CN113469220A - Track identification method and related device and equipment - Google Patents
Track identification method and related device and equipment Download PDFInfo
- Publication number
- CN113469220A CN113469220A CN202110642505.8A CN202110642505A CN113469220A CN 113469220 A CN113469220 A CN 113469220A CN 202110642505 A CN202110642505 A CN 202110642505A CN 113469220 A CN113469220 A CN 113469220A
- Authority
- CN
- China
- Prior art keywords
- track
- points
- identified
- identification
- corner points
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
Abstract
The application discloses a track identification method, a related device and equipment, wherein the track identification method comprises the following steps: acquiring a plurality of track points of a track to be identified; determining a plurality of initial corner points of a track to be identified from a plurality of track points; obtaining a plurality of identification points with the same number as the reference corner points of the track template based on the initial corner points; and determining the type of the track to be identified by using the plurality of identification points and the reference corner points. According to the scheme, the accuracy and the efficiency of track identification can be improved.
Description
Technical Field
The present application relates to the field of trajectory recognition technologies, and in particular, to a trajectory recognition method, and a related apparatus and device.
Background
With the rapid development of science and technology, trajectory recognition is more and more widely applied in various industries. For example, handwriting input methods, games in which you draw me or guess me, pre-learning education in learning to learn characters, and the like all have certain dependence on track recognition technology.
At present, a common trajectory identification method usually utilizes statistics and machine learning principles to train a large number of samples to obtain a neural network or a classifier, and then uses the neural network or the classifier to perform type division on unknown samples.
However, in practice, if the types of the tracks to be identified are increased, a large number of samples need to be obtained again to train the neural network or the classifier, and this method is time-consuming and labor-consuming and is difficult to have practical value.
Disclosure of Invention
The application provides a track identification method, a related device and equipment, which aim to solve the problem of low track identification efficiency in the prior art.
The application provides a track identification method, which comprises the following steps: acquiring a plurality of track points of a track to be identified; determining a plurality of initial corner points of a track to be identified from a plurality of track points; obtaining a plurality of identification points with the same number as the reference corner points of the track template based on the initial corner points; and determining the type of the track to be identified by using the plurality of identification points and the reference corner points.
Wherein the step of obtaining a plurality of identification points of which the number is the same as that of the reference corner points of the track template based on the initial corner points comprises the following steps: in response to the number of the initial corner points being smaller than the number of the reference corner points, dividing a connecting line between every two adjacent initial corner points of the track to be identified equally by utilizing the multiple of the reference corner points relative to the number of the initial corner points to obtain at least one dividing point; and merging the equipartition points and the initial corner points into identification points.
Wherein, the step of obtaining a plurality of identification points with the same number as the reference corner points of the track template based on the initial corner points further comprises: in response to the fact that the multiple of the reference corner point relative to the number of the initial corner points is non-integral multiple, increasing the number of the first equipartition times for the connecting line of two adjacent initial corner points of the first remaining number of the to-be-recognized tracks in sequence according to the connecting line sequence of each initial corner point by utilizing the multiple and the remainder of the reference corner point relative to the number of the initial corner points so as to equipartite the connecting line of every two adjacent initial corner points of the to-be-recognized tracks to obtain at least one equipartition point; and merging the equipartition points and the initial corner points into identification points.
The method comprises the following steps of determining a plurality of initial corner points of a track to be identified from a plurality of track points, wherein the steps comprise: sequentially connecting the track points according to the track sequence of the track to be recognized to obtain a plurality of track vectors; and comparing the included angle between every two adjacent track vectors with a preset angle, and determining all track points between the two adjacent track vectors with the included angles larger than the preset angle as initial corner points.
Wherein the step of obtaining a plurality of identification points of which the number is the same as that of the reference corner points of the track template based on the initial corner points comprises the following steps: and increasing the angle of the preset angle in response to the number of the initial corner points being greater than the number of the reference corner points, so that the number of the initial corner points is less than the number of the reference corner points.
The method comprises the following steps of determining the type of a track to be identified by utilizing a plurality of identification points and reference corner points, wherein the steps comprise: and comparing the coordinates of the plurality of identification points with the coordinates of the reference corner points to determine the type of the track to be identified.
The method comprises the following steps of comparing coordinate information of a plurality of identification points with coordinate information of a reference corner point, and determining the type of a track to be identified, wherein the steps comprise: and responding to the situation that the number of the track templates is more than 1, and sequentially comparing each identification point with the corresponding reference corner points on the track template one by one according to the sequence of the tracks so as to determine the type of the track to be identified.
Wherein, responding to the number of the track templates larger than 1, sequentially comparing the identification points with the reference corner points of each track template respectively, and determining the type of the track to be identified, comprising the following steps: sequentially comparing the plurality of identification points with the reference corner point of each track template respectively to obtain the similarity between the track to be identified and each track template; and determining the type of the track template with the highest similarity as the type of the track to be identified.
