory parameter, and the third preset behavior trajectory parameter.

Further, the system further comprises:

a first input unit configured to input the first perspective element analysis result as input information to the probability classification trajectory matching model;

a twelfth obtaining unit, configured to obtain, by training the probability classification trajectory matching model through multiple sets of training data, each set of training data in the multiple sets of training data including: the first visual angle element analysis result and identification information used for identifying a preset track probability value; and training the probability grading track matching model to a convergence state, and acquiring output information, wherein the output information comprises the first track probability value corresponding to the first visual angle element analysis result.

Further, the system further comprises:

a thirteenth obtaining unit, configured to perform distributed display of spatial coordinates on the pre-driving behavior parameter to obtain a pre-driving behavior parameter curve;

a fourteenth obtaining unit, configured to render the preset behavior trajectory parameter sequence to the spatial coordinate, and perform curve fitting on the pre-driving behavior parameter curve to obtain a first curve fitting result;

a fifteenth obtaining unit, configured to obtain the first target parameter set coinciding with the pre-driving behavior parameter according to the first curve fitting result;

a sixteenth obtaining unit, configured to obtain the first target behavior parameter corresponding to the first target parameter set.

Further, the system further comprises:

a seventeenth obtaining unit configured to obtain head posture information of a target user in the second angle element set;

a first judgment unit configured to judge whether the head posture information and the target moving direction are kept coincident;

a first determination unit configured to determine a first correction parameter based on the head posture information if the head posture information and the target moving direction do not coincide;

a first correction unit configured to perform result correction on the second perspective element analysis result according to the first correction parameter.

EXAMPLE III

Based on the same inventive concept as the database-based parameter result prediction method in the previous embodiment, the present application further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method according to the first embodiment.

Exemplary electronic device

The electronic device of the present application is described below with reference to figure 5,

based on the same inventive concept as the database-based parameter result prediction method in the foregoing embodiment, the present application also provides a database-based parameter result prediction system, including: a processor coupled to a memory, the memory storing a program that, when executed by the processor, causes the system to perform the steps of the method of embodiment one.

The electronic device 300 includes: processor 302, communication interface 303, memory 301. Optionally, the electronic device 300 may also include a bus architecture 304. Wherein, the communication interface 303, the processor 302 and the memory 301 may be connected to each other through a bus architecture 304; the bus architecture 304 may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The bus architecture 304 may be divided into an address bus, a data bus, a control bus, and the like. For ease of illustration, only one thick line is shown in FIG. 5, but this is not intended to represent only one bus or type of bus.

Processor 302 may be a CPU, microprocessor, ASIC, or one or more integrated circuits for controlling the execution of programs in accordance with the teachings of the present application.

The communication interface 303 may be any device, such as a transceiver, for communicating with other devices or communication networks, such as an ethernet, a Radio Access Network (RAN), a Wireless Local Area Network (WLAN), a wired access network, and the like.

The memory 301 may be, but is not limited to, a ROM or other type of static storage device that can store static information and instructions, a RAM or other type of dynamic storage device that can store information and instructions, an electrically erasable Programmable read-only memory (EEPROM), a compact-read-only-memory (CD-ROM) or other optical disk storage, optical disk storage (including compact disk, laser disk, optical disk, digital versatile disk, blu-ray disk, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer. The memory may be self-contained and coupled to the processor through a bus architecture 304. The memory may also be integral to the processor.

The memory 301 is used for storing computer-executable instructions for executing the present application, and is controlled by the processor 302 to execute. The processor 302 is configured to execute computer-executable instructions stored in the memory 301, so as to implement a database-based parameter result prediction method provided by the above-mentioned embodiment of the present application.

