CN114781149A - Method and system for automatically acquiring scene element information - Google Patents

Abstract

The invention relates to a method and a system for automatically acquiring scene element information, wherein a known and correct scene element sample set is established, and a scene picture to be identified is acquired; then determining a scene element identification task issuing strategy, wherein the identification task issuing strategy comprises a plurality of identification tasks, and each identification task comprises a plurality of known scene element samples and at least one scene picture to be identified; finally, the marking accuracy of the marking personnel on the known scene element sample is obtained, and if the marking accuracy is larger than a preset threshold value, multiple marking results of the marking personnel on the scene picture to be marked are obtained; and if the marking results are the same for multiple times, judging that the marking results are valid, and adding the marked scene picture to the scene element sample set. According to the invention, correct sample data sets are automatically increased step by step, the automatic driving scene data element definition is established, the scene database is quickly established, and effective support is provided for automatic driving development perception and decision-making algorithms.

Description

Method and system for automatic acquisition of scene element information

technical field

The invention relates to the technical field of automatic driving, in particular to a method and system for automatically acquiring scene element information of an automatic driving scene library.

Background technique

The automated driving simulation test technology can accelerate the development process and greatly reduce the development cost in the whole vehicle development process. Simulation test, as one of the main means of ICV product testing, complements the real vehicle test, and is gradually being paid attention to. However, how to quickly establish a system simulation scene library and fully verify the driving assistance function has become a major challenge. At present, the typical simulation scene library mainly comes from the data collected from public roads, and the description of the elements contained in the scene is limited to the results of sensor perception, which cannot be fully covered. All dynamic and static elements that affect perception and regulation.

SUMMARY OF THE INVENTION

Aiming at the technical problems existing in the prior art, the present invention provides a method and system for automatically acquiring scene element information. Through an online graphic identification verification system, a picture verification service is released for online websites and online APPs, so that scene elements are performed in a crowdsourcing manner. Collect verification services for the purpose of automatically identifying scene elements.

The technical scheme that the present invention solves the above-mentioned technical problems is as follows:

In a first aspect, the present invention provides a method for automatically acquiring scene element information, including:

Establish a known correct sample set of scene elements, and obtain the scene picture to be identified;

determining a scene element identification task release strategy, the identification task release strategy includes a plurality of identification tasks, and the identification task includes a plurality of known scene element samples and at least one scene picture to be identified;

Obtain the labeling accuracy rate of the known scene element samples by the labeling personnel, and if the labeling accuracy rate is greater than a preset threshold, obtain the results of multiple labeling of the scene pictures by the labeling personnel;

If multiple times of marking results are the same, it is determined that the marking result is valid, and the marked scene picture is added to the scene element sample set.

Further, the identification task includes 2 known scene element samples and 1 scene picture to be identified.

Further, if the marking of the two known scene element samples by the marking staff is correct, the multiple marking results of the scene pictures to be marked by the marking staff are acquired.

Further, the method also includes:

After marking each scene picture to be marked, the counter is incremented by 1 and the marking result is recorded;

If the labeling results are consistent for 5 consecutive times, it is determined that the labeling result is valid;

If any mark result is inconsistent within five times, the counter will be cleared.

In a second aspect, the present invention provides a system for automatically acquiring scene element information, including:

The data acquisition module is used to establish a known correct sample set of scene elements and acquire the scene pictures to be identified;

a strategy formulation module, configured to determine a scene element identification task release strategy, the identification task release strategy includes a plurality of identification tasks, and the identification task includes a plurality of known scene element samples and at least one scene picture to be identified;

The labeling comparison module is used to obtain the labeling accuracy rate of the known scene element samples by the labeling personnel. If the labeling accuracy rate is greater than the preset threshold, obtain the results of multiple labeling of the scene pictures by the labeling personnel; if the labeling results are multiple times If the same, it is determined that the marking result is valid, and the marked scene picture is added to the scene element sample set.

In a third aspect, the present invention provides an electronic device, comprising:

memory for storing computer software programs;

The processor is configured to read and execute the computer software program, thereby implementing the method for automatically acquiring scene element information according to the first aspect of the present invention.

In a fourth aspect, the present invention provides a non-transitory computer-readable storage medium, in which a computer software program for implementing the method for automatically acquiring scene element information described in the first aspect of the present invention is stored.