The method comprises the following steps of sequentially comparing a plurality of identification points with the reference corner points of each track template respectively to obtain the similarity between a track to be identified and each track template, wherein the steps comprise: calculating the similarity R between the to-be-identified track of the to-be-identified track and each track template through the following formula:
T1*T2＝T11.x*T21.x+T11.y*T21.y...T1D.x*T2D.x+T1D.y*T2D.y (2)
wherein, T1 is the track to be identified, T2 is the track template, and D is the number of identification points or reference corner points.
Before the step of determining the type of the track to be identified by using the plurality of identification points and the reference corner points, the method comprises the following steps: and calculating a central point among the plurality of identification points, and mapping the coordinates of the plurality of identification points into a preset range by taking the central point as an origin to perform normalization processing on the plurality of identification points.
The present application further provides a trajectory recognition device, including: the acquisition module is used for acquiring a plurality of track points of the track to be identified; the corner module is used for determining a plurality of initial corner points of the track to be identified from a plurality of track points; the identification module is used for obtaining a plurality of identification points which are the same as the reference corner points of the track template in number based on the initial corner points; and the determining module is used for determining the type of the track to be identified by utilizing the plurality of identification points and the reference corner point.
The present application further provides an electronic device, which includes a memory and a processor coupled to each other, wherein the processor is configured to execute program instructions stored in the memory to implement any one of the above-mentioned trajectory recognition methods.
The present application also provides a computer readable storage medium having stored thereon program instructions which, when executed by a processor, implement the trajectory recognition method of any of the above.
According to the scheme, a plurality of initial corner points of the track to be identified are determined from a plurality of track points by acquiring the plurality of track points of the track to be identified. And then obtaining a plurality of identification points with the same number as the reference corner points of the track template based on the initial corner points, and finally determining the type of the track to be identified by utilizing the plurality of identification points and the reference corner points, wherein the type of the track to be identified can be directly determined through the track template, the accuracy of track identification is improved, meanwhile, the step of training a classifier based on a training sample to identify the track is avoided, the training consumption is reduced, and the speed and the efficiency of track identification are improved.
Drawings
FIG. 1 is a schematic flow chart diagram illustrating an embodiment of a trajectory recognition method of the present application;
FIG. 2 is a schematic flow chart diagram illustrating another embodiment of the trajectory recognition method of the present application;
FIG. 3 is a schematic diagram of one embodiment of a wire sharing;
FIG. 4 is a block diagram of an embodiment of the trajectory recognition device of the present application;
FIG. 5 is a block diagram of an embodiment of an electronic device of the present application;
FIG. 6 is a block diagram of an embodiment of a computer-readable storage medium of the present application.
Detailed Description
The following describes in detail the embodiments of the present application with reference to the drawings attached hereto.
In the following description, for purposes of explanation and not limitation, specific details are set forth such as particular system structures, interfaces, techniques, etc. in order to provide a thorough understanding of the present application.
The terms "system" and "network" are often used interchangeably herein. The term "and/or" herein is merely an association describing an associated object, and there may be three relationships, e.g., a and/or B, and: a exists alone, A and B exist simultaneously, and B exists alone. In addition, in this document, the character "/", generally, the former and latter related objects are in an "or" relationship. Further, herein, "more" than two or more than two.
Referring to fig. 1, fig. 1 is a schematic flowchart illustrating an embodiment of a trajectory recognition method according to the present application.
Specifically, the method may include the steps of:
step S11: and acquiring a plurality of track points of the track to be identified.
And acquiring a plurality of track points of the track to be identified. The trajectory to be recognized may include human body interaction trajectories such as a motion trajectory of a gesture, a motion trajectory of limbs, and the like, and a specific target object of the trajectory to be recognized is not limited herein. The method comprises the steps of identifying a human body interaction track to be identified, obtaining a track type, determining track meaning, and applying the track meaning to various track identification application scenes, such as: handwriting input methods, you draw me guess games, etc.
The specific mode of acquiring the track to be identified may be acquired through a sensor, a camera, a GPS positioning system, a beidou positioning system, or other devices, and the specific acquisition mode may be determined based on the type of the target object of the track to be identified, which is not limited herein.
Step S12: a plurality of initial corner points of the track to be identified are determined from the plurality of track points.
And determining a plurality of initial corner points on the track to be recognized from a plurality of track points of the track to be recognized. Wherein, the corner point refers to a bending point on the track to be identified. According to the embodiment, the track can be smoothed to a certain extent by performing subsequent identification and judgment through the initial corner points of the track to be identified, the characteristics of the track are reserved, the accuracy of subsequent track identification is improved, the calculated amount of track identification is reduced, and the efficiency of track identification is improved.
