CN116932518A

CN116932518A - Positioning point quality evaluation method, device, equipment and computer readable storage medium

Info

Publication number: CN116932518A
Application number: CN202210327137.2A
Authority: CN
Inventors: 廖杰; 李子希; 章文涛; 余重玲
Original assignee: Fengtu Technology Shenzhen Co Ltd
Current assignee: Fengtu Technology Shenzhen Co Ltd
Priority date: 2022-03-30
Filing date: 2022-03-30
Publication date: 2023-10-24

Abstract

The embodiment of the application provides a positioning point quality evaluation method, a positioning point quality evaluation device, positioning point quality evaluation equipment and a computer readable storage medium, wherein the positioning point quality evaluation method comprises the following steps: acquiring a target positioning point to be evaluated; performing deviation rectifying treatment on the target locating points according to preset standard track information to obtain deviation rectifying locating points; and determining a quality evaluation result of the target positioning point according to the target positioning point and the deviation correcting positioning point. According to the technical scheme provided by the embodiment of the application, the deviation rectifying locating point is obtained by carrying out deviation rectifying treatment on the target locating point, and because the deviation rectifying locating point contains the characteristics of the target locating point in the longitudinal dimension, namely the characteristic of richer dimension of the target locating point is fused to realize quality assessment, the analysis of the target locating point can be further realized, so that a more accurate assessment result is obtained, and the screening of high-quality locating point data is more convenient.

Description

Positioning point quality evaluation method, device, equipment and computer readable storage medium

Technical Field

The embodiment of the application relates to the technical field of data quality evaluation, in particular to a positioning point quality evaluation method, a positioning point quality evaluation device, positioning point quality evaluation equipment and a computer readable storage medium.

Background

With the popularity of positioning devices, there are various schemes for acquiring track information of a target object. Typically, the track information is composed of a series of anchor points obtained by continuous sampling. However, due to the influence of the factors of the acquisition equipment and the uncontrollable interference of external conditions, in the process of actually sampling the track information, the quality of the positioning points obtained by actually sampling is uneven, and the problems of drift, redundancy, deletion, dislocation and the like are often caused compared with the real data. In order to provide high-quality traffic information service, quality evaluation is often required to be performed on the positioning point data acquired by various modes, the positioning point data with low quality is removed, and only the positioning point data with high quality is reserved.

However, the quality evaluation of the positioning point data is usually only performed by transversely comparing the difference between the acquired positioning point data, so that the quality evaluation result of the positioning point data is not ideal, and the finally determined high-quality positioning point data is still not accurate enough.

Disclosure of Invention

The embodiment of the application provides a locating point quality assessment method, a locating point quality assessment device, locating point quality assessment equipment and a computer readable storage medium, and aims to solve the technical problems that an assessment result of an existing locating point quality assessment method is not ideal enough, and finally high-quality locating point data determined by the existing locating point quality assessment method is not accurate enough.

In one aspect, an embodiment of the present application provides a positioning point quality evaluation method, including:

acquiring a target positioning point to be evaluated;

performing deviation rectifying treatment on the target locating points according to preset standard track information to obtain deviation rectifying locating points;

and determining a quality evaluation result of the target positioning point according to the target positioning point and the deviation correcting positioning point.

As an optional embodiment of the present application, the determining, according to the target positioning point and the deviation correcting positioning point, a quality evaluation result of the target positioning point includes:

extracting sequence characteristics of the deviation correcting locating points according to the parameter information of the deviation correcting locating points; the parameter information at least comprises one of position information, speed information and time information;

determining deviation rectifying characteristics of the deviation rectifying locating points according to the difference between the parameter information of the target locating points and the parameter information of the deviation rectifying locating points;

and inputting the sequence characteristics and the deviation correction characteristics into a trained network model for processing, and outputting a quality evaluation result of the target positioning point.

As an optional embodiment of the present application, before the sequence feature and the correction feature are input to a trained network model for processing and the quality evaluation result of the target positioning point is output, the method includes:

Determining the track length corresponding to the target positioning point according to the position information of the adjacent positioning points in the target positioning point;

acquiring an associated network model corresponding to the track length; the association network model is obtained by training based on the track length of a training sample and a training label in advance;

setting the associated network model as a trained network model

As an optional embodiment of the present application, the extracting the sequence feature of the deviation correcting locating point according to the parameter information of the deviation correcting locating point includes:

determining a first local characteristic of the deviation correcting locating point according to the difference of parameter information of adjacent locating point pairs in the deviation correcting locating point;

determining a first global feature of the deviation correcting locating point according to a data statistics result of the parameter information of the deviation correcting locating point; the data statistics result at least comprises one of an average number, a median and a mode;

and setting the first local feature and the first global feature as sequence features of the deviation correcting locating points.

As an optional embodiment of the present application, the determining, according to a difference between the parameter information of the target positioning point and the parameter information of the deviation correcting positioning point, a deviation correcting feature of the deviation correcting positioning point includes:

Setting the difference value between the parameter information of the target locating point and the parameter information of the deviation correcting locating point as a second local characteristic of the deviation correcting locating point;

determining a second global feature of the deviation correcting locating point according to the data statistics result of the difference value; the data statistics result at least comprises one of an average number, a median and a mode;

and setting the second local feature and the second global feature as correction features of the correction positioning points.

As an alternative embodiment of the present application, the trained network model includes a trained long-short term memory neural network and a trained full connection layer;

inputting the sequence feature and the deviation correction feature into a trained network model for processing, and outputting the quality evaluation result of the target positioning point comprises the following steps:

splicing the local features in the sequence features and the local features in the deviation correction features to obtain the local features of the target positioning points;

splicing the global features in the sequence features and the global features in the deviation correcting features to obtain global features of the target positioning points;

inputting the local characteristics of the target positioning points into the trained long-short-period memory neural network, and outputting the processed local characteristics;

And inputting the processed local features and the global features of the target positioning points into the trained full-connection layer, and outputting the quality evaluation result of the target positioning points.

dividing the target positioning points and the deviation correcting positioning points respectively to obtain a target positioning point set and a deviation correcting positioning point set corresponding to the target positioning point set;

determining quality evaluation results of all the target positioning point sets according to the target positioning points in the target positioning point sets and the deviation correcting positioning points in the deviation correcting positioning point sets;

and determining the quality evaluation result of the target positioning point according to the quality evaluation result of each target positioning point set.

On the other hand, the embodiment of the application also provides a positioning point quality assessment method, which comprises the following steps:

the acquisition module is used for acquiring a target positioning point to be evaluated;

the deviation rectifying module is used for carrying out deviation rectifying processing on the target locating points according to preset standard track information to obtain deviation rectifying locating points;

And the evaluation module is used for determining a quality evaluation result of the target positioning point according to the target positioning point and the deviation correcting positioning point.