The beneficial effects of the present invention are as follows: the present invention first provides a sample set of correct scene elements, obtains an online graph verification system, establishes a graph result comparison algorithm, and publishes the scene recognition service in the form of a picture verification method service in an online website or APP, thereby Obtain a large number of scene picture identification results, and quickly build a scene database. Through the present invention, the correct sample data set is gradually added automatically, the definition of the data element of the automatic driving scene is established, the information of the long-term and temporary obstacles or traffic participation types in the scene is covered, the scene database is quickly established, and the perception and decision-making algorithm for the development of automatic driving is provided. Effective support.

Description of drawings

1 is a schematic flowchart of a method for automatically acquiring scene element information according to an embodiment of the present invention;

2 is a schematic structural diagram of a system for automatically acquiring scene element information according to an embodiment of the present invention;

3 is a schematic diagram of an embodiment of an electronic device provided by an embodiment of the present invention;

FIG. 4 is a schematic diagram of an embodiment of a computer-readable storage medium provided by an embodiment of the present invention.

Detailed ways

The principles and features of the present invention will be described below with reference to the accompanying drawings. The examples are only used to explain the present invention, but not to limit the scope of the present invention.

As shown in FIG. 1 , an embodiment of the present invention provides a method for automatically acquiring scene element information, including:

S1: Establish a known correct sample set of scene elements, and acquire a scene picture to be identified.

S2: Determine a scene element identification task release strategy, where the identification task release strategy includes a plurality of identification tasks, and the identification tasks include a plurality of known scene element samples and at least one scene picture to be identified. The identification task in this embodiment includes two known scene element samples and one scene picture to be identified.

S3, obtain the labeling accuracy rate of the known scene element samples by the labeling personnel, and if the labeling accuracy rate is greater than a preset threshold, obtain the multiple labeling results of the labeling scene pictures by the labeling personnel; if the multiple labeling results are the same, it is determined that the The marked result is valid, and the marked scene picture is added to the scene element sample set.

In this embodiment, if the labeling personnel are correct in labeling the two known scene element samples, multiple times of labeling results of the scene pictures to be labelled by the labeling personnel are obtained.

After marking each scene picture to be marked, the counter is incremented by 1 and the marking result is recorded;

If the labeling results are consistent for 5 consecutive times, it is determined that the labeling result is valid;

If any mark result is inconsistent within five times, the counter will be cleared.

Through the above process, gradually increase the correct sample data set automatically, establish the definition of the data element of the automatic driving scene, cover the long-term, temporary obstacle or traffic participation type information in the scene, quickly establish the scene database, and provide the development of perception and decision-making algorithms for automatic driving. Effective support.

As shown in FIG. 2 , an embodiment of the present invention further provides a system for automatically acquiring scene element information, including:

The data acquisition module is used to establish a known correct sample set of scene elements and acquire the scene pictures to be identified;

a strategy formulation module, configured to determine a scene element identification task release strategy, the identification task release strategy includes a plurality of identification tasks, and the identification task includes a plurality of known scene element samples and at least one scene picture to be identified;

The labeling comparison module is used to obtain the labeling accuracy rate of the known scene element samples by the labeling personnel. If the labeling accuracy rate is greater than the preset threshold, obtain the results of multiple labeling of the scene pictures by the labeling personnel; if the labeling results are multiple times If the same, it is determined that the marking result is valid, and the marked scene picture is added to the scene element sample set.

Please refer to FIG. 3 , which is a schematic diagram of an embodiment of an electronic device provided by an embodiment of the present invention. As shown in FIG. 3 , an embodiment of the present invention provides an electronic device 500, which includes a memory 510, a processor 520, and a computer program 511 stored in the memory 520 and running on the processor 520, and the processor 520 executes the computer program At 511, implement the following steps:

Establish a known correct sample set of scene elements, and obtain the scene picture to be identified;

determining a scene element identification task release strategy, the identification task release strategy includes a plurality of identification tasks, and the identification task includes a plurality of known scene element samples and at least one scene picture to be identified;

Obtain the labeling accuracy rate of the known scene element samples by the labeling personnel, and if the labeling accuracy rate is greater than a preset threshold, obtain the results of multiple labeling of the scene pictures by the labeling personnel;

If multiple times of marking results are the same, it is determined that the marking result is valid, and the marked scene picture is added to the scene element sample set.

