# US 2023/0053952 A1 · Method and apparatus for evaluating motion state of traffic tool, device, and medium

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2023/0053952 A1
- Document ID: US20230053952A1
- Status: Published application
- Kind: A1
- Application number: 17/978,113
- Filed: 2022-10-31
- Published: 2023-02-23
- Earliest claimed: 2021-01-25
- Assignees: Tencent Technology (Shenzhen) Company Limited
- Applicants: Tencent Technology (Shenzhen) Company Limited
- Inventors: Jinglan SU
- Web page: https://patentgrep.com/patent/US20230053952A1

## CPC

- G06T 7/246: using feature-based methods, e.g. the tracking of corners or segments
- G06T 7/248: {involving reference images or patches}
- G01C 21/32: Structuring or formatting of map data
- G06T 5/50: using two or more images, e.g. averaging or subtraction
- G06T 5/70: Denoising; Smoothing
- G06T 7/251: {involving models}
- G01C 11/06: by comparison of two or more pictures of the same area
- G06T 2207/20021: Dividing image into blocks, subimages or windows
- G06T 2207/20212: Image combination
- G06T 2207/30252: Vehicle exterior; Vicinity of vehicle
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G01C: MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY (measuring liquid level G01F; radio navigation, determining distance or velocity by use of propagation effects, e.g. Doppler effects, propagation time, of radio waves, analogous arrangements using other waves G01S)

## Abstract

This application provides a method for evaluating a motion state of a vehicle performed by a computer device. The method includes: obtaining target image data captured by cameras on the vehicle, and combining every two neighboring image frames in the target image data into N image frame groups; obtaining matching feature points between two image frames included in each of the N image frame groups; constructing target mesh plane figures respectively corresponding to the N image frame groups according to the matching feature points in each image frame group; and determining the motion state of the vehicle according to the target mesh plane figures respectively corresponding to the N image frame groups. By adopting the embodiments of this application, the evaluation accuracy of the motion state of the vehicle may be improved.

## Claims

### Claim 1

A method for evaluating a motion state of a vehicle, performed by a computer device and comprising:
obtaining target image data captured by cameras on the vehicle, and combining every two neighboring image frames in the target image data into N image frame groups, N being a positive integer;
obtaining matching feature points between two image frames comprised in each image frame group in the N image frame groups;
constructing target mesh plane figures respectively corresponding to the N image frame groups according to the matching feature points in each image frame group; and
determining the motion state of the vehicle according to the target mesh plane figures respectively corresponding to the N image frame groups.

### Claim 2 (depends on 1)

The method according to claim 1, wherein the obtaining captured by cameras on the vehicle, and combining every two neighboring image frames in the target image data into N image frame groups comprises:
obtaining original image data captured by the cameras on the vehicle, and performing denoising processing on the original image data to obtain denoised image data;
performing distortion removing processing on the denoised image data according to a device parameter corresponding to the cameras, to obtain the target image data corresponding to the vehicle; and
dividing the target image data into N+1 image frames, and combining every two neighboring image frames in the N+1 image frames to obtain the N image frame groups.

### Claim 3 (depends on 1)

The method according to claim 1, wherein the obtaining matching feature points between two image frames comprised in each image frame group in the N image frame groups comprises:
obtaining an image frame group Gi in the N image frame groups, wherein the image frame group Gi comprises an image frame Ti and an image frame Ti+1, and i is a positive integer less than or equal to N;
obtaining a first feature point set in the image frame Ti and a second feature point set in the image frame Ti+1; and
performing feature point matching on the first feature point set and the second feature point set to obtain matching feature points between the image frame Ti and the image frame Ti+1.

### Claim 4 (depends on 1)

The method according to claim 1, wherein the constructing target mesh plane figures respectively corresponding to the N image frame groups according to the matching feature points in each image frame group comprises:
obtaining q matching feature points between an image frame Ti and an image frame Ti+1 comprised in an image frame group Gi, and determining matching coordinate information respectively corresponding to the q matching feature points according to image coordinate information of the q matching feature points in the image frame Ti and the image frame Ti+1 respectively, wherein the image frame group Gi belongs to the N image frame groups, i is a positive integer less than or equal to N, and q is a positive integer;
obtaining an initial polygon linked list associated with the q matching feature points according to the matching coordinate information respectively corresponding to the q matching feature points;
obtaining, for a matching feature point cr in the q matching feature points, an associated polygon matching the matching feature point cr from the initial polygon linked list, wherein a circumscribed circle of the associated polygon comprises the matching feature point cr, and r is a positive integer less than or equal to q;
deleting a common side of the associated polygon, and connecting the matching feature point cr to vertexes of the associated polygon to generate a candidate polygon corresponding to the matching feature point cr;
converting the candidate polygon into a target polygon corresponding to the matching feature point cr, and adding the target polygon to the initial polygon linked list, wherein a circumscribed circle of the target polygon does not comprise any matching feature point; and
determining the initial polygon linked list as a target polygon linked list when target polygons respectively corresponding to the q matching feature points are all added to the initial polygon linked list, and generating a target mesh plane figure corresponding to the image frame group Gi according to the target polygon linked list.

