IP Library Granted Patent US 10,217,005
Granted Patent B2
US 10,217,005 · App. 15/628,667 · Granted Feb 26, 2019

Method, apparatus and device for generating target detection information

Inventors: Wei Liu (Shenyang, CN); Xin Yang (Shenyang, CN); Lu Wei (Shenyang, CN)
Assignees: NEUSOFT CORPORATION; NEUSOFT REACH AUTOMOTIVE TECHNOLOGY (SHANGHAI) CO., LTD.
G06K9/00791G01C11/04G01S13/726G01S13/865G01S13/867G01S13/91G06K9/6215G06K2209/21
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 10,217,005
App. No.
15/628,667
Granted
Feb 26, 2019
Kind
B2
Abstract

Provided is a method for generating target detection information, including detecting target objects around the vehicle by multiple different types of sensors, and determining the detection targets representing a same target object, detected by the different types of sensors, by spatial position and time tracing. With taking the target object as the detection result, the target detection information generated for the detection result includes a spatial matching confidence of the detection result in the current detection period, a time matching confidence of the detection result, and target detection information on each of the detection targets representing the detection result, collected by the sensor detecting the detection target in a dimension corresponding to the sensor.

Claims (33)

1. A method for generating target detection information, comprising:

determining, in a unified plane coordinate system, spatial positions of initial detection targets detected by a plurality of different types of sensors in a current detection period;

matching spatial positions of initial detection targets to be matched detected by every two of the plurality of different types of sensors in the current detection period, and determining the initial detection targets under a detection result as result detection targets, wherein spatial positions of any two of the result detection targets under the detection result are matched with each other; and

generating target detection information of the detection result, wherein the target detection information of the detection result comprises target detection information on each of the result detection targets, which is collected by the sensor detecting the result detection target in a dimension corresponding to the sensor, and wherein each of the result detection targets is the initial detection target detected by the sensor.

2. The method according to claim 1 , further comprising:

calculating a spatial matching confidence of the detection result in the current detection period based on the number of successful matches among the spatial positions of all the result detection targets in the current detection period, wherein the target detection information of the detection result further comprises the spatial matching confidence of the detection result in the current detection period.

3. The method according to claim 2 , wherein generating target detection information of the detection result comprises:

generating the target detection information of the detection result in a case that the spatial matching confidence of the detection result in the current detection period is greater than a preset spatial confidence threshold.

4. The method according to claim 2 , further comprising:

determining a time matching confidence of the detection result by performing a weighting operation on spatial matching confidences of the detection result in a plurality of recent detection periods, wherein the target detection information of the detection result further comprises the time matching confidence of the detection result.

5. The method according to claim 4 , wherein generating target detection information of the detection result comprises: generating the target detection information of the detection result in a case that the time matching confidence of the detection result is greater than a preset time confidence threshold.

6. The method according to claim 4 , wherein determining the time matching confidence of the detection result by performing the weighting operation on the spatial matching confidences of the detection result in the plurality of recent detection periods comprises:

adding results obtained by multiplying a spatial matching confidence in each of the plurality of recent detection periods by a weighting coefficient corresponding to the detection period, to obtain the time matching confidence of the detection result, wherein the closer the detection period is to a current time instant, the greater the weighting coefficient corresponding to the detection period is.

7. The method according to claim 6 , wherein the weighting coefficient corresponding to each of the detection periods is a normalized weighting coefficient.

8. The method according to claim 2 , wherein the spatial matching confidence of the detection result in the current detection period is a ratio of the number of successful matches corresponding to the detection result to the total number of matches, wherein the number of successful matches corresponding to the detection result is the number of successful matches among the spatial positions of all the result detection targets in the current detection period, and the total number of matches is the number of matches among all the initial detection targets to be matched in the current detection period.

9. The method according to claim 1 , wherein it is determined that spatial positions of two of the initial detection targets to be matched are matched with each other, in a case that a Euclidean distance between the spatial positions of the two of the initial detection targets to be matched is within a preset distance threshold.

10. The method according to claim 1 , further comprising:

determining a screening range in the plane coordinate system based on a current speed of a vehicle; and

determining the initial detection targets to be matched based on the screening range, wherein spatial positions of the initial detection targets to be matched are within the screening range.

11. The method according to claim 10 , wherein the screening range is a range of all spatial positions able to be passed by the vehicle in a case that a turning radius exceeds a current radius threshold and a path does not exceed a current path threshold, wherein the current radius threshold is determined based on the current speed of the vehicle and a lateral acceleration threshold, and the current path threshold is determined based on the current speed of the vehicle and a preset time threshold.

12. The method according to claim 1 , wherein the multiple different types of sensors comprises at least one of a camera, a radar and a laser radar, a dimension corresponding to the camera is an image characteristic of a detected target object, a dimension corresponding to the radar is a relative speed of the detected target object with respect to the vehicle, and a dimension corresponding to the laser radar is a profile of the detected target object.

13. An apparatus for generating target detection information, comprising:

a first determining unit configured to determine, in a unified plane coordinate system, spatial positions of initial detection targets detected by a plurality of different types of sensors in a current detection period;

a second determining unit configured to match spatial positions of initial detection targets to be matched detected by every two of the plurality of different types of sensors in the current detection period, and determine the initial detection targets under a detection result as result detection targets, wherein spatial positions of any two of the result detection targets under the detection result are matched with each other; and

a generating unit configured to generate target detection information of the detection result, wherein the target detection information of the detection result comprises target detection information on each of the result detection targets, which is collected by the sensor detecting the result detection target in a dimension corresponding to the sensor, and each of the result detection targets is the initial detection target detected by the sensor.

14. A device for generating target detection information, comprising a processor, a storage, a communication interface and a bus system, wherein

the bus system is configured to couple the processor, the storage and the communication interface together;

the communication interface is configured to implement communication connection between the device and at least one other device;

the storage is configured to storage instructions; and

the processor is configured to read the instructions stored in the storage, to perform steps of:

determining, in a unified plane coordinate system, spatial positions of initial detection targets detected by a plurality of different types of sensors in a current detection period;

matching spatial positions of initial detection targets to be matched detected by every two of the plurality of different types of sensors in the current detection period, and determining the initial detection targets under a detection result as result detection targets, wherein spatial positions of any two of the result detection targets under the detection result are matched with each other; and

generating target detection information of the detection result, wherein the target detection information of the detection result comprises target detection information on each of the result detection targets, which is collected by the sensor detecting the result detection target in a dimension corresponding to the sensor, and each of the result detection targets is the initial detection target detected by the sensor.

Assignments (2)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Sep 14, 2020
From: NEUSOFT CORPORATION
To: NEUSOFT REACH AUTOMOTIVE TECHNOLOGY (SHANGHAI) CO.,LTD.
Reel/Frame 053766/0158 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jun 21, 2017
From: LIU, WEI; YANG, XIN; WEI, LU
To: NEUSOFT CORPORATION; NEUSOFT REACH AUTOMOTIVE TECHNOLOGY (SHANGHAI) CO., LTD.
Reel/Frame 042921/0895 →
Priority Claims (1)
CN 2016 1 1259559 · Dec 30, 2016 · national
Continuity (1)
Related Publication 20180189575A1 · Jul 5, 2018
Cited By (4)
US 12,290,984 US 12,350,884 US 12,570,045 US 12,686,175