In a specific application scenario, the method for determining a plurality of initial corner points on a trajectory to be identified may include: connecting two adjacent track points on the track to be identified to obtain a plurality of connected track vectors, judging whether an included angle between every two adjacent track vectors exceeds a preset angle, and if so, determining a corner point between every two adjacent track vectors as an initial corner point. In another specific application scenario, the method for determining a plurality of initial corner points on a trajectory to be identified may further include: connecting two adjacent track points on the track to be recognized to obtain a plurality of connected track vectors, respectively judging whether the track difference between each two adjacent track vectors and the trend line of the track to be recognized exceeds a preset difference, and if so, confirming the corner point between the two adjacent track vectors as an initial corner point. In this embodiment, the method for determining a plurality of initial corner points on the trajectory to be identified is not limited herein.
Step S13: a number of identification points is derived based on the initial corner points, which is the same as the number of reference corner points of the trajectory template.
In this step, a plurality of identification points that are the same as the reference corner points of the track template can be obtained based on the initial corner points, so that the number of the identification points of the track to be identified is the same as the number of the reference corner points of the track template.
The manner of acquiring the plurality of identification points may include increasing or decreasing the number of initial corner points by adjusting the manner of determining the initial corner points in step S12, newly increasing identification points based on the features between the initial corner points or newly increasing identification points according to a preset rule, and the like, and the specific manner of acquiring the plurality of identification points is not limited herein.
Step S14: and determining the type of the track to be identified by using the plurality of identification points and the reference corner points.
And determining the type of the track to be recognized by using the plurality of recognition points of the track to be recognized and the plurality of reference corner points of the track template. Specifically, the similarity between a plurality of identification points of the to-be-identified track and a plurality of reference corner points of the track template may be determined to determine the type of the to-be-identified track.
In a specific application scenario, relative positions between a plurality of identification points of a to-be-identified track and relative positions between a plurality of reference corner points of a track template can be compared, and when the similarity between the relative positions between the plurality of identification points and the relative positions between the plurality of reference corner points meets a preset requirement, the type of the track template can be determined as the type of the to-be-identified track. The preset requirement may be set based on an actual situation, and is not limited herein.
In another specific application scenario, the coordinates of a plurality of identification points of the track to be identified and a plurality of reference corner points of the track template can also be obtained and compared, and when the similarity between the plurality of identification points and the plurality of reference corner points meets the preset requirement, the type of the track template can be determined as the type of the track to be identified. The method for determining the type of the track to be identified by using the multiple identification points and the reference corner point in this embodiment is not limited herein.
By the method, the track identification method of the embodiment firstly obtains a plurality of track points of the track to be identified, and then determines a plurality of initial corner points of the track to be identified from the plurality of track points. And then obtaining a plurality of identification points with the same number as the reference corner points of the track template based on the initial corner points, and finally determining the type of the track to be identified by utilizing the plurality of identification points and the reference corner points, wherein the type of the track to be identified can be directly determined through the track template, the accuracy of track identification is improved, meanwhile, the step of training a neural network or a classifier based on a training sample to identify the track is avoided, the training consumption is reduced, and the speed and the efficiency of track identification are improved.
Referring to fig. 2, fig. 2 is a schematic flowchart illustrating a trajectory recognition method according to another embodiment of the present application. Specifically, the method may include the steps of:
step S21: and acquiring a plurality of track points of the track to be identified.
The method comprises the steps of obtaining a plurality of track points of a track to be identified, and obtaining each track point and coordinates thereof by identifying key points of the track to be identified after obtaining the track to be identified of a target object in a specific application scene. The key point identification can be carried out through a trained deep neural network or a manual identification mode and the like.
In this example, the first step is performed by (T)1、T2、...、TN) And representing the track points and the coordinates of the track to be recognized according to the track sequence of the track to be recognized, wherein N is the number of the track points. The track to be recognized according to this embodiment is a track including at least 3 track points, and when the track points of the track to be recognized are 1 or 2, the comparison and determination in step S25 may be directly performed.
Step S22: and sequentially connecting all track points according to the track sequence of the track to be identified to obtain a plurality of track vectors, comparing the included angle between every two adjacent track vectors with a preset angle, and determining the track points between all two adjacent track vectors with the included angles larger than the preset angle as initial corner points.
And sequentially connecting the track points according to the track sequence of the track to be recognized to obtain a plurality of track vectors. And comparing the included angle between every two adjacent track vectors with a preset angle theta, and determining all track points between the two adjacent track vectors with the included angles larger than the preset angle theta as initial corner points. The preset angle θ may be 0 degree, 5 degrees, and the like, and may be specifically set based on the number requirement or actual situation of the identification points.
The method for comparing the included angle between the two track vectors with the preset angle theta comprises the steps of translating one of the two track vectors to enable the starting points of the two track vectors to coincide, judging whether the included angle of the two track vectors after the starting points coincide exceeds the preset angle, and if the included angle exceeds the preset angle, determining a track point between the two track vectors as an initial corner point.