On the other hand, the embodiment of the application also provides a positioning point quality evaluation device, which comprises a processor, a memory and a positioning point quality evaluation program stored in the memory and capable of running on the processor, wherein the processor executes the positioning point quality evaluation program to realize the steps in the positioning point quality evaluation method.

In another aspect, an embodiment of the present application further provides a computer readable storage medium, where a positioning point quality evaluation program is stored on the computer readable storage medium, where the positioning point quality evaluation program is executed by a processor to implement the steps in the positioning point quality evaluation method described above.

According to the technical scheme provided by the embodiment of the application, the deviation correcting processing is carried out on the target positioning points to obtain the deviation correcting positioning points, and compared with the technical scheme in the prior art that the quality evaluation is carried out on the target positioning points by simply transversely comparing the differences between different target positioning points, the quality evaluation is finally carried out on the target positioning points according to the target positioning points and the deviation correcting positioning points by integrating the characteristics of different dimensions of the target positioning points, so that the analysis on the target positioning points can be more deeply realized, more accurate evaluation results are obtained, and the screening of high-quality positioning point data is more convenient.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the description of the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

Fig. 1 is a schematic diagram of an implementation scenario of quality assessment of a positioning point according to an embodiment of the present application;

fig. 2 is a schematic flow chart of a positioning point quality evaluation method according to an embodiment of the present application;

FIG. 3 is a flowchart illustrating a step of determining a quality evaluation result according to a target positioning point and a deviation correcting positioning point according to an embodiment of the present application;

FIG. 4 is a flowchart illustrating steps for determining a quality evaluation result according to a sequence feature and a correction feature according to an embodiment of the present application;

FIG. 5 is a flowchart illustrating steps for extracting sequence features according to an embodiment of the present application;

FIG. 6 is a flowchart illustrating a step of extracting correction features according to an embodiment of the present application;

FIG. 7 is a flowchart illustrating steps for determining a quality assessment result using a specific network model according to an embodiment of the present application;

FIG. 8 is a flowchart illustrating another step of determining a quality evaluation result according to a target positioning point and a deviation correcting positioning point according to an embodiment of the present application;

fig. 9 is a schematic structural diagram of a positioning point quality evaluation device according to an embodiment of the present application;

fig. 10 is a schematic structural diagram of a positioning point quality evaluation device according to an embodiment of the present application.

Detailed Description

The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be encompassed by the present application.

In the embodiments of the present application, the term "exemplary" is used to mean "serving as an example, instance, or illustration. Any embodiment described as "exemplary" in this disclosure is not necessarily to be construed as preferred or advantageous over other embodiments. The following description is presented to enable any person skilled in the art to make and use the application. In the following description, details are set forth for purposes of explanation. It will be apparent to one of ordinary skill in the art that the present application may be practiced without these specific details. In other instances, well-known structures and processes have not been described in detail so as not to obscure the description of the application with unnecessary detail. Thus, the present application is not intended to be limited to the embodiments shown, but is to be accorded the widest scope consistent with the principles and features disclosed herein.

The embodiment of the application provides a positioning point quality evaluation method, a positioning point quality evaluation device, positioning point quality evaluation equipment and a computer readable storage medium, and the positioning point quality evaluation method, the positioning point quality evaluation device and the positioning point quality evaluation equipment are respectively described in detail below.

The positioning point quality evaluation method in the embodiment of the application is deployed on the positioning point quality evaluation device in the form of a program, the positioning point quality evaluation device is installed in the positioning point quality evaluation equipment in the form of a processor, and after the positioning point quality evaluation device in the positioning point quality evaluation equipment acquires a target positioning point to be evaluated, the positioning point quality evaluation device executes the program corresponding to the positioning point quality evaluation method to execute the following steps: and carrying out deviation rectifying processing on the target locating points according to preset standard track information to obtain deviation rectifying locating points, determining a quality evaluation result of the target locating points according to the target locating points and the deviation rectifying locating points, and outputting the quality evaluation result.

As shown in fig. 1, fig. 1 is a schematic view of an implementation scenario of anchor point quality assessment provided by an embodiment of the present application, where the schematic view of the implementation scenario of anchor point quality assessment provided by the embodiment of the present application includes a target terminal 100, and an anchor point acquisition system 200 and an anchor point quality assessment device 300 installed on the target terminal. The target terminal 100 mainly refers to a mobile phone mobile terminal, a vehicle terminal, and other terminal devices that need to provide traffic information services. The positioning point acquisition system 200 includes a plurality of different data acquisition devices, and each data acquisition device periodically completes the acquisition of positioning information of the target terminal, so as to obtain a series of positioning points which can be used for describing the moving track of the mobile terminal. Specifically, the data acquisition device may include a GPS positioning device, an inertial navigation positioning device, and the like, and may also include other devices with positioning data acquisition functions, which are not described herein.

Furthermore, due to the performance difference of the data acquisition devices, the positioning points acquired by different data acquisition devices have different degrees of deviation, namely different quality, the positioning points acquired by different data acquisition devices are input into the positioning point quality evaluation device 300, the positioning point quality evaluation device 300 performs quality evaluation on the positioning point data acquired by each data acquisition device, the quality evaluation of each positioning point data is output, and the positioning point data with the best quality evaluation is screened out from the quality evaluation and provided for the target terminal, so that high-quality traffic information service is realized.

It should be noted that, the schematic view of the implementation scenario of the setpoint quality assessment described in the above embodiment is for more clearly describing the technical solution of the embodiment of the present application, and does not constitute a limitation of the technical solution of the setpoint quality assessment provided by the embodiment of the present application.

Based on the implementation scene diagram of the anchor point quality evaluation, a specific embodiment of an anchor point quality evaluation method is provided.

As shown in fig. 2, fig. 2 is a schematic step flow diagram of a positioning point quality evaluation method according to an embodiment of the present application, where the positioning point quality evaluation method according to the embodiment of the present application includes steps 201 to 203:

And 201, acquiring a target positioning point to be evaluated.

In combination with the implementation scene schematic diagram provided by the foregoing, it can be known that the embodiment of the application is mainly used for screening out the positioning points with better quality, that is, the positioning points closer to the real track, from a series of positioning points acquired by different data acquisition devices. In other words, in the embodiment of the present application, there are multiple groups of actually acquired target positioning points to be evaluated, which correspond to different data acquisition devices, and each group of target positioning points is formed by a series of positioning points that are periodically acquired by the corresponding data acquisition device and can be used for describing the movement track of the target terminal. Specifically, in the embodiment of the application, a processing procedure of the positioning point quality evaluation device on a group of target positioning points is described by taking a group of target positioning points as an example, and the quality evaluation results of the group of target positioning points can be obtained by adopting the same processing procedure aiming at other target positioning points, so that a group of target positioning points with optimal quality evaluation can be conveniently screened and reserved from the quality evaluation results of each group of target positioning points in the follow-up process, and the method is used for providing high-quality traffic information service.