### Claim 5 (depends on 1)

The method according to claim 1, wherein the determining the motion state of the vehicle according to the target mesh plane figures respectively corresponding to the N image frame groups comprises:
obtaining a target mesh plane figure Hi corresponding to an image frame group Gi in the N image frame groups, wherein the target mesh plane figure Hi comprises M connected but non-overlapping polygons, a circumscribed circle of each of the polygons does not comprise any matching feature point, M is a positive integer, and i is a positive integer less than or equal to N;
obtaining average coordinate change values respectively corresponding to the M polygons, sorting the average coordinate change values respectively corresponding to the M polygons, and determining the sorted average coordinate change values as a first coordinate sequence of the target mesh plane figure Hi;
determining a statistical value associated with the target mesh plane figure Hi according to the first coordinate sequence; and
determining the motion state of the vehicle as a static state when statistical values respectively associated with N target mesh plane figures are all less than a first threshold.

### Claim 6 (depends on 1)

The method according to claim 1, wherein the determining the motion state of the vehicle according to the target mesh plane figures comprises:
obtaining a target mesh plane figure Hi corresponding to an image frame group Gi in the N image frame groups, wherein the image frame group Gi comprises an image frame Ti and an image frame Ti+1, the target mesh plane figure Hi comprises M connected but non-overlapping polygons, a circumscribed circle of each of the polygons does not comprise any matching feature point, M is a positive integer, and i is a positive integer less than or equal to N;
obtaining a first direction of each side of the M polygons in the image frame Ti, and obtaining a second direction of each side of the M polygons in the image frame Ti+1;
determining a direction change value corresponding to each side of the M polygons according to the first direction and the second direction;
generating a direction change sequence associated with the M polygons according to direction change values respectively corresponding to the M polygons, and determining a direction average value and a direction median associated with the target mesh plane figure Hi according to the direction change sequence; and
determining the motion state of the vehicle as a linear motion state when direction average values respectively associated with N target mesh plane figures and direction medians respectively associated with the N target mesh plane figures are all less than a second threshold.

### Claim 7 (depends on 1)

The method according to claim 1, further comprising:
when the motion state of the vehicle is in a static state, determining positioning information of the vehicle according to a positioning and navigation system mounted on the vehicle; and
when the motion state of the vehicle is in a linear motion state, determining a motion direction of the vehicle according to the linear motion state, and determining the positioning information of the vehicle according to the motion direction and the positioning and navigation system.

### Claim 8

A computer device, comprising a memory and a processor,
the processor being connected to the memory, the memory being configured to store a computer program, and the processor being configured to execute the computer program and cause the computer device to perform a method for evaluating a motion state of a vehicle including:
obtaining target image data captured by cameras on the vehicle, and combining every two neighboring image frames in the target image data into N image frame groups, N being a positive integer;
obtaining matching feature points between two image frames comprised in each image frame group in the N image frame groups;
constructing target mesh plane figures respectively corresponding to the N image frame groups according to the matching feature points in each image frame group; and
determining the motion state of the vehicle according to the target mesh plane figures respectively corresponding to the N image frame groups.

### Claim 9 (depends on 8)

The computer device according to claim 8, wherein the obtaining captured by cameras on the vehicle, and combining every two neighboring image frames in the target image data into N image frame groups comprises:
obtaining original image data captured by the cameras on the vehicle, and performing denoising processing on the original image data to obtain denoised image data;
performing distortion removing processing on the denoised image data according to a device parameter corresponding to the cameras, to obtain the target image data corresponding to the vehicle; and
dividing the target image data into N+1 image frames, and combining every two neighboring image frames in the N+1 image frames to obtain the N image frame groups.