In a specific application scenario, when two adjacent track vectors T are adjacent2-3And T3-4When the track vectors of (1.1) - (2.2) and (2.2) - (3.1) are respectively used, two adjacent track vectors T are present2-3And T3-4The included angle between the two adjacent track vectors is 90 degrees and is greater than the preset angle of 0 degree, so that the track point T between the two adjacent track vectors3And (2.2) determining the initial corner points.
After a plurality of initial corner points of the track to be identified are determined by comparing the included angle between every two adjacent track vectors with the preset angle theta, the initial corner points can be connected in sequence again to obtain a plurality of initial corner point vectors. Comparing the included angle between every two adjacent initial corner point vectors with the preset angle theta again, reserving all the initial corner points between the two adjacent initial corner point vectors with the included angles larger than the preset angle theta, and removing all the initial corner points between the two adjacent initial corner point vectors with the included angles smaller than the preset angle theta. The preset angle theta is the same as the preset angle theta of the track vector comparison, so that the initial corner points are repeatedly operated to remove the influence of burr protrusions on the track to be identified, and only the real initial corner points are reserved.
The coordinates of the initial corner points are determined based on the coordinates of the track points after the initial corner points are obtained.
Step S23: a number of identification points is derived based on the initial corner points, which is the same as the number of reference corner points of the trajectory template.
And obtaining a plurality of identification points with the same number as the reference corner points of the track template based on the initial corner points, specifically, adding points based on the number of the reference corner points of the track template on the basis of the initial corner points so that the added points and the initial corner points are added to be the same as the reference corner points of the track template.
In a specific application scene, after a plurality of initial corner points of a track to be recognized are obtained, the initial corner points are sequentially connected according to the sequence of the track to be recognized, whether the number of the preset corner points on the track to be recognized is smaller than that of the reference corner points or not is judged, if the number of the preset corner points on the track to be recognized is smaller than or equal to that of the reference corner points, the connecting lines between every two adjacent initial corner points are equally divided, and a plurality of recognition points of the track to be recognized are obtained.
If it is greater than the preset number, the rule of determining the initial corner points in step S22 is modified to reduce the number of initial corner points, for example, the preset angle θ is adjusted so that the number of initial corner points is less than or equal to the number of reference corner points. The number of the reference corner points may be 10, 20, 50, and the like, and the adjustment setting is specifically performed based on a preset number, which is not limited herein.
The acquisition mode of the reference corner points of the track template is the same as the acquisition mode of a plurality of identification points of the track to be identified, and the acquisition mode is obtained by acquiring a plurality of track points of the track template, determining a plurality of initial corner points from the track points, adjusting the number of the initial corner points to be the same as the preset number, and then carrying out normalization processing. The specific obtaining process may refer to the obtaining process of multiple identification points of the track to be identified, and is not described herein again. The track template identification method comprises the steps that the track template identification method comprises the steps of identifying tracks to be identified and track templates, wherein specific quantity of preset quantity is fixed and cannot be modified once the tracks to be identified and the track templates are identified, so that the quantity of identification points obtained by the tracks to be identified and the track templates is the same, and the quantity of the identification points is preset, so that comparison and identification are facilitated.
In a specific application scene, in response to the fact that the number of the initial corner points is smaller than the number of the reference corner points, dividing a connecting line between every two adjacent initial corner points of the track to be identified equally by utilizing the multiple of the reference corner points relative to the number of the initial corner points to obtain at least one dividing point; and merging the average points and the initial corner points into identification points, thereby obtaining the identification points with the same number as the reference corner points, namely the identification points with the preset number.
Referring to fig. 3, fig. 3 is a schematic diagram of an embodiment of the connection sharing.
The embodiment includes 4 initial corner points, wherein the 4 initial corner points are sequentially connected to obtain a vector group a, if the 4 initial corner points 1 need to be divided into 8 identification points, that is, the number of the reference corner points is 8, each vector in the vector group a is divided into 2 equal parts based on the multiple of the reference corner points relative to the number of the initial corner points 1, so as to obtain a vector group b, 4 newly added average division points 2 are obtained based on the starting points and the key points of the equally divided vectors, and the 4 average division points 2 in the vector group b and the original 4 initial corner points 1 are combined into 8 identification points.
In a specific application scene, in response to the fact that the multiple of the reference corner point relative to the number of the initial corner points is non-integral multiple, increasing the number of first equipartition times for connecting lines of two adjacent initial corner points of a plurality of previous parts of the track to be recognized according to the sequence of connecting lines of the initial corner points by utilizing the multiple and remainder of the reference corner point relative to the number of the initial corner points, and equipartiting the connecting lines of every two adjacent initial corner points of the track to be recognized to obtain at least one equipartition point; and merging the averaging points and the initial corner points into identification points, thereby obtaining the identification points with the same number as the reference corner points, namely the identification points with the preset number.
