SYSTEM AND METHOD FOR ANNOTATING VIDEO

• US 20140092252A1
• Filed: 05/10/2012
• Published: 04/03/2014
• Est. Priority Date: 05/12/2011
• Status: Abandoned Application

• Current Assignee: Magna Electronics Incorporated (Magna International Incorporated)
• Original Assignee: Magna Electronics Incorporated (Magna International Incorporated)
• Inventors: Nix, Axel
• Application Number:
  US14/116,859
• Publication Number:
  US 20140092252A1
• Time in Patent Office: Days
• US Class Current: 348/148
• CPC Class Codes: G06F 11/3668 Software testing software testing in telephone exchanges H04M3/242, testing of hardware G06F11/22
• CPC Class Codes: G06F 18/217 Validation; Performance evaluation; Active pattern learning techniques
• CPC Class Codes: H04N 7/181 for receiving images from a plurality of remote sources

Abstract

A method of providing test data for a vehicle may use the test data to verify the performance of a system in the vehicle under different environmental conditions (such as at night, while it is raining, during high glare conditions, during fog and/or the like). The method entails driving a test vehicle through a selected set of environmental conditions. Environment data, such as, for example, images, are captured while driving the test vehicle. The environment data relates to the environment outside the test vehicle. The environment data may be recorded to a memory. Additionally, environmental condition data relating to the environmental conditions outside the vehicle while it is being driven may be recorded to the memory.

Claims

1. A method of determining a success rate of a sensing system when determining properties of the environment outside a vehicle, said method comprising:

(a) driving a test vehicle through a set of environmental conditions;

(b) capturing first environment data relating to the environment outside the test vehicle using a first environment sensing system, during step (a);

(c) determining properties of the environment;

(d) recording to a memory (i) the first environment data, (ii) the determined properties of the environment, (iii) vehicle data relating to at least one property of the vehicle while the vehicle is driven during step (a), and (iv) environmental condition data relating to the environmental conditions outside the vehicle while the vehicle is driven during step (a);

(e) providing a processor and software executable by the processor;

(f) using the software to determine properties of the environment based on the first environment data;

(g) determining a success rate of the software at determining the properties of the environment based on the determination of properties of the environment made in step (c);

(h) determining whether the success rate determined in step (g) exceeds a selected success rate; and

(i) iteratively repeating steps (f), (g) and (h) using modified software until the success rate determined at step (g) exceeds the selected success rate.

2. A method as claimed in claim 1, wherein a first iteration of step (g) takes place during step (a).

3. A method as claimed in claim 1, wherein the first environment data comprises images captured from a camera on board the vehicle.

4. A method as claimed in claim 1, wherein the properties of the environment include the presence of obstacles in the path of the vehicle.

5. A method as claimed in claim 1, wherein the properties of the environment include the presence of pedestrians in the path of the vehicle.

6. A method as claimed in claim 1, wherein the environmental condition data includes at least one datum selected from the group of environmental condition data consisting of temperature, dew point, amount of cloud coverage, humidity, time for sunrise, time for sunset, a level of precipitation and a type of precipitation.

7. A method as claimed in claim 1, wherein the vehicle data includes at least one datum selected from the group of vehicle data consisting of vehicle speed, vehicle steering angle, operational state of windshield wipers, and operational state of vehicle headlights.

8. A method as claimed in claim 1, wherein at least some of the environmental condition data is obtained from the internet via a wireless connection thereto on board the vehicle, and is based at least in part on positional information relating to the vehicle.

9. A method as claimed in claim 8, wherein the positional information is taken from a GPS sensor on board the vehicle.

10. A method as claimed in claim 8, wherein the positional information for the vehicle is recorded to the memory in step (e).

11. A method as claimed in claim 1, wherein the environmental condition data covers a plurality of different environmental conditions, and wherein step (f) includes:

(j) selecting a subset of environmental conditions covered by the environmental condition data; and

(k) using software to determine properties of the environment based on a subset of the first environment data taken under environmental conditions corresponding to the environmental conditions selected in step (j).