### Claim 10 (depends on 8)

The computer device according to claim 8, wherein the obtaining matching feature points between two image frames comprised in each image frame group in the N image frame groups comprises:
obtaining an image frame group Gi in the N image frame groups, wherein the image frame group Gi comprises an image frame Ti and an image frame Ti+1, and i is a positive integer less than or equal to N;
obtaining a first feature point set in the image frame Ti and a second feature point set in the image frame Ti+1; and
performing feature point matching on the first feature point set and the second feature point set to obtain matching feature points between the image frame Ti and the image frame Ti+1.

### Claim 11 (depends on 8)

The computer device according to claim 8, wherein the constructing target mesh plane figures respectively corresponding to the N image frame groups according to the matching feature points in each image frame group comprises:
obtaining q matching feature points between an image frame Ti and an image frame Ti+1 comprised in an image frame group Gi, and determining matching coordinate information respectively corresponding to the q matching feature points according to image coordinate information of the q matching feature points in the image frame Ti and the image frame Ti+1 respectively, wherein the image frame group Gi belongs to the N image frame groups, i is a positive integer less than or equal to N, and q is a positive integer;
obtaining an initial polygon linked list associated with the q matching feature points according to the matching coordinate information respectively corresponding to the q matching feature points;
obtaining, for a matching feature point cr in the q matching feature points, an associated polygon matching the matching feature point cr from the initial polygon linked list, wherein a circumscribed circle of the associated polygon comprises the matching feature point cr, and r is a positive integer less than or equal to q;
deleting a common side of the associated polygon, and connecting the matching feature point cr to vertexes of the associated polygon to generate a candidate polygon corresponding to the matching feature point cr;
converting the candidate polygon into a target polygon corresponding to the matching feature point cr, and adding the target polygon to the initial polygon linked list, wherein a circumscribed circle of the target polygon does not comprise any matching feature point; and
determining the initial polygon linked list as a target polygon linked list when target polygons respectively corresponding to the q matching feature points are all added to the initial polygon linked list, and generating a target mesh plane figure corresponding to the image frame group Gi according to the target polygon linked list.

### Claim 12 (depends on 8)

The computer device according to claim 8, wherein the determining the motion state of the vehicle according to the target mesh plane figures respectively corresponding to the N image frame groups comprises:
obtaining a target mesh plane figure Hi corresponding to an image frame group Gi in the N image frame groups, wherein the target mesh plane figure Hi comprises M connected but non-overlapping polygons, a circumscribed circle of each of the polygons does not comprise any matching feature point, M is a positive integer, and i is a positive integer less than or equal to N;
obtaining average coordinate change values respectively corresponding to the M polygons, sorting the average coordinate change values respectively corresponding to the M polygons, and determining the sorted average coordinate change values as a first coordinate sequence of the target mesh plane figure Hi;
determining a statistical value associated with the target mesh plane figure Hi according to the first coordinate sequence; and
determining the motion state of the vehicle as a static state when statistical values respectively associated with N target mesh plane figures are all less than a first threshold.

### Claim 13 (depends on 8)

The computer device according to claim 8, wherein the determining the motion state of the vehicle according to the target mesh plane figures comprises:
obtaining a target mesh plane figure Hi corresponding to an image frame group Gi in the N image frame groups, wherein the image frame group Gi comprises an image frame Ti and an image frame Ti+1, the target mesh plane figure Hi comprises M connected but non-overlapping polygons, a circumscribed circle of each of the polygons does not comprise any matching feature point, M is a positive integer, and i is a positive integer less than or equal to N;
obtaining a first direction of each side of the M polygons in the image frame Ti, and obtaining a second direction of each side of the M polygons in the image frame Ti+1;
determining a direction change value corresponding to each side of the M polygons according to the first direction and the second direction;
generating a direction change sequence associated with the M polygons according to direction change values respectively corresponding to the M polygons, and determining a direction average value and a direction median associated with the target mesh plane figure Hi according to the direction change sequence; and
determining the motion state of the vehicle as a linear motion state when direction average values respectively associated with N target mesh plane figures and direction medians respectively associated with the N target mesh plane figures are all less than a second threshold.

### Claim 14 (depends on 8)

The computer device according to claim 8, wherein the method further comprises:
when the motion state of the vehicle is in a static state, determining positioning information of the vehicle according to a positioning and navigation system mounted on the vehicle; and
when the motion state of the vehicle is in a linear motion state, determining a motion direction of the vehicle according to the linear motion state, and determining the positioning information of the vehicle according to the motion direction and the positioning and navigation system.