In a specific application scenario, assuming that the number of initial corner points is 4 and the reference corner point/preset number is 9, the multiple of the reference corner point relative to the initial corner points is 2, and the remainder is 1. Adding an equal division number of times for the connecting line of the two adjacent initial corner points in the sequence of the connecting lines of the initial corner points, wherein the number of times is 3, the number of times for the other connecting lines is 2, the number of times for the connecting lines after the connecting lines are equally divided is 3+2+2+2, the number of added equal division points is 2+1+1, and the original 4 initial corner points are combined to form 9 identification points. If the number of the initial corner points is 4, the reference corner point/preset number is 10, the multiple is 2, the remainder is 2, the number of times of adding one time of averaging to the connecting line of the first two adjacent initial corner points is 3, the number of times of averaging of the remaining other connecting lines is 2, the number of times of averaging of each connecting line after averaging is 3+3+2+2 in sequence, the number of times of adding 2+2+1+1, and the original 4 initial corner points are combined to form 10 identification points.
Step S24: and calculating a central point among the plurality of identification points, and mapping the coordinates of the plurality of identification points into a preset range by taking the central point as an origin to perform normalization processing on the plurality of identification points.
After the preset number of identification points of the track to be identified is determined, the coordinate ranges of the identification points of the tracks to be identified are different due to the dynamic uncertainty of the track to be identified, so that the normalization processing is performed on the plurality of identification points of the track to be identified, and the subsequent coordinate comparison is facilitated. And the plurality of reference corner points of each track template are subjected to the same normalization processing in advance and then subjected to coordinate comparison.
Specifically, the normalization processing method comprises the following steps: the method comprises the steps of firstly calculating the central points of all identification points of a track to be identified, taking the central points as original points, and mapping the coordinates of a plurality of identification points into a preset range, thereby carrying out normalization processing on the plurality of identification points. The calculation method of the central point comprises the following steps: adding the horizontal coordinates of all the identification points, and dividing the sum by the number of the identification points to obtain the horizontal coordinate of the central point; and adding the vertical coordinates of all the identification points, and dividing the sum by the number of the identification points to obtain the vertical coordinate of the central point, thereby obtaining the complete coordinate of the central point.
The preset range comprises ranges of [ -1, 1], [ -2, 2], [ -10, 10] and the like, and coordinates of all the identification points are mapped into the preset range in an equal proportion mode, so that subsequent comparison calculation is facilitated, and calculation amount is reduced. Wherein, when the preset range is [ -1, 1], the comparison calculation amount can be further reduced, and the comparison efficiency is improved.
Step S25: and comparing the coordinates of the plurality of identification points with the coordinates of the reference corner points to determine the type of the track to be identified.
And comparing the coordinates of the identification points after normalization processing with the coordinates of the reference corner points after normalization processing to determine the type of the track to be identified.
And in response to the fact that the number of the track templates is larger than 1, sequentially comparing the identification points with the corresponding reference corner points on the track templates one by one according to the sequence of the tracks so as to determine the type of the track to be identified. Specifically, the plurality of identification points can be sequentially compared with the reference corner point of each track template to obtain the similarity between the track to be identified and each track template; and determining the type of the track template with the highest similarity as the type of the track to be identified.
The similarity R between the track to be recognized of the track to be recognized and each track template is calculated through the following formula:
T1*T2＝T11.x*T21.x+T11.y*T21.y...T1D.x*T2D.x+T1D.y*T2D.y (2)
wherein, T1 is the track to be identified, T2 is the track template, and D is the preset number, i.e. the number of identification points or reference corner points. T11.xFor the abscissa of the first recognition point on the trajectory to be recognized, T21.xThe abscissa of the first reference corner point on the track template, T1D.xFor the abscissa of the D-th recognition point on the trajectory to be recognized, T2D.xFor the D-th reference corner point on the track templateCoordinates, other coordinate symbols and so on.
And sequentially calculating the similarity R between the track to be identified and each track template through the two formulas, and determining the type of the track template with the maximum similarity R with the track to be identified as the type of the track to be identified.
In a specific application scenario, when the types of the track templates are 3, namely, a letter a, a letter B and a letter C, respectively, the similarity R between the track template of the letter a and the track to be recognized is 2.1, the similarity R between the track template of the letter B and the track to be recognized is 1.4, and the similarity R between the track template of the letter C and the track to be recognized is 2.0, and the recognition result of the track to be recognized is the letter a.
The template tracks are a plurality of template tracks which are pre-stored in a template library before the tracks to be identified are identified. The template track may include multiple types, such as characters, numbers, patterns, letters, and the like, and the template track stored in the template library is the track type that can be identified. The template tracks stored in the template library are obtained by the same processing method as the tracks to be identified, namely the template tracks are obtained by the steps of obtaining track points, determining initial corner points, adjusting the number of identification points based on the preset number, performing normalization processing and the like in the steps of S21-S24 in sequence.