Typically, these positioning points are in the form of coordinates, describing the location information of these points. Of course, as an alternative embodiment of the present application, in addition to the coordinates of the parameter information describing the position, the positioning points will also typically carry other parameter information, such as a time stamp describing the parameter information of time, a real-time acquisition speed describing the parameter information of speed, etc. The time stamps of the target positioning points can determine the sequence of the target positioning points acquired, so that the sequence relation of the target positioning points can be determined.

In particular, the parameter information of the anchor point may be described by a vector of high dimension. For example, a certain positioning point may be described by a vector (a, B, C, D), which indicates that the coordinate of the positioning point is (a, B), the moment of acquiring the positioning point is C, and the speed of the target terminal when acquiring the positioning point is D.

Further, it should be noted that, for any parameter information, different anchor points need to be unified in terms of units and definitions, for example, as a possible solution, regarding time information, considering that a day includes 86400 seconds, a current time may be described by a value between [0, 86400 ], for example, a value of C is 0 may indicate that the sampling time is zero, and a value of C is 43200 may indicate that the sampling time is 12 noon. And for the speed information, if the speed of the target terminal is 10m/s when the positioning point is acquired, the numerical value of D is 10, and if the speed of the target terminal is 8m/s when the positioning point is acquired, the numerical value of D is 8.

202, performing deviation rectifying processing on the target locating points according to preset standard track information to obtain deviation rectifying locating points.

In the embodiment of the application, the standard track information describes a real road in a digital form, specifically, in a computer program, the standard track information usually exists in the form of an equation describing two road boundaries or in the form of an equation describing a road center line plus a numerical value describing a road width value, for the former, the position relationship between a locating point and the road can be determined by substituting the coordinates of the locating point into the equation describing the two road boundaries, so as to judge whether the locating point deviates from the road, and for the latter, the position relationship between the locating point and the road can be determined by calculating the distance between the locating point and the road center line and according to the relationship between the distance and the numerical value describing the road width value, so as to judge whether the locating point deviates from the road.

In the embodiment of the application, the deviation correction processing is performed on the target positioning points according to the standard track information, namely, the deviation of part of the positioning points which are obviously deviated from the road in the target positioning points is corrected to the road, and the positioning points which are in the range of the road are not adjusted. Specifically, as a feasible embodiment of the present application, the method of calculating the vertical line may be used to determine the drop foot of the target positioning point relative to the standard track, if the standard track information is in the form of equation of the road center line plus the numerical value for describing the road width value, the drop foot is the deviation correcting positioning point, if the standard track information is in the form of equation for describing two road boundaries, the drop foot of the target positioning point relative to the standard track is two, the drop foot of the target positioning point relative to the two road boundaries is respectively, and at this time, the midpoint of the two drop feet is the deviation correcting positioning point.

Specifically, taking the example that the target positioning points include M, N, P, Q, at this time, the deviation correcting positioning points should theoretically include M ', N', P ', Q', but only the target positioning points N and Q are found to deviate from the road obviously by comparing with the standard track information, and the targets M and P are on the road, that is, the target positioning point M and the deviation correcting positioning point M 'corresponding thereto are identical points, the target positioning point P and the deviation correcting positioning point P' corresponding thereto are identical points, that is, the deviation correcting positioning points M ', N', P ', Q' can be basically and equally understood to be formed by the target positioning points M and P not deviating from the road and the positioning points N 'and Q' obtained by deviation correcting the target positioning points deviating from the road.

It should be noted that, the correction is only performed on the position information in the positioning points, and is not performed on other parameter information, that is, compared with the corresponding positioning points in the target positioning points, such as N and Q, the correction positioning points N 'and Q' are only different in coordinate information, and other parameter information remains the same.

In addition, it should be noted that, the positioning point of the deviation correcting on the road obtained in the embodiment of the present application may also be not real positioning information, and a certain deviation may still exist in the position information.

In the embodiment of the present application, the implementation scheme of performing the deviation rectifying process on the target locating point according to the preset standard track information to obtain the deviation rectifying locating point may refer to the following fig. 8 and the explanation thereof.

And 203, determining a quality evaluation result of the target positioning point according to the target positioning point and the deviation correcting positioning point.

In the embodiment of the application, the quality evaluation result of the target positioning point can be used for describing the quality of the target positioning point, and in general, 3 grades can be used for describing, for example, the quality and the difference, and of course, specific quantization indexes can also be used, for example, a percentile can be used, and the higher the score is, the better the quality of the target positioning point is. Further, the better the quality of the target positioning points, the closer the group of the target positioning points to the real track, and the high-quality traffic information service can be provided by the group of the target positioning points later.

Compared with the prior art that the quality evaluation of the target positioning points is realized only by relying on the data of each group of the target positioning points, the technical scheme provided by the embodiment of the application additionally combines the information of the deviation correcting positioning points after deviation correction to realize the quality evaluation result of the target positioning points, enriches the data dimension of the target positioning points, and can evaluate the quality of the track more comprehensively. Specifically, the quality evaluation result of determining the target positioning point according to the target positioning point and the deviation correction positioning point can be realized based on the technical means of feature extraction, and the specific implementation scheme can refer to the following fig. 3 and the explanation thereof.

Further, as an alternative embodiment of the present application, the anchor point quality evaluation device may further divide the target anchor points for describing the complete moving track of the target terminal into a plurality of sets, for example, divide the target anchor points according to the set number of anchor points, that is, divide the continuously preset number of the target anchor points into the same set, and individually perform quality evaluation with respect to the anchor points in each set, thereby obtaining a quality evaluation result of each track segment, so as to facilitate subsequent combination of different track segments with optimal quality evaluation results according to the quality evaluation result of each track segment in each set of target anchor points, thereby obtaining better track information. A specific implementation may be found in the following description of fig. 7 and its explanation.

Fig. 3 is a schematic flowchart of a step of determining a quality evaluation result according to a target positioning point and a deviation correcting positioning point according to an embodiment of the present application, which is described in detail below.

In the embodiment of the application, an implementation scheme for realizing a quality evaluation result based on feature extraction of positioning points is provided, which specifically comprises steps 301 to 303:

and 301, extracting the sequence characteristics of the deviation correcting locating points according to the parameter information of the deviation correcting locating points.

In the embodiment of the present application, as known from the foregoing description, the parameter information of the deviation correcting locating point is consistent with the parameter of the target locating point, and may include position information, time information, speed information, and so on, so that the sequence feature of the deviation correcting locating point may be extracted based on at least one of the position information, the speed information, and the time information of the deviation correcting locating point.

