Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

A method and device for determining the recognition performance of a traffic light recognition system

A technology for identifying performance and identifying systems, applied in the computer field, can solve problems such as long test period, high test cost, and significant impact on safety, and achieve accurate and efficient evaluation and reduce test cost

Active Publication Date: 2018-06-22
BAIDU ONLINE NETWORK TECH (BEIJIBG) CO LTD
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Whether an unmanned vehicle can correctly identify traffic lights during driving has a great impact on driving safety, so a complete set of test methods and devices are needed to test whether the traffic light recognition system of unmanned vehicles can accurately identify traffic lights at intersections, as mentioned above It can be seen that unmanned vehicles need to use telephoto cameras, positioning systems based on GPS (Global Positioning System, Global Positioning System) / IMU (Inertial measurement unit, inertial measurement unit) and radar equipment and high-precision map data when performing traffic light recognition. , faced with the problems of complex test environment deployment, high test cost and long test cycle during the test process

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A method and device for determining the recognition performance of a traffic light recognition system
  • A method and device for determining the recognition performance of a traffic light recognition system
  • A method and device for determining the recognition performance of a traffic light recognition system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0022] figure 1 The flow chart of the method for determining the recognition performance of the traffic light recognition system provided by Embodiment 1 of the present invention, this embodiment is applicable to the situation where the recognition performance of the traffic light recognition system is tested, and the method can be performed by a computing device such as a server, an intelligent terminal or a customized On-board system to perform. Exemplarily, the traffic light recognition system can be applied to unmanned vehicles. When the unmanned vehicle is driving on the road, it is necessary to accurately identify the traffic lights on the road through the traffic light recognition system, and determine whether to drive through the intersection according to the instructions of the traffic lights.

[0023] Such as figure 1 Specifically, the following steps are shown:

[0024] Step 101: Receive collected video information including traffic lights, and determine the true ...

Embodiment 2

[0041] figure 2 The flow chart of the method for determining the recognition performance of the traffic light recognition system provided by Embodiment 2 of the present invention. On the basis of Embodiment 1, this embodiment provides a specific method for collecting video images of traffic lights and identifying them for the traffic light recognition system. Performance is tested to obtain the method for identifying performance. Such as figure 2 shown, including:

[0042] Step 201: Receive video information including traffic lights collected by a camera device in a specified scene, and perform virtualization processing on the video information to obtain video information of traffic lights in a preset scene.

[0043] Exemplarily, the camera device is fixed in a designated scene to collect video information, and at the same time, it is used to simulate the process of collecting video information during the driving process of a real vehicle, and the designated scene is deter...

Embodiment 3

[0053] image 3 The structural diagram of the device for determining the recognition performance of the traffic light recognition system provided by Embodiment 3 of the present invention specifically includes:

[0054] The video information processing module 1 is used to receive video information collected including traffic lights, and determine the true value data of the video information;

[0055] The simulation generation module 2 is used to generate a simulation positioning system according to the vehicle state information, and generate a simulation high-precision map according to the facility information of the traffic lights;

[0056] The recognition content acquisition module 3 is used to obtain the recognition content obtained by the traffic light recognition system according to the video information, the simulated positioning system and the simulated high-precision map;

[0057] The recognition performance determining module 4 is configured to determine the recogniti...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a method and a device for determining recognition performance of a traffic lamp recognition system. The method comprises the steps of: receiving acquired video information containing a traffic lamp, and determining truth-value data of the video information; generating a simulated positioning system according to vehicle state information, and generating a simulated high-precision map according to facility information of the traffic lamp; acquiring recognition content of the traffic lamp recognition system according to the video information, the simulated positioning system and the simulated high-precision map; and determining recognition performance of the traffic lamp recognition system according to the truth-value data and the recognition content. The method and the device achieve the fast, accurate and efficient evaluation of the traffic lamp recognition system, and greatly reduce the test cost.

Description

technical field [0001] Embodiments of the present invention relate to computer technology, and in particular to a method and device for evaluating a traffic light recognition system. Background technique [0002] With the development of artificial intelligence, smart cars have gradually entered the public's field of vision. Smart cars are also called unmanned vehicles, which sense the surrounding environment of the vehicle through on-board sensors, and the central control system of the vehicle comprehensively controls the steering and speed of the vehicle based on information such as roads and obstacles perceived by the sensors, so that the vehicle can be safely and reliably on the road. drive. [0003] Unmanned vehicles need to effectively identify traffic lights while driving on the road. The specific identification process is: use a telephoto camera to collect video images of traffic lights, determine the current position and attitude of the vehicle (orientation, rotatio...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G08G1/0962
CPCG08G1/09623
Inventor 胡太群
Owner BAIDU ONLINE NETWORK TECH (BEIJIBG) CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Eureka Blog
Learn More
PatSnap group products