When the track types which are not available in the template library are met, the new track types are processed through the processing steps of the step S21-the step S24, then the newly added track templates comprising the reference track points in the preset number are obtained and stored in the template library, the recognition method of the embodiment can recognize the corresponding tracks only by adding the newly added track types subsequently without training a neural network or a classifier, and accordingly the speed and the efficiency of recognizing various tracks are increased.
By the method, the included angle between every two adjacent track vectors on the track to be identified is compared with the preset angle, all track points between the two adjacent track vectors with the included angles larger than the preset angle are determined as the initial corner points, and the type of the track to be identified is judged through the corner features, so that the track features are reserved, and the complexity of comparison calculation is simplified; then a plurality of identification points with the same number as the reference corner points of the track template are obtained based on the initial corner points, so as to unify the number specification of the identification points and the reference corner points, facilitate the subsequent comparison calculation, then calculating the central point among the plurality of identification points, taking the central point as an origin point, mapping the coordinates of the plurality of identification points to a preset range, the embodiment can improve the accuracy and efficiency of the track recognition, needs to train a neural network or a classifier, and needs to add a new track type subsequently, corresponding tracks can be identified, and therefore the speed and efficiency of identifying various tracks are further improved.
Referring to fig. 4, fig. 4 is a schematic diagram of a frame of an embodiment of a trajectory recognition device according to the present application. The trajectory recognition means 40 comprises an acquisition module 41, a corner module 42, a recognition module 43 and a determination module 44. An obtaining module 41, configured to obtain a plurality of track points of a track to be identified; the corner module 42 is configured to determine a plurality of initial corner points of the track to be identified from the plurality of track points; an identification module 43, configured to obtain a plurality of identification points, which are the same as the reference corner points of the track template, based on the initial corner points; a determining module 44, configured to determine the type of the track to be identified by using the plurality of identification points and the reference corner point.
The identification module 43 is further configured to, in response to that the number of the initial corner points is smaller than the number of the reference corner points, equally divide a connecting line between every two adjacent initial corner points of the trajectory to be identified by using a multiple of the reference corner points with respect to the number of the initial corner points, so as to obtain at least one average division point; and merging the equipartition points and the initial corner points into identification points.
The identification module 43 is further configured to, in response to that the multiple of the reference corner point relative to the number of the initial corner points is a non-integer multiple, sequentially increase, by using the multiple and the remainder of the reference corner point relative to the number of the initial corner points, a first division number of times for dividing a connecting line of two adjacent initial corner points of a first remaining number of tracks to be identified according to a connecting line sequence of each initial corner point, so as to divide a connecting line of every two adjacent initial corner points of the tracks to be identified equally to obtain at least one division point; and merging the equipartition points and the initial corner points into identification points.
The corner module 42 is further configured to sequentially connect the track points according to the track sequence of the track to be identified to obtain a plurality of track vectors; and comparing the included angle between every two adjacent track vectors with a preset angle, and determining all track points between the two adjacent track vectors with the included angles larger than the preset angle as initial corner points.
The identification module 43 is further configured to increase the angle of the preset angle in response to the number of initial corner points being greater than the number of reference corner points, so that the number of initial corner points is less than the number of reference corner points.
The determining module 44 is further configured to compare the coordinates of the multiple identification points with the coordinates of the reference corner points, and determine the type of the track to be identified.
The determining module 44 is further configured to, in response to that the number of the track templates is greater than 1, sequentially compare the identification points with the reference corner points of each track template one by one, so as to determine the type of the track to be identified.
The determining module 44 is further configured to sequentially compare the plurality of identification points with the reference corner point of each track template, so as to obtain similarity between the track to be identified and each track template; and determining the type of the track template with the highest similarity as the type of the track to be identified.
The determining module 44 is further configured to calculate similarity R between the to-be-identified track of the to-be-identified track and each track template by the following formula:
T1*T2＝T11.x*T21.x+T11.y*T21.y...T1D.x*T2D.x+T1D.y*T2D.y (2)
wherein, T1 is the track to be identified, T2 is the track template, and D is the number of identification points or reference corner points.
The determining module 44 is further configured to calculate a central point between the plurality of identification points, and map the coordinates of the plurality of identification points into a preset range with the central point as an origin, so as to perform normalization processing on the plurality of identification points.
According to the scheme, the accuracy and the efficiency of track identification can be improved.
Referring to fig. 5, fig. 5 is a schematic diagram of a frame of an embodiment of an electronic device according to the present application. The electronic device 50 comprises a memory 51 and a processor 52 coupled to each other, and the processor 52 is configured to execute program instructions stored in the memory 51 to implement the steps of any of the above-described embodiments of the trajectory recognition method. In one particular implementation scenario, electronic device 50 may include, but is not limited to: a microcomputer, a server, and the electronic device 50 may also include a mobile device such as a notebook computer, a tablet computer, and the like, which is not limited herein.