Further, the sequence features mainly include local features for describing single anchor points or a small section of track information determined based on the parameter information of the adjacent deviation correcting anchor points, and global features for describing complete track information determined based on the parameter information of all deviation correcting anchor points, and at this time, the implementation scheme for specifically extracting the sequence features can refer to the following fig. 5 and the explanation content thereof.

In addition, it should be noted that, in order to better extract the feature of the correction positioning point, the correction positioning point includes not only the positioning point after correction, but also the positioning point that is not subjected to correction processing in the target positioning point, and the embodiment provided above is taken as an example, that is, in this step, the correction positioning point includes M, N ', P, Q'.

302, determining deviation correcting characteristics of the deviation correcting locating points according to the difference between the parameter information of the target locating points and the parameter information of the deviation correcting locating points.

In addition to separately extracting features of the correction positioning points, the embodiment of the application additionally considers the difference between the target positioning points and the correction positioning points, taking the target positioning points M, N, P, Q and the correction positioning points M ', N', P ', Q' as examples. The difference between the target positioning point and the deviation correcting positioning point mainly refers to the difference between the parameter information of M ' in the deviation correcting positioning point and the parameter information of M in the deviation correcting positioning point, the difference between the parameter information of N ' in the deviation correcting positioning point and the parameter information of P ' in the deviation correcting positioning point and the parameter information of Q in the deviation correcting positioning point. It should be noted that, for the target positioning points, such as M and P, which are not deviated from the road, since the corresponding deviation correcting positioning points M 'and P' are the same as themselves, that is, the target positioning point M and the corresponding deviation correcting positioning point M 'are the same point, the target positioning point P and the corresponding deviation correcting positioning point P' are the same point deviation correcting positioning point M, and therefore, for these positioning points, the difference value of the parameter information can be regarded as 0. Specifically, taking the example that the deviation correcting feature includes a deviation correcting distance, where the deviation correcting distance refers to a linear distance between the position information of the deviation correcting locating point and the target locating point, for example, MM ', NN ', and so on, and since the deviation correcting locating point M ' and the target locating point M are the same point, a value of the linear distance MM ' between the position information of the deviation correcting locating point M ' and the target locating point M is 0. And further arranging the correction distances corresponding to the correction positioning points in sequence to obtain correction characteristics.

Furthermore, similar to the sequence features described above, the above correction features only take into account the difference between the correction positioning points and the target positioning points alone, and in fact, the correction features may take into account the difference between the correction positioning points and the target positioning points as a whole, including some integral features, such as the correction ratio of the positioning points, the average correction distance, and so on. The deviation correcting proportion of the locating points can be understood as the proportion of the locating points subjected to deviation correcting treatment to all the locating points. For example, taking the foregoing target positioning point M, N, P, Q as an example, the target positioning points include four total target positioning points, wherein two target positioning points N and Q are subjected to deviation correction processing, and two target positioning points M and P are not subjected to deviation correction processing, the deviation correction ratio of the positioning points is 50%, and the average deviation correction distance can be understood as the ratio of the sum of the deviation correction distances to the number of positioning points, that is, the average deviation correction distance is (MM '+nn' +pp '+qq')/4. The implementation of the correction feature obtained by specific extraction can be referred to later in fig. 6 and the explanation thereof.

303, inputting the sequence feature and the deviation correction feature into a trained network model for processing, and outputting a quality evaluation result of the target positioning point.

In the embodiment of the application, the trained network model is based on the thought of machine learning, and a great number of annotated training samples are utilized to pre-train a functional relationship which can be used for describing the internal connection between the characteristics and the quality evaluation results of the locating points. After the network model is trained, the sequence characteristics and the deviation correction characteristics are input into the network model for corresponding processing, and then the quality evaluation result of the target positioning point can be directly output. Among these, the ideas of machine learning include, but are not limited to, random forests, support vector machines, and neural networks, among others. In addition, the training process is similar to the quality evaluation process, the training samples need to be subjected to the same correction and feature extraction, so that in order to facilitate understanding of the implementation manner of training to obtain a trained network model in the application, the implementation process of training to obtain a model function by using the idea of machine learning is briefly described, and in particular, taking a deep neural network model as an example, the specific training process is as follows:

(1) constructing an initialized depth network model, wherein parameters in the model are randomly initialized;

(2) acquiring a training sample, wherein the training sample comprises a historical sampling locating point and a corresponding prediction quality evaluation label;

(3) Performing deviation rectifying processing on the historical sampling locating points according to preset standard track information to obtain training deviation rectifying locating points, and obtaining sequence characteristics and deviation rectifying characteristics of the training deviation rectifying locating points based on parameter information of the training deviation rectifying locating points and differences between the parameter information of the training deviation rectifying locating points and the parameter information of the historical sampling locating points;

(3) inputting the sequence characteristics of the training deviation correcting locating points and the deviation correcting characteristics into the initialized depth network model, and outputting a response quality assessment result;

(4) updating parameters in the depth network model according to the difference between the response quality evaluation result and the prediction quality evaluation label and a back propagation algorithm until the sequence characteristics of the training deviation correction positioning points and the deviation correction characteristics are input into the updated depth network model, and setting the current updated depth network model as a trained network model when the difference between the output response quality evaluation result and the prediction quality evaluation label is smaller than a preset threshold value.

Further, as an alternative embodiment of the present application, the anchor point quality assessment device may further adopt different machine learning trained network functions to better implement quality assessment of the target anchor point based on the type of the target anchor point. For example, for a positioning point with a more complex track, that is, a larger data volume, or a positioning point with a cross in the track, a neural network with a richer structure can be adopted, for example, through a long-short-period memory neural network, the sequence of each positioning point is fully considered to obtain a more accurate quality evaluation result. For anchor points with relatively simple track, i.e. with less data, a model with relatively simple structure, such as a random forest model, can be used to improve the efficiency of anchor point quality evaluation. A specific implementation may be found in the following description of fig. 4 and its explanation.

According to the technical scheme provided by the embodiment of the application, the sequence characteristics and the deviation correcting characteristics are respectively extracted from the two dimensions of the deviation correcting locating points and the difference between the deviation correcting locating points and the target locating points, so that a more accurate quality evaluation result can be obtained when the extracted characteristics are processed by utilizing the function model trained based on the machine learning thought.

Fig. 4 is a schematic flowchart of a step of determining a quality evaluation result according to a sequence feature and a deviation correction feature according to an embodiment of the present application, which is described in detail below.

In the embodiment of the application, an implementation scheme for selecting a proper network model to evaluate the quality of a target positioning point based on the track length of the target positioning point is provided, and specifically, the implementation scheme comprises the following steps:

401, determining the track length corresponding to the target positioning point according to the position information of the adjacent positioning point pair in the target positioning point.