Those of ordinary skill in the art will understand that: the various numbers of the first, second, etc. mentioned in this application are for convenience of description and are not intended to limit the scope of this application nor to indicate the order of precedence. "and/or" describes the association relationship of the associated objects, meaning that there may be three relationships, e.g., a and/or B, which may mean: a exists alone, A and B exist simultaneously, and B exists alone. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship. "at least one" means one or more. At least two means two or more. "at least one," "any," or similar expressions refer to any combination of these items, including any combination of singular or plural items. For example, at least one (one ) of a, b, or c, may represent: a, b, c, a-b, a-c, b-c, or a-b-c, wherein a, b, c may be single or multiple.

In the above embodiments, the implementation may be wholly or partially realized by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the procedures or functions described in accordance with the present application are generated, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, from one website site, computer, server, or data center to another website site, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device including one or more available media integrated servers, data centers, and the like. The usable medium may be a magnetic medium (e.g., floppy Disk, hard Disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., Solid State Disk (SSD)), among others.

The various illustrative logical units and circuits described in this application may be implemented or operated through the design of a general purpose processor, a digital signal processor, an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof. A general-purpose processor may be a microprocessor, but, in the alternative, the processor may be any conventional processor, controller, microcontroller, or state machine. A processor may also be implemented as a combination of computing devices, e.g., a digital signal processor and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a digital signal processor core, or any other similar configuration.

The steps of a method or algorithm described in this application may be embodied directly in hardware, in a software element executed by a processor, or in a combination of the two. The software cells may be stored in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. For example, a storage medium may be coupled to the processor such the processor can read information from, and write information to, the storage medium. In the alternative, the storage medium may be integral to the processor. The processor and the storage medium may reside in an ASIC, which may be disposed in a terminal. In the alternative, the processor and the storage medium may reside in different components within the terminal. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

Although the present application has been described in conjunction with specific features and embodiments thereof, it will be evident that various modifications and combinations can be made thereto without departing from the spirit and scope of the application. Accordingly, the specification and figures are merely exemplary of the application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of the application. It will be apparent to those skilled in the art that various changes and modifications may be made in the present application without departing from the scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the present application and its equivalent technology, it is intended that the present application include such modifications and variations.

Claims

1. A database-based method for predicting parameter outcomes, the method comprising:

building a behavior prediction analysis database, wherein a multi-view prediction analysis model is embedded in the behavior prediction analysis database;

the method comprises the steps that an image acquisition device based on a target vehicle acquires images of the target vehicle within a preset circumference range in a multi-angle mode to obtain a multi-angle image set;

extracting key elements from the multi-angle image set to obtain a key element image set;

uploading the key element image set as input information to the multi-view prediction analysis model to perform interpretation analysis of different views, and obtaining multi-view element analysis results;

uploading the multi-view element analysis result to a probability grading track matching model for data processing, and generating a preset behavior track parameter sequence;

obtaining a pre-driving behavior parameter of the target vehicle;

performing parameter association degree matching on the preset behavior track parameter sequence and the pre-driving behavior parameters to generate first target behavior parameters corresponding to a first target parameter set;

and sending the first target behavior parameters to an intelligent driving end of the target vehicle, and monitoring the driving of the target vehicle.

2. The method as claimed in claim 1, wherein said performing key element extraction on said set of multi-angle images comprises:

the multi-angle image set is split in multiple angles to obtain a first angle image, a second angle image and a third angle image;

acquiring in-vehicle audio information of the target vehicle based on the audio acquisition device of the target vehicle to obtain first audio data;

generating a first angle element set by extracting key elements from the first angle image and the first audio data;

extracting and analyzing the movement track of the target user set in the second angle image to obtain a second angle element set;

and extracting and analyzing the driving states of the target vehicle set in the third angle image to obtain a third angle element set.

3. The method of claim 2, wherein the obtaining multiple perspective element analysis results comprises:

the multi-view prediction analysis model corresponds to the multi-angle image set one by one, and comprises a first view prediction analysis model, a second view prediction analysis model and a third view prediction analysis model;

uploading the second angle element set to the second view angle prediction analysis model, and performing prediction analysis on the target moving direction in the second angle element set to obtain a first analysis result;

performing predictive analysis on the target moving speed in the second angle element set to obtain a second analysis result;

and performing data union processing on the first analysis result and the second analysis result to generate a second visual angle element analysis result.