### Claim 15

A non-transitory computer-readable storage medium, storing a computer program, the computer program, adapted to be loaded and executed by a processor of a computer device, and causing the computer device to perform a method for evaluating a motion state of a vehicle including:
obtaining target image data captured by cameras on the vehicle, and combining every two neighboring image frames in the target image data into N image frame groups, N being a positive integer;
obtaining matching feature points between two image frames comprised in each image frame group in the N image frame groups;
constructing target mesh plane figures respectively corresponding to the N image frame groups according to the matching feature points in each image frame group; and
determining the motion state of the vehicle according to the target mesh plane figures respectively corresponding to the N image frame groups.

### Claim 16 (depends on 15)

The non-transitory computer-readable storage medium according to claim 15, wherein the obtaining captured by cameras on the vehicle, and combining every two neighboring image frames in the target image data into N image frame groups comprises:
obtaining original image data captured by the cameras on the vehicle, and performing denoising processing on the original image data to obtain denoised image data;
performing distortion removing processing on the denoised image data according to a device parameter corresponding to the cameras, to obtain the target image data corresponding to the vehicle; and
dividing the target image data into N+1 image frames, and combining every two neighboring image frames in the N+1 image frames to obtain the N image frame groups.

### Claim 17 (depends on 15)

The non-transitory computer-readable storage medium according to claim 15, wherein the obtaining matching feature points between two image frames comprised in each image frame group in the N image frame groups comprises:
obtaining an image frame group Gi in the N image frame groups, wherein the image frame group Gi comprises an image frame Ti and an image frame Ti+1, and i is a positive integer less than or equal to N;
obtaining a first feature point set in the image frame Ti and a second feature point set in the image frame Ti+1; and
performing feature point matching on the first feature point set and the second feature point set to obtain matching feature points between the image frame Ti and the image frame Ti+1.

### Claim 18 (depends on 15)

The non-transitory computer-readable storage medium according to claim 15, wherein the determining the motion state of the vehicle according to the target mesh plane figures respectively corresponding to the N image frame groups comprises:
obtaining a target mesh plane figure Hi corresponding to an image frame group Gi in the N image frame groups, wherein the target mesh plane figure Hi comprises M connected but non-overlapping polygons, a circumscribed circle of each of the polygons does not comprise any matching feature point, M is a positive integer, and i is a positive integer less than or equal to N;
obtaining average coordinate change values respectively corresponding to the M polygons, sorting the average coordinate change values respectively corresponding to the M polygons, and determining the sorted average coordinate change values as a first coordinate sequence of the target mesh plane figure Hi;
determining a statistical value associated with the target mesh plane figure Hi according to the first coordinate sequence; and
determining the motion state of the vehicle as a static state when statistical values respectively associated with N target mesh plane figures are all less than a first threshold.

### Claim 19 (depends on 15)

The non-transitory computer-readable storage medium according to claim 15, wherein the determining the motion state of the vehicle according to the target mesh plane figures comprises:
obtaining a target mesh plane figure Hi corresponding to an image frame group Gi in the N image frame groups, wherein the image frame group Gi comprises an image frame Ti and an image frame Ti+1, the target mesh plane figure Hi comprises M connected but non-overlapping polygons, a circumscribed circle of each of the polygons does not comprise any matching feature point, M is a positive integer, and i is a positive integer less than or equal to N;
obtaining a first direction of each side of the M polygons in the image frame Ti, and obtaining a second direction of each side of the M polygons in the image frame Ti+1;
determining a direction change value corresponding to each side of the M polygons according to the first direction and the second direction;
generating a direction change sequence associated with the M polygons according to direction change values respectively corresponding to the M polygons, and determining a direction average value and a direction median associated with the target mesh plane figure Hi according to the direction change sequence; and
determining the motion state of the vehicle as a linear motion state when direction average values respectively associated with N target mesh plane figures and direction medians respectively associated with the N target mesh plane figures are all less than a second threshold.

### Claim 20 (depends on 15)

The non-transitory computer-readable storage medium according to claim 15, wherein the method further comprises:
when the motion state of the vehicle is in a static state, determining positioning information of the vehicle according to a positioning and navigation system mounted on the vehicle; and
when the motion state of the vehicle is in a linear motion state, determining a motion direction of the vehicle according to the linear motion state, and determining the positioning information of the vehicle according to the motion direction and the positioning and navigation system.