In the embodiment of the application, adjacent positioning point pairs in the target positioning points refer to two positioning points which are continuously acquired from the time dimension, and the two positioning points can be specifically determined based on the time information of each target positioning point. For example, taking the foregoing target anchor point including M, N, P, Q as an example, if M, N, P, Q is acquired sequentially, M and N, N and P, P and Q are adjacent anchor point pairs.

Further, as an optional embodiment of the present application, according to the position information of each adjacent positioning point, the distance between two adjacent positioning points in the pair of positioning points can be calculated, and then the distances between the positioning points in the pair of adjacent positioning points are summed to obtain the track length corresponding to the target positioning point, that is, the track length is the sum of the distance MN between the positioning points M and N, the distance NP between the positioning points N and P, and the distance PQ between the positioning points P and Q, that is, the track length is (mn+np+pq), and of course, the track length corresponding to the target positioning point can also be calculated by other manners.

And 402, acquiring an associated network model corresponding to the track length, wherein the associated network model is obtained by training in advance based on the track length of the training sample and the training label.

In the embodiment of the application, under normal conditions, the track length can be used for describing the track to a certain extent, namely the complexity of the positioning points, specifically, the longer the track length is, the more complex the track can be considered, and corresponding to the positioning points with different complexity, the corresponding network model can be trained and obtained for processing. Specifically, in the model training process, the anchor point quality evaluation device trains in advance by using sample anchor point data with different complexity to obtain corresponding network models, and associates and stores the different network models with average track lengths which can be used for describing the complexity of the corresponding sample anchor point data in a preset database. In this way, in the process of quality evaluation by the positioning point quality evaluation device, the corresponding network model can be selected from a plurality of different network models as the associated network model by utilizing the size relation between the track length corresponding to the target positioning point and each average track length in the preset database, so as to improve the final quality evaluation effect.

403, setting the associated network model as a trained network model.

In the embodiment of the application, as can be seen from the description of the steps, the associated network model is a network model which is obtained through training and can perform effective quality evaluation on the positioning point data with similar complexity, the associated network model is set as a trained network model, and the quality evaluation on the target positioning point can be effectively realized through the network model.

According to the technical scheme provided by the embodiment of the application, the network models aiming at the complexity of different positioning points are obtained through training, and further, the associated network model corresponding to the track length of the target positioning point is obtained to be used as the network model for processing the target positioning point, so that the more accurate quality evaluation result of the target positioning point can be obtained.

As shown in fig. 5, fig. 5 is a schematic flow chart of steps for extracting sequence features according to an embodiment of the present application, which is described in detail below.

In the embodiment of the application, an implementation scheme for extracting sequence features from local and overall angles is provided, specifically, the implementation scheme comprises steps 501-503:

501, determining a first local feature of the deviation correcting locating point according to the difference of parameter information of adjacent locating point pairs in the deviation correcting locating point.

In the embodiment of the application, similar to the definition of the adjacent positioning points in the target positioning points, the adjacent positioning point pairs in the deviation correcting positioning points refer to two continuous positioning points in terms of time dimension. For example, taking the foregoing target anchor point including M, N, P, Q and the deviation correcting anchor point including M ', N ', P ', Q ', if M, N, P, Q is acquired sequentially, the adjacent anchor point pair in the deviation correcting anchor point includes M ' and N ', N ' and P ' and Q '.

Further, as a possible embodiment of the present application, the first local feature mainly includes a position sequence feature, a time sequence feature, and a velocity sequence feature. Specifically, based on the position information of the adjacent correction positioning points, the position sequence features of the correction positioning points can be extracted, for example, the straight line distances between every two adjacent correction positioning points are sequentially arranged, and the formed vector is the position sequence feature, wherein the sequence is also based on the time dimension information of each adjacent correction positioning point pair, and details are not repeated herein, and the specific implementation manner can refer to examples provided later. Similarly, based on the time information and the speed information of the adjacent deviation correcting locating points, the time sequence features and the speed sequence features of the deviation correcting locating points can be respectively extracted, for example, the time intervals between every two adjacent deviation correcting locating points are sequentially arranged, and the formed vector is the time sequence feature. And the difference values of the speeds of every two adjacent deviation correcting locating points are sequentially arranged, and the formed vector is the speed sequence characteristic. And sequentially splicing the position sequence features, the time sequence features and the speed sequence features to obtain the local features of the deviation correcting locating points.

For example, taking the case that the correction positioning points include M ', N', P ', Q', the linear distances between every two adjacent correction positioning points are sequentially arranged, so as to obtain the position sequence feature (d ₁ ，d ₂ ，d ₃ ) Wherein d ₁ ，d ₂ ，d ₃ The linear distance between the positioning points M ' and N ', the linear distance between the positioning points N ' and P ', and the linear distance between the positioning points P and Q ', respectively, can obtain time sequence characteristics (t ₁ ，t ₂ ，t ₃ ) And velocity sequence characteristics (v ₁ ，v ₂ ，v ₃ ) Wherein t is ₁ ，t ₂ ，t ₃ Respectively, the time interval between the locating points M ' and N ', the time interval between the locating points N ' and P ', the time interval between the locating points P and Q ', v ₁ ，v ₂ ，v ₃ The speed difference between the positioning points M ' and N ', the speed difference between the positioning points N ' and P ', and the speed difference between the positioning points P and Q ', respectively. The local characteristic of the finally obtained deviation rectifying locating point is (d) ₁ ，d ₂ ，d ₃ ，t ₁ ，t ₂ ，t ₃ ，v ₁ ，v ₂ ，v ₃ )。

Of course, besides the above-mentioned several features, the local features may also include other information obtained by extracting features by using a track segment formed between adjacent correction positioning points as a unit, for example, based on a ratio of a linear distance between adjacent correction positioning points to a time interval, an average speed between the two correction positioning points may be obtained, and similarly, based on a ratio of a speed difference between adjacent correction positioning points to a time interval, an average acceleration between the two correction positioning points may be obtained.

502, determining a first global feature of the deviation correcting locating point according to a data statistics result of the parameter information of the deviation correcting locating point.

In the embodiment of the application, the data statistics result of the parameter information refers to a result obtained by performing data statistics on the parameter information of all deviation correcting locating points, and specifically comprises at least one of average number, median and mode. For example, as an alternative embodiment of the present application, the global characteristics of the fix points include a travel length, a travel time, an average travel speed, and the like of the complete track determined based on the parameter information of all fix points.

503, setting the first local feature and the first global feature as the sequence feature of the deviation correcting positioning point.