In particular, the processor 52 is configured to control itself and the memory 51 to implement the steps of any of the above-described embodiments of the speech detection method. Processor 52 may also be referred to as a CPU (Central Processing Unit). Processor 52 may be an integrated circuit chip having signal processing capabilities. The Processor 52 may also be a general purpose Processor, a Digital Signal Processor (DSP), 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. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. In addition, the processor 52 may be commonly implemented by an integrated circuit chip.
According to the scheme, the accuracy and the efficiency of track identification can be improved.
Referring to fig. 6, fig. 6 is a block diagram illustrating an embodiment of a computer-readable storage medium according to the present application. The computer readable storage medium 60 stores program instructions 601 capable of being executed by a processor, the program instructions 601 being for implementing the steps of any of the above-described embodiments of the trajectory recognition method.
According to the scheme, the accuracy and the efficiency of track identification can be improved.
In the several embodiments provided in the present application, it should be understood that the disclosed method and apparatus may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, a division of a module or a unit is merely one type of logical division, and an actual implementation may have another division, for example, a unit or a component may be combined or integrated with another system, or some features may be omitted, or not implemented. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of devices or units through some interfaces, and may be in an electrical, mechanical or other form.
Units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on network elements. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.
In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be substantially implemented or contributed to by the prior art, or all or part of the technical solution may be embodied in a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, a network device, or the like) or a processor (processor) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
Claims (13)
1. A track recognition method is characterized by comprising the following steps:
acquiring a plurality of track points of a track to be identified;
determining a plurality of initial corner points of the track to be identified from the plurality of track points;
obtaining a plurality of identification points with the same number as the reference corner points of the track template based on the initial corner points;
and determining the type of the track to be identified by using the plurality of identification points and the reference corner points.
2. The trajectory recognition method according to claim 1, wherein the step of deriving a number of recognition points based on the initial corner points, which is equal to a number of reference corner points of the trajectory template, comprises:
in response to the fact that the number of the initial corner points is smaller than the number of the reference corner points, dividing a connecting line between every two adjacent initial corner points in the track to be identified equally by utilizing the multiple of the reference corner points relative to the number of the initial corner points to obtain at least one dividing point;
merging the averaging points and the initial corner points into the identification points.
3. The trajectory recognition method according to claim 1, wherein the step of obtaining a number of recognition points based on the initial corner points, which is equal to a number of reference corner points of the trajectory template, further comprises:
in response to that the multiple of the reference corner point relative to the initial corner point number is non-integral multiple, sequentially increasing the number of primary averaging points to the connecting lines of two adjacent initial corner points of the first remaining number of the to-be-identified track according to the connecting line sequence of each initial corner point by utilizing the multiple and remainder of the reference corner point relative to the initial corner point number so as to equally divide the connecting lines of every two adjacent initial corner points in the to-be-identified track to obtain at least one averaging point;
merging the averaging points and the initial corner points into the identification points.
4. The trajectory recognition method of claim 1, wherein the step of determining a plurality of initial corner points of the trajectory to be recognized from the plurality of trajectory points comprises:
sequentially connecting the track points according to the track sequence of the track to be recognized to obtain a plurality of track vectors;
and comparing the included angle between every two adjacent track vectors with a preset angle, and determining the track points between the two adjacent track vectors with the included angles larger than the preset angle as the initial corner points.
5. The trajectory recognition method according to claim 4, wherein the step of deriving a number of recognition points based on the initial corner points, which is equal to the number of reference corner points of the trajectory template, comprises:
and increasing the angle of the preset angle in response to the number of the initial corner points being greater than the number of the reference corner points, so that the number of the initial corner points is less than the number of the reference corner points.
6. The trajectory identification method according to claim 1, wherein the step of determining the type of the trajectory to be identified by using the plurality of identification points and the reference corner point comprises:
and comparing the coordinates of the identification points with the coordinates of the reference corner points to determine the type of the track to be identified.
7. The track identification method according to claim 1 or 6, wherein the step of comparing the coordinate information of the plurality of identification points with the coordinate information of the reference corner point to determine the type of the track to be identified comprises:
and responding to the fact that the number of the track templates is larger than 1, and sequentially comparing the identification points with the corresponding reference corner points on the track templates one by one according to the sequence of the tracks so as to determine the type of the track to be identified.
8. The trajectory identification method according to claim 7, wherein the step of sequentially comparing the identification points with the reference corner points of each trajectory template in response to the number of the trajectory templates being greater than 1, and determining the type of the trajectory to be identified comprises:
sequentially comparing the plurality of identification points with the reference corner point of each track template respectively to obtain the similarity between the track to be identified and each track template;
and determining the type of the track template with the highest similarity as the type of the track to be identified.
9. The trajectory identification method according to claim 7, wherein the step of sequentially comparing the plurality of identification points with the reference corner point of each trajectory template to obtain the similarity between the trajectory to be identified and each trajectory template comprises:
calculating the similarity R between the to-be-identified track of the to-be-identified track and each track template through the following formula:
T1*T2＝T11.x*T21.x+T11.y*T21.y...T1D.x*T2D.x+T1D.y*T2D.y (2)
wherein, T1 is the track to be identified, T2 is the track template, and D is the number of identification points or reference corner points.