12. A method as claimed in claim 1, wherein step (c) takes place during step (a).

13. A method as claimed in claim 1, wherein the vehicle data includes bearing information for the vehicle and wherein the environmental condition data includes the position of the sun in the sky during step (a).

14. A method as claimed in claim 1, further comprising recording to the memory in step (e) the time of day during step (a).

15. A method as claimed in claim 1, further comprising capturing second environment data relating to the environment outside the test vehicle using a second environment sensing system, during step (a), and wherein step (d) further includes recording the second environment data to the memory, and wherein the determination of properties of the environment in step (c) are made at least in part on the second environment data.

16-21. -21. (canceled)

22. A method of determining a success rate of a sensing system when determining properties of the environment outside a vehicle, said method comprising:

(a) driving a test vehicle through a set of environmental conditions;

(b) during step (a), capturing first environment data relating to the environment outside the test vehicle using a first environment sensing system, wherein said first environment sensing system comprises at least one camera;

(c) during step (a), capturing second environment data using a second environment sensing system;

(d) recording to a memory (i) the first environment data, (ii) the second environment data, (iii) vehicle data relating to at least one property of the vehicle while the vehicle is driven during step (a), and (iv) environmental condition data relating to the environmental conditions outside the vehicle while the vehicle is driven during step (a);

(e) determining properties of the environment based at least in part on the second environment data;

(f) providing a processor and software executable by the processor;

(g) using the software to determine properties of the environment based on the first environment data;

(h) determining a success rate of the software at determining the properties of the environment based on a comparison of the determination of properties of the environment in step (g) and the determination of properties of the environment made in step (e);

(i) determining whether the success rate determined in step (h) exceeds a selected success rate; and

(j) iteratively repeating steps (g), (h) and (i) using modified software until the success rate determined at step (h) exceeds the selected success rate.

23. A method as claimed in claim 22, wherein the properties of the environment include at least one of the presence of obstacles in the path of the vehicle and the presence of pedestrians in the path of the vehicle.

24. A method as claimed in claim 22, wherein at least one of (i) the environmental condition data includes at least one datum selected from the group of environmental condition data consisting of temperature, dew point, amount of cloud coverage, humidity, time for sunrise, time for sunset, a level of precipitation and a type of precipitation, (ii) the vehicle data includes at least one datum selected from the group of vehicle data consisting of vehicle speed, vehicle steering angle, operational state of windshield wipers, and operational state of vehicle headlights, and (iii) at least some of the environmental condition data is obtained from the internet via a wireless connection thereto on board the vehicle, and is based at least in part on positional information relating to the vehicle.

25. A method of determining a success rate of a sensing system when determining properties of the environment outside a vehicle, said method comprising:

(a) driving a test vehicle through a set of environmental conditions;

(b) during step (a), capturing first environment data relating to the environment outside the test vehicle using a first environment sensing system, wherein said first environment sensing system comprises at least one camera;

(c) during step (a), capturing second environment data using a second environment sensing system, wherein said second environment sensing system comprises at least one radar sensor;

(d) determining properties of the environment based at least in part on the second environment data, wherein the determined properties of the environment include the presence of an object in the path of the vehicle;

(e) providing a processor and software executable by the processor;

(f) using the software to determine properties of the environment based on the first environment data;

(g) determining a success rate of the software at determining the properties of the environment based on a comparison of the determination of properties of the environment in step (f) and the determination of properties of the environment made in step (d);

(h) determining whether the success rate determined in step (g) exceeds a selected success rate; and

(i) iteratively repeating steps (f), (g) and (h) using modified software until the success rate determined at step (g) exceeds the selected success rate.

26. A method as claimed in claim 25, comprising capturing environmental condition data relating to the environmental conditions outside the vehicle while the vehicle is driven during step (a), wherein determining a success rate of the software in step (g) comprises determining a success rate of the software during exposure of the vehicle to different environmental conditions while the vehicle is driven during step (a).