In the embodiment of the application, since the first local feature and the first global feature exist in a vector manner, the extracted first global feature is spliced after the first local feature, so that a high-dimensional vector with a dimension equal to the sum of the dimension of the first local feature and the dimension of the first global feature can be obtained, and the high-dimensional vector is the sequence feature of the deviation correcting positioning point, wherein the first local feature is determined based on the difference between adjacent deviation correcting positioning points, describes the feature information of each section of local track, and the first global feature is determined based on the data statistics result of all deviation correcting positioning parameter information, and describes the feature information of the complete track. For example, as an alternative embodiment of the present application, in the process of determining the quality evaluation result by using the sequence feature, the sequence feature and the correction feature are input into the network model with a specific structure for processing. A specific implementation may be found in the following description of fig. 7 and its explanation.

Fig. 6 is a schematic flowchart of a step of determining a correction characteristic according to an embodiment of the present application, which is described in detail below.

In the embodiment of the application, the implementation scheme for extracting the correction features from the local and integral angles is also provided, and specifically includes steps 601-603:

601, setting a difference value between the parameter information of the target positioning point and the parameter information of the deviation correcting positioning point as a second local characteristic of the deviation correcting positioning point.

In the embodiment of the present application, as known from the foregoing description associated with step 202, since the correction is only performed on the position information in the positioning point, and is not performed on other parameter information, the difference between the parameter information of the target positioning point and the parameter information of the correction positioning point is mainly calculated according to the position information of the target positioning point and the position information of the correction positioning point, and the correction distance between each calculated correction positioning point and the corresponding target positioning point is calculated.

In the embodiment of the application, specifically, the difference value between the parameter information of the target positioning point and the parameter information of the deviation correcting positioning point is arranged according to the sequential relation of the deviation correcting positioning points, so that the second local characteristic of the deviation correcting positioning point can be obtained. The sequence relationship of the correction positioning points is defined based on the time dimension, that is, the sequence relationship is determined according to the time stamp of the correction positioning point, which is similar to the adjacent positioning point pair in the correction positioning points, and the embodiment of the application is not repeated here. And based on the sequence relation of the deviation correcting locating points, arranging the obtained deviation correcting values to obtain the second local characteristics of the deviation correcting locating points. Specifically, taking the example that the target positioning point includes M, N, P, Q and the deviation correcting positioning point includes M ', N', P ', Q', the second local feature is (MM ', NN', PP ', QQ').

And 602, determining a second global feature of the deviation correcting locating point according to the data statistics result of the difference value.

In the embodiment of the present application, similar to the global feature in the sequence feature determined in the step 502, the data statistics result of the difference value refers to a result obtained by performing data statistics on all parameter information of the difference value, and specifically includes at least one of an average number, a median number and a mode number. For example, as an alternative embodiment of the present application, the second global feature comprises the average deskew distance, i.e. the ratio of the sum of deskew distances to the number of anchor points, i.e. (MM '+nn' +pp '+qq')/4.

603, setting the second local feature and the second global feature as deviation rectifying features of the deviation rectifying locating points.

In the embodiment of the application, as the second local feature and the second global feature are both in a vector manner, the extracted second global feature is spliced after the second local feature, so that a high-dimensional vector with the dimension equal to the sum of the dimension of the second local feature and the dimension of the second global feature can be obtained, and the high-dimensional vector is the correction feature of the correction positioning point, wherein the second local feature is determined based on the difference between the correction positioning point and the target positioning point, the correction feature of each positioning point is independently described, and the second global feature is determined based on the difference between all the correction positioning points and the target positioning point, and the correction feature of all the positioning points is integrally described, so that the finally obtained correction feature comprises the correction feature extracted from two different dimensions of each positioning point and all the positioning points, and can be used for processing to obtain a more accurate quality evaluation result.

Fig. 7 is a flowchart of a step of determining a quality evaluation result by using a specific network model according to an embodiment of the present application, which is described in detail below.

In the embodiment of the application, a realization scheme for processing sequence characteristics and deviation correction characteristics by using a network model formed by a trained long-short-period memory neural network and a full-connection layer to obtain a quality evaluation result is provided, which specifically comprises steps 701-704:

701, splicing the local features in the sequence features and the local features in the deviation rectifying features to obtain the local features of the target positioning points.

In the embodiment of the present application, the local features in the sequence features are already shown in the foregoing fig. 5, and specifically, reference may be made to the step 501 and the explanation thereof, while the local features in the deviation correcting features are already shown in the foregoing fig. 6, and specifically, reference may be made to the step 602 and the explanation thereof, and the embodiment of the present application will not be repeated here.

Specifically, local features in the sequence features, such as position sequence features, time sequence features and speed sequence features, are spliced with local features in the deviation correcting features, such as deviation correcting distances corresponding to the deviation correcting positioning points, so that the local features of the target positioning points can be obtained.

And 702, splicing the global features in the sequence features and the global features in the deviation correcting features to obtain the global features of the target positioning points.

In the embodiment of the present application, the global features in the sequence features are already given in the foregoing fig. 5, and specifically, reference may be made to step 502 and the explanation thereof, while the global features in the deviation correcting features are already given in the foregoing fig. 6, and specifically, reference may be made to step 603 and the explanation thereof, which are not repeated herein.

In the embodiment of the application, specifically, global features in the sequence features, such as the running length, running time and average running speed of a complete track, global features in the deviation correcting features, such as the deviation correcting proportion of locating points, and the total length of the deviation correcting path are spliced to obtain the global features of the target locating points.

And 703, inputting the local characteristics of the target positioning points into the trained long-short-period memory neural network, and outputting the processed local characteristics.

In the embodiment of the application, based on the sequence of each positioning point and each section of track in the complete track, the local characteristic sequence corresponding to each positioning point and each section of track is input into a trained Long Short-Term Memory (LSTM) neural network, and the Long-Term Memory neural network can effectively learn the dependency relationship between the front local characteristic and the rear local characteristic, so that the processed local characteristic is obtained.

And 704, inputting the processed local features and the global features of the target positioning points into the trained full-connection layer, and outputting the quality evaluation result of the target positioning points.

In the embodiment of the application, the processed local features and the global features of the target positioning points are spliced and then input into the trained full-connection layer, and the quality evaluation result of the target positioning points can be finally obtained.

It should be noted that, in the embodiment of the present application, the model parameters of the long-short-period memory neural network mentioned in step 703 and the full-connection layer mentioned in step 704 are all obtained by training based on the idea of machine learning, and the training optimization process of the model parameters in the embodiment of the present application is not repeated here.

According to the technical scheme provided by the embodiment of the application, the local characteristics and the global characteristics processed by the long-term memory neural network are used as the input of the full-connection layer, so that the learning capability of the model on the track characteristics is effectively improved, and the effect of locating point quality evaluation is further improved.