4. The method of claim 3, wherein the multi-perspective element analysis results comprise a first perspective element analysis result, a second perspective element analysis result, and a third perspective element analysis result, and wherein the generating the preset behavior trace parameter sequence comprises:

based on the probability grading track matching model, carrying out track matching training on the first visual angle element analysis result, the second visual angle element analysis result and the third visual angle element analysis result in sequence to generate a first track probability value, a second track probability value and a third track probability value;

obtaining a first preset behavior track parameter corresponding to the first track probability value, a second preset behavior track parameter corresponding to the second track probability value and a third preset behavior track parameter corresponding to the third track probability value by performing parameter conversion on the track probability value;

and generating the preset behavior track parameter sequence according to the first preset behavior track parameter, the second preset behavior track parameter and the third preset behavior track parameter.

5. The method of claim 4, wherein the method comprises:

inputting the first view element analysis result into the probability grading track matching model as input information;

the probability classification track matching model is obtained by training a plurality of groups of training data, and each group of training data in the plurality of groups of training data comprises: the first visual angle element analysis result and identification information used for identifying a preset track probability value;

and training the probability grading track matching model to a convergence state, and obtaining output information, wherein the output information comprises the first track probability value corresponding to the first visual angle element analysis result.

6. The method of claim 1, wherein the performing of parameter correlation matching on the sequence of preset behavior trajectory parameters and the pre-driving behavior parameters comprises:

carrying out distributed display of space coordinates on the pre-driving behavior parameters to obtain a pre-driving behavior parameter curve;

rendering the preset behavior track parameter sequence to the space coordinate, and performing curve fitting on the pre-driving behavior parameter curve to obtain a first curve fitting result;

according to the first curve fitting result, obtaining the first target parameter set coincident with the pre-driving behavior parameters;

and obtaining the first target behavior parameters corresponding to the first target parameter set.

7. The method of claim 3, wherein the method comprises:

obtaining head pose information of a target user in the second angle element set;

judging whether the head posture information and the target moving direction are consistent or not;

if the head posture information is not consistent with the target moving direction, determining a first correction parameter based on the head posture information;

and according to the first correction parameter, performing result correction on the second visual angle element analysis result.

8. A database-based parametric result prediction system, the system comprising:

the system comprises a first building unit, a second building unit and a third building unit, wherein the first building unit is used for building a behavior prediction analysis database, and a multi-view prediction analysis model is embedded in the behavior prediction analysis database;

the system comprises a first obtaining unit, a second obtaining unit and a third obtaining unit, wherein the first obtaining unit is used for carrying out multi-angle collection on images of a target vehicle within a preset circumference range based on an image collecting device of the target vehicle to obtain a multi-angle image set;

a second obtaining unit, configured to perform key element extraction on the multi-angle image set to obtain a key element image set;

a third obtaining unit, configured to upload the key element image set as input information to the multi-view predictive analysis model to perform interpretation analysis on different views, so as to obtain a multi-view element analysis result;

the first generation unit is used for uploading the multi-view element analysis result to a probability grading track matching model for data processing to generate a preset behavior track parameter sequence;

a fourth obtaining unit configured to obtain a pre-driving behavior parameter of the target vehicle;

the first matching unit is used for performing parameter association degree matching on the preset behavior track parameter sequence and the pre-driving behavior parameters to generate first target behavior parameters corresponding to a first target parameter set;

and the first monitoring unit is used for sending the first target behavior parameters to an intelligent driving end of the target vehicle and monitoring the driving of the target vehicle.

9. A database-based parametric result prediction system, comprising: a processor coupled to a memory, the memory for storing a program that, when executed by the processor, causes a system to perform the steps of the method of any of claims 1 to 7.

10. A computer-readable storage medium, characterized in that the storage medium has stored thereon a computer program which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 7.