# US 2022/0036727 A1 · Traffic condition prediction system and traffic condition prediction method

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

- Number: US 2022/0036727 A1
- Document ID: US20220036727A1
- Status: Published application
- Kind: A1
- Application number: 17/019,249
- Filed: 2020-09-12
- Published: 2022-02-03
- Earliest claimed: 2020-07-29
- Assignees: Institute for Information Industry
- Applicants: Institute For Information Industry
- Inventors: Yen-Lin CHEN; Hsiu-Chih CHEN; Chao-Wei YU; Chieh-Min CHANG; Meng-Tsan LI
- Web page: https://patentgrep.com/patent/US20220036727A1

## CPC

- G08G 1/0133: {for classifying traffic situation}
- G06F 18/22: Matching criteria, e.g. proximity measures
- G06T 7/73: using feature-based methods
- G06V 10/40: Extraction of image or video features
- G06V 10/46: Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features (colour feature extraction G06V10/56)
- G06V 10/761: {Proximity, similarity or dissimilarity measures}
- G06V 20/52: Surveillance or monitoring of activities, e.g. for recognising suspicious objects (recognising microscopic objects G06V20/69)
- G06V 20/54: of traffic, e.g. cars on the road, trains or boats
- G08G 1/0116: {from roadside infrastructure, e.g. beacons}
- G08G 1/0129: {for creating historical data or processing based on historical data}
- G08G 1/0145: {for active traffic flow control}
- G08G 1/04: using optical or ultrasonic detectors
- G06T 2207/30232: Surveillance
- G06T 2207/30236: Traffic on road, railway or crossing
- G06T 2207/30241: Trajectory
- G08G: TRAFFIC CONTROL SYSTEMS (guiding railway traffic, ensuring the safety of railway traffic B61L; arrangement of road signs or traffic signals E01F9/00; radar or analogous systems, sonar systems, lidar systems specially adapted for traffic control G01S13/91, G01S15/88, G01S17/88; {radar or analogous systems, sonar systems, lidar systems specially adapted for anti-collision purposes G01S13/93, G01S15/93, G01S17/93})
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

A traffic condition prediction system and a traffic condition prediction method are disclosed. The method includes: determining a center of circle in a surveillance image of a target traffic scene; determining a first circle based on the center of circle and a first radius; extracting a plurality of first feature points along the circumference of the first circle according to a first preset sampling frequency; generating a scene feature of the target traffic scene at least based on the first feature points; determining whether the scene feature and a scene feature of another traffic scene are similar; and when determining that they are similar, predicting traffic condition of the target traffic scene through a prediction model used for predicting traffic condition of the other traffic scene. The scene feature of the target traffic scene and that of the other traffic scene are generated in a same way.

## Claims

### Claim 1

A traffic condition prediction system, comprising:
a storage, being configured to store a first scene feature of a first traffic scene, a first prediction model for predicting traffic condition of the first traffic scene, and a surveillance image of a second traffic scene; and
a processor electrically connected to the storage, being configured to:
determine a center of circle in the surveillance image;
determine a first circle based on the center of circle and a first radius;
extract a plurality of first feature points on the surveillance image along the circumference of the first circle according to a first preset sampling frequency;
generate a second scene feature of the second traffic scene at least according to the plurality of first feature points;
determine whether the first scene feature is similar to the second scene feature; and
when determining that the first scene feature is similar to the second scene feature, predict traffic condition of the second traffic scene through the first prediction model;
wherein the first scene feature and the second scene feature are generated in a same way.

### Claim 2 (depends on 1)

The traffic condition prediction system of claim 1, wherein:
the processor is further configured to:
determine a second circle based on the center of circle and a second radius; and
extract a plurality of second feature points on the surveillance image along the circumference of the second circle according to a second preset sampling frequency; and
the processor generates the second scene feature according to the plurality of first feature points and the plurality of second feature points.

### Claim 3 (depends on 1)

The traffic condition prediction system of claim 1, wherein each of the first scene feature and the second scene feature is a numerical sequence, and the processor compares the two numerical sequences based on a rotation invariance to determine whether the first scene feature and the second scene feature are similar.