Please refer to FIG. 4, which is a schematic diagram of an embodiment of a computer-readable storage medium provided by an embodiment of the present invention. As shown in FIG. 4 , this embodiment provides a computer-readable storage medium 600 on which a computer program 611 is stored. When the computer program 611 is executed by a processor, the following steps are implemented:

Establish a known correct sample set of scene elements, and obtain the scene picture to be identified;

determining a scene element identification task release strategy, the identification task release strategy includes a plurality of identification tasks, and the identification task includes a plurality of known scene element samples and at least one scene picture to be identified;

Obtain the labeling accuracy rate of the known scene element samples by the labeling personnel, and if the labeling accuracy rate is greater than a preset threshold, obtain the results of multiple labeling of the scene pictures by the labeling personnel;

If multiple times of marking results are the same, it is determined that the marking result is valid, and the marked scene picture is added to the scene element sample set.

It should be noted that, in the foregoing embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to the relevant descriptions of other embodiments.

As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.

The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block in the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to the processor of a general purpose computer, special purpose computer, embedded computer or other programmable data processing device to produce a machine such that the instructions executed by the processor of the computer or other programmable data processing device produce Means implementing the functions specified in one or more of the flowcharts and/or one or more blocks of the block diagrams.

These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory result in an article of manufacture comprising instruction means, the instructions The apparatus implements the functions specified in the flow or flow of the flowcharts and/or the block or blocks of the block diagrams.

These computer program instructions can also be loaded on a computer or other programmable data processing device to cause a series of operational steps to be performed on the computer or other programmable device to produce a computer-implemented process such that The instructions provide steps for implementing the functions specified in the flow or blocks of the flowcharts and/or the block or blocks of the block diagrams.

Although preferred embodiments of the present invention have been described, additional changes and modifications to these embodiments may occur to those skilled in the art once the basic inventive concepts are known. Therefore, the appended claims are intended to be construed to include the preferred embodiment and all changes and modifications that fall within the scope of the present invention.

It will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the spirit and scope of the invention. Thus, provided that these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.

Claims (7)

1. A method for automatically acquiring scene element information is characterized by comprising the following steps:

establishing a known and correct scene element sample set, and acquiring a scene picture to be identified;

determining a scene element identification task issuing strategy, wherein the identification task issuing strategy comprises a plurality of identification tasks, and the identification tasks comprise a plurality of known scene element samples and at least one scene picture to be identified;

acquiring the marking accuracy of a marker on a known scene element sample, and acquiring a multi-marking result of the marker on a scene picture to be identified if the marking accuracy is greater than a preset threshold;

and if the marking results are the same for multiple times, judging that the marking results are effective, and adding the marked scene pictures to the scene element sample set.

2. The method according to claim 1, wherein the identification task comprises 2 known scene element samples and 1 scene picture to be identified.

3. The method according to claim 2, wherein if the labeling of the 2 known scene element samples by the labeling person is correct, a plurality of labeling results of the labeling person for the scene picture to be identified are obtained.

4. The method of claim 1 or 3, further comprising:

after marking the scene picture to be identified each time, adding 1 to a counter and recording a marking result;

if the marking results of 5 continuous times are consistent, judging that the marking results are valid;

and if the marking result of any one time in the five times is inconsistent, clearing the counter.

5. An automatic scene element information acquisition system, comprising:

the data acquisition module is used for establishing a known and correct scene element sample set and acquiring a scene picture to be identified;

the system comprises a strategy making module, a scene element identification task issuing module and a scene element identification task processing module, wherein the strategy making module is used for determining a scene element identification task issuing strategy, the identification task issuing strategy comprises a plurality of identification tasks, and the identification tasks comprise a plurality of known scene element samples and at least one scene picture to be identified;

the marking comparison module is used for acquiring the marking accuracy of the marking personnel on the known scene element samples, and acquiring a plurality of marking results of the marking personnel on the scene picture to be marked if the marking accuracy is greater than a preset threshold value; and if the marking results are the same for multiple times, judging that the marking results are valid, and adding the marked scene picture to the scene element sample set.

6. An electronic device, comprising:

a memory for storing a computer software program;

a processor, configured to read and execute the computer software program, so as to implement the method for automatically acquiring scene element information according to any one of claims 1 to 4.

7. A non-transitory computer-readable storage medium, wherein a computer software program for implementing the automatic scene element information obtaining method according to any one of claims 1 to 4 is stored in the storage medium.

Images