Fig. 8 is a schematic flowchart of another step of determining a quality evaluation result according to a target positioning point and a deviation correcting positioning point according to an embodiment of the present application, which is described in detail below.

In the embodiment of the application, the method specifically comprises the steps 801 to 803:

801, dividing the target positioning points and the deviation correcting positioning points respectively to obtain a target positioning point set and a deviation correcting positioning point set corresponding to the target positioning point set.

In the embodiment of the application, dividing the target positioning points refers to dividing the target positioning points used for representing the complete track into a plurality of target positioning point sets describing a plurality of partial tracks according to a preset rule, wherein each target positioning point set comprises a plurality of target positioning points, and the preset rule can be that the track length reaches a certain length, or the running time reaches a certain duration, and the like. For example, taking the foregoing target anchor point containing M, N, P, Q as an example, dividing the target anchor point may refer to dividing the target anchor point into two sets of (M, N, P) and (P, Q), and describing the track of M to N to P and the track of P to Q, respectively. Of course, the foregoing is merely an example, and is not meant to limit the embodiments of the present application, that is, the actual target anchor point set may include more anchor points, rather than only two anchor points.

Correspondingly, after the target positioning points are divided to obtain a plurality of target positioning point sets, the deviation correcting positioning points can be divided into corresponding deviation correcting positioning point sets based on the same division rule. For example, taking the example of dividing the target anchor point M, N, P, Q into two sets of (M, N, P) and (P, Q), dividing the deviation correcting anchor points M, N ', P, Q' can obtain two sets of (M, N ', P) and (P, Q') of deviation correcting anchor points, which describe the track of M-N 'to P and the track of P-Q', respectively.

802, determining quality evaluation results of the target positioning point sets according to the positioning points in the target positioning point sets and the positioning points in the deviation correcting positioning point sets.

In the embodiment of the present application, the target anchor point in each target anchor point set is regarded as a new target anchor point, the correction anchor point in the corresponding correction anchor point set is regarded as a new correction anchor point, and then the quality evaluation result of each target anchor point set can be obtained by processing according to the embodiments provided in fig. 3 to 6, where the quality evaluation result of each target anchor point set is the quality evaluation result of the corresponding track, for example, the quality evaluation result of the target anchor point set (M, N, P) is the quality evaluation result of the track of M to N to P.

After the quality evaluation results of the tracks of each section are obtained, the quality evaluation results of similar tracks in the target positioning points acquired by different data acquisition devices can be respectively compared, track segments with optimal evaluation results are screened out, and then the track segments with optimal evaluation results are combined to obtain better track information.

803, determining a quality evaluation result of the target positioning point according to the quality evaluation result of each target positioning point set.

In the embodiment of the application, further, the quality evaluation result of the complete track can be comprehensively obtained by combining the quality evaluation results of all the sections of tracks.

According to the technical scheme provided by the embodiment of the application, the quality evaluation results of the track segments are obtained by carrying out quality evaluation on the track segments, so that the quality evaluation results of the complete track can be determined comprehensively according to the quality evaluation results of the track segments, and the optimal track segments in the data acquisition devices can be combined conveniently according to the quality evaluation results of the track segments acquired by the different data acquisition devices, so that more accurate track data can be provided.

In order to better implement the anchor point quality assessment method provided by the embodiment of the application, the embodiment of the application also provides an anchor point quality assessment device based on the anchor point quality assessment method. Fig. 9 is a schematic structural diagram of a positioning point quality evaluation device according to an embodiment of the present application, as shown in fig. 9. Specifically, the locating point quality evaluation device comprises:

An acquisition module 901, configured to acquire a target positioning point to be evaluated;

the deviation rectifying module 902 is configured to rectify the target positioning point according to preset standard track information to obtain a deviation rectifying positioning point;

and the evaluation module 903 is configured to determine a quality evaluation result of the target positioning point according to the target positioning point and the deviation correction positioning point.

In some embodiments of the application, the evaluation module includes:

the sequence feature extraction sub-module is used for extracting the sequence features of the deviation correcting locating points according to the parameter information of the deviation correcting locating points; the parameter information at least comprises one of position information, speed information and time information;

the deviation correcting feature extraction sub-module is used for determining deviation correcting features of the deviation correcting locating points according to the difference between the parameter information of the target locating points and the parameter information of the deviation correcting locating points;

and the model processing submodule is used for inputting the sequence characteristics and the deviation rectifying characteristics into a trained network model for processing and outputting a quality evaluation result of the target positioning point.

In some embodiments of the application, the evaluation module further comprises:

the track length determining submodule is used for determining the track length corresponding to the target locating point according to the position information of the adjacent locating point pair in the target locating point;

The model acquisition sub-module is used for acquiring an associated network model corresponding to the track length; the association network model is obtained by training based on the track length of a training sample and a training label in advance;

and the model setting sub-module is used for setting the associated network model as a trained network model.

In some embodiments of the present application, the sequence feature extraction submodule includes:

the first local feature extraction unit is used for determining the first local feature of the deviation correcting locating point according to the difference of the parameter information of the adjacent locating point pair in the deviation correcting locating point;

the first global feature extraction unit is used for determining global features of the deviation rectifying locating points according to data statistics results of parameter information of the deviation rectifying locating points;

the sequence feature setting unit is used for setting the first local feature and the second global feature as sequence features of the deviation correcting locating points; the data statistics result at least comprises one of an average number, a median and a mode;

in some embodiments of the present application, the correction feature extraction submodule includes:

the second local feature extraction unit is used for setting the difference value between the parameter information of the target positioning point and the parameter information of the deviation correcting positioning point as the second local feature of the deviation correcting positioning point;

The second global feature extraction unit is used for determining the second global feature of the deviation correcting locating point according to the data statistics result of the difference value; the data statistics result at least comprises one of an average number, a median and a mode;

and the deviation correcting feature setting unit is used for setting the second local feature and the second global feature as deviation correcting features of the deviation correcting locating points.

In some embodiments of the present application, the trained network model includes a trained long-term and short-term memory neural network and a trained full-connection layer, and the model processing submodule includes:

the first splicing unit is used for splicing the local features in the sequence features and the local features in the deviation correcting features to obtain the local features of the target positioning points;

the second splicing unit is used for splicing the global features in the sequence features and the global features in the deviation correcting features to obtain the global features of the target positioning points;

the first network processing unit is used for inputting the local characteristics of the target positioning points into the trained long-short-period memory neural network and outputting the processed local characteristics;

and the second network processing unit is used for inputting the processed local characteristics and the global characteristics of the target positioning points into the trained full-connection layer and outputting the quality evaluation result of the target positioning points.

In some embodiments of the application, the evaluation module includes:

the dividing sub-module is used for respectively dividing the target positioning points and the deviation correcting positioning points to obtain a target positioning point set and a deviation correcting positioning point set corresponding to the target positioning point set;

the segmentation evaluation sub-module is used for determining the quality evaluation result of each target positioning point set according to the positioning points in the target positioning point set and the positioning points in the deviation correction positioning point set;

and the comprehensive evaluation sub-module is used for determining the quality evaluation result of the target positioning point according to the quality evaluation result of each target positioning point set.

The embodiment of the application also provides a positioning point quality evaluation device, as shown in fig. 10, and fig. 10 is a schematic structural diagram of the positioning point quality evaluation device provided by the embodiment of the application.

The positioning point quality evaluation device comprises a memory, a processor and a positioning point quality evaluation program which is stored in the memory and can run on the processor, wherein the processor realizes the steps in the positioning point quality evaluation method provided by any embodiment of the application when executing the positioning point quality evaluation program.

Specifically, the present application relates to a method for manufacturing a semiconductor device. The setpoint quality assessment device may include one or more processors 1001 of a processing core, one or more memories 1002 of a storage medium, a power supply 1003, and an input unit 1004, among other components. It will be appreciated by those skilled in the art that the anchor point quality assessment device structure shown in fig. 10 does not constitute a limitation of the anchor point quality assessment device, and may include more or fewer components than shown, or may combine certain components, or may be a different arrangement of components. Wherein:

The processor 1001 is a control center of the anchor point quality assessment apparatus, connects respective parts of the entire anchor point quality assessment apparatus using various interfaces and lines, and performs various functions of the anchor point quality assessment apparatus and processes data by running or executing software programs and/or modules stored in the memory 1002 and calling data stored in the memory 1002, thereby performing overall monitoring of the anchor point quality assessment apparatus. Optionally, the processor 1001 may include one or more processing cores; preferably, the processor 1001 may integrate an application processor and a modem processor, wherein the application processor mainly processes an operating system, a user interface, an application program, and the like, and the modem processor mainly processes wireless communication. It will be appreciated that the modem processor described above may not be integrated into the processor 1001.

The memory 1002 may be used to store software programs and modules, and the processor 1001 executes various functional applications and data processing by executing the software programs and modules stored in the memory 1002. The memory 1002 may mainly include a storage program area that may store an operating system, application programs required for at least one function (such as a sound playing function, an image playing function, etc.), and a storage data area; the storage data area may store data created according to the use of the anchor point quality assessment device, etc. In addition, memory 1002 may include high-speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory 1002 may also include a memory controller to provide the processor 1001 with access to the memory 1002.

The setpoint quality assessment device further comprises a power supply 1003 for powering the various components, preferably the power supply 1003 is logically connected to the processor 1001 by a power management system, whereby the functions of managing charging, discharging, and power consumption are performed by the power management system. The power supply 1003 may also include one or more of any of a direct current or alternating current power supply, a recharging system, a power failure detection circuit, a power converter or inverter, a power status indicator, and the like.

The setpoint quality assessment device may further comprise an input unit 1004, which input unit 1004 may be used for receiving input numerical or character information and generating keyboard, mouse, joystick, optical or trackball signal inputs in connection with user settings and function control.

Although not shown, the anchor point quality evaluation device may further include a display unit or the like, which is not described herein. In particular, in this embodiment, the processor 1001 in the positioning point quality assessment device loads executable files corresponding to the processes of one or more application programs into the memory 1002 according to the following instructions, and the processor 1001 runs the application programs stored in the memory 1002, so as to implement the steps in the positioning point quality assessment method provided by any embodiment of the present application.

To this end, embodiments of the present application provide a computer-readable storage medium, which may include: read Only Memory (ROM), random access Memory (RAM, random Access Memory), magnetic or optical disk, and the like. The computer readable storage medium stores a setpoint quality assessment program which, when executed by a processor, implements the steps in the setpoint quality assessment method provided by any of the embodiments of the present application.

In the foregoing embodiments, the descriptions of the embodiments are focused on, and the portions of one embodiment that are not described in detail in the foregoing embodiments may be referred to in the foregoing detailed description of other embodiments, which are not described herein again.

In the implementation, each unit or structure may be implemented as an independent entity, or may be implemented as the same entity or several entities in any combination, and the implementation of each unit or structure may be referred to the foregoing method embodiments and will not be repeated herein.

The specific implementation of each operation above may be referred to the previous embodiments, and will not be described herein.

The above describes in detail a positioning point quality evaluation method provided by the embodiment of the present application, and specific examples are applied herein to illustrate the principle and implementation of the present application, and the description of the above embodiment is only used to help understand the method and core idea of the present application; meanwhile, as those skilled in the art will have variations in the specific embodiments and application scope in light of the ideas of the present application, the present description should not be construed as limiting the present application.

Claims

1. A method for evaluating quality of a positioning point, comprising:

acquiring a target positioning point to be evaluated;

2. The anchor point quality assessment method according to claim 1, wherein the determining the quality assessment result of the target anchor point according to the target anchor point and the deviation correcting anchor point comprises:

3. The anchor point quality assessment method according to claim 2, wherein before inputting the sequence feature and the correction feature into a trained network model for processing and outputting the quality assessment result of the target anchor point, the method comprises:

Determining the track length corresponding to the target locating point according to the position information of the adjacent locating point pair in the target locating point;

the associated network model is set to a trained network model.

4. The anchor point quality evaluation method according to claim 2, wherein the extracting the sequence feature of the anchor point according to the parameter information of the anchor point comprises:

5. The anchor point quality assessment method according to claim 2, wherein the determining the deviation correcting feature of the deviation correcting anchor point according to the difference between the parameter information of the target anchor point and the parameter information of the deviation correcting anchor point includes:

6. The anchor point quality assessment method of claim 2, wherein the trained network model comprises a trained long-short term memory neural network and a trained full connection layer;

the sequence features and the deviation rectifying features are input into a trained network model to be processed, and the quality evaluation result of the target positioning point is output, wherein the quality evaluation result comprises:

7. The anchor point quality assessment method according to any one of claims 1 to 6, wherein the determining the quality assessment result of the target anchor point according to the target anchor point and the deviation correcting anchor point includes:

determining quality evaluation results of the target positioning point sets according to the positioning points in the target positioning point sets and the positioning points in the deviation correcting positioning point sets;

8. A method for evaluating quality of a positioning point, comprising:

9. A setpoint quality assessment device, characterized in that it comprises a processor, a memory and a setpoint quality assessment program stored in the memory and executable on the processor, the processor executing the setpoint quality assessment program to implement the steps in the setpoint quality assessment method of any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon a setpoint quality assessment program, which is executed by a processor to implement the steps in the setpoint quality assessment method of any one of claims 1 to 7.