### Claim 4 (depends on 1)

The traffic condition prediction system of claim 1, wherein the surveillance image of the second traffic scene is taken by a camera.

### Claim 5

A traffic condition prediction system, comprising:
a storage, being configured to store a first scene feature of a first traffic scene, a first prediction model for predicting traffic condition of the first traffic scene, at least one actual vehicle trajectory related to a second traffic scene, and a surveillance image of the second traffic scene; and
a processor electrically connected to the storage, being configured to:
determine a center of circle in the surveillance image;
determine a first circle based on the center of circle and a first radius;
extract a plurality of first feature points on the surveillance image along the circumference of the first circle according to a first preset sampling frequency;
generate a second scene feature of the second traffic scene at least according to the plurality of first feature points;
determine whether the first scene feature is similar to the second scene feature;
when determining that the first scene feature is similar to the second scene feature, predict at least one vehicle trajectory related to the second traffic scene through the first prediction model, and determine whether the at least one actual vehicle trajectory is similar to the at least one predicted vehicle trajectory; and
when determining that the at least one actual vehicle trajectory is similar to the at least one predicted vehicle trajectory, predict the traffic condition of the second traffic scene through the first prediction model;
wherein the first scene feature and the second scene feature are generated in a same way.

### Claim 6 (depends on 5)

The traffic condition prediction system of claim 5, wherein:
the processor is further configured to:
determine a second circle based on the center of circle and a second radius; and
extract a plurality of second feature points on the surveillance image along the circumference of the second circle according to a second preset sampling frequency; and
the processor generates the second scene feature according to the plurality of first feature points and the plurality of second feature points.

### Claim 7 (depends on 5)

The traffic condition prediction system of claim 5, wherein each of the first scene feature and the second scene feature is a numerical sequence, and the processor compares the two numerical sequences based on a rotation invariance to determine whether the first scene feature and the second scene feature are similar.

### Claim 8 (depends on 5)

The traffic condition prediction system of claim 5, wherein the surveillance image of the second traffic scene is taken by a camera.

### Claim 9 (depends on 5)

The traffic condition prediction system of claim 5, wherein each of the at least one actual vehicle trajectory is represented by a binary code sequence, and each of the at least one predicted vehicle trajectory is also represented by a binary code sequence.

### Claim 10 (depends on 9)

The traffic condition prediction system of claim 9, wherein each of all the binary code sequences representing the at least one actual vehicle trajectory and the at least one predicted vehicle trajectory is defined by a first symbol and a second symbol, the first symbol represents a presence of lane switch, and the second symbol represents an absence of lane switch.

### Claim 11

A traffic condition prediction method for a traffic condition prediction system, the traffic condition prediction system storing a first scene feature of a first traffic scene, a first prediction model for predicting traffic condition of the first traffic scene, and a surveillance image of a second traffic scene, the traffic condition prediction method comprising:
determining a center of circle in the surveillance image;
determining a first circle based on the center of circle and a first radius;
extracting a plurality of first feature points on the surveillance image along the circumference of the first circle according to a first preset sampling frequency;
generating a second scene feature of the second traffic scene at least according to the plurality of first feature points;
determining whether the first scene feature is similar to the second scene feature; and
when determining that the first scene feature is similar to the second scene feature, predicting traffic condition of the second traffic scene through the first prediction model;
wherein the first scene feature and the second scene feature are generated in a same way.

### Claim 12 (depends on 11)

The traffic condition prediction method of claim 11, further comprising:
determining a second circle based on the center of circle and a second radius; and
extracting a plurality of second feature points on the surveillance image along the circumference of the second circle according to a second preset sampling frequency;
wherein the traffic condition prediction system generates the second scene feature according to the plurality of first feature points and the plurality of second feature points.

### Claim 13 (depends on 11)

The traffic condition prediction method of claim 11, wherein each of the first scene feature and the second scene feature is a numerical sequence, and the traffic condition prediction method further comprises:
comparing the two numerical sequences based on a rotation invariance to determine whether the first scene feature and the second scene feature are similar.

### Claim 14 (depends on 11)

The traffic condition prediction method of claim 11, wherein the surveillance image of the second traffic scene is taken by a camera.