10. The trajectory recognition method of claim 1, wherein the step of determining the type of the trajectory to be recognized using the plurality of recognition points and the reference corner point is preceded by the step of:
and calculating a central point among the plurality of identification points, and mapping the coordinates of the plurality of identification points into a preset range by taking the central point as an origin point so as to perform normalization processing on the plurality of identification points.
11. A trajectory recognition device, characterized in that the trajectory recognition device comprises:
the acquisition module is used for acquiring a plurality of track points of the track to be identified;
the corner module is used for determining a plurality of initial corner points of the track to be identified from the plurality of track points;
the identification module is used for obtaining a plurality of identification points which are the same as the reference corner points of the track template in number based on the initial corner points;
and the determining module is used for determining the type of the track to be identified by utilizing the plurality of identification points and the reference corner point.
12. An electronic device comprising a memory and a processor coupled to each other, the processor being configured to execute program instructions stored in the memory to implement the trajectory recognition method according to any one of claims 1 to 10.
13. A computer-readable storage medium having stored thereon program instructions, which when executed by a processor, implement the trajectory recognition method of any one of claims 1 to 10.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110642505.8A CN113469220A (en) | 2021-06-09 | 2021-06-09 | Track identification method and related device and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110642505.8A CN113469220A (en) | 2021-06-09 | 2021-06-09 | Track identification method and related device and equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113469220A true CN113469220A (en) | 2021-10-01 |
Family
ID=77869443
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110642505.8A Pending CN113469220A (en) | 2021-06-09 | 2021-06-09 | Track identification method and related device and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113469220A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114879656A (en) * | 2022-03-30 | 2022-08-09 | 深圳拓邦股份有限公司 | Method and device for adjusting operation direction of intelligent mower, electronic equipment and storage medium |
| CN117114971A (en) * | 2023-08-01 | 2023-11-24 | 北京城建设计发展集团股份有限公司 | Pixel map-to-vector map conversion method and system |
-
2021
- 2021-06-09 CN CN202110642505.8A patent/CN113469220A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114879656A (en) * | 2022-03-30 | 2022-08-09 | 深圳拓邦股份有限公司 | Method and device for adjusting operation direction of intelligent mower, electronic equipment and storage medium |
| CN117114971A (en) * | 2023-08-01 | 2023-11-24 | 北京城建设计发展集团股份有限公司 | Pixel map-to-vector map conversion method and system |
| CN117114971B (en) * | 2023-08-01 | 2024-03-08 | 北京城建设计发展集团股份有限公司 | Pixel map-to-vector map conversion method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110232311B (en) | Method and device for segmenting hand image and computer equipment | |
| WO2021026805A1 (en) | Adversarial example detection method and apparatus, computing device, and computer storage medium | |
| US10445602B2 (en) | Apparatus and method for recognizing traffic signs | |
| US9984280B2 (en) | Object recognition system using left and right images and method | |
| US10565713B2 (en) | Image processing apparatus and method | |
| US20200125876A1 (en) | Method and Device for License Plate Positioning | |
| US8090151B2 (en) | Face feature point detection apparatus and method of the same | |
| CN106446862A (en) | Face detection method and system | |
| JP2016006679A (en) | Face recognition method, apparatus, and computer-readable recording medium for implementing method | |
| CN113469220A (en) | Track identification method and related device and equipment | |
| US10528844B2 (en) | Method and apparatus for distance measurement | |
| US20190228209A1 (en) | Lip movement capturing method and device, and storage medium | |
| CN110852311A (en) | Three-dimensional human hand key point positioning method and device | |
| CN112733767B (en) | Human body key point detection method and device, storage medium and terminal equipment | |
| CN112085701A (en) | Face ambiguity detection method and device, terminal equipment and storage medium | |
| US20190012561A1 (en) | Fast curve matching for tattoo recognition and identification | |
| Ismail et al. | Efficient enhancement and matching for iris recognition using SURF | |
| CN111582027B (en) | Identity authentication method, identity authentication device, computer equipment and storage medium | |
| WO2017206144A1 (en) | Estimation of human orientation in images using depth information | |
| CN110232381B (en) | License plate segmentation method, license plate segmentation device, computer equipment and computer readable storage medium | |
| CN112364807B (en) | Image recognition method, device, terminal equipment and computer readable storage medium | |
| US20030044067A1 (en) | Apparatus and methods for pattern recognition based on transform aggregation | |
| US20170220898A1 (en) | Select type of test image based on similarity score to database image | |
| CN108288023B (en) | Face recognition method and device | |
| CN110287786B (en) | Vehicle information identification method and device based on artificial intelligence anti-interference |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |