Database construction method, positioning method, and related device

1. A database construction method, comprising:

determining a target image set that meets a preset image coincidence degree requirement, wherein the target image set comprises at least one image, and wherein each of the at least one image corresponds to a type of natural condition information;

obtaining a scenario key point information set based on the target image set and the natural condition information;

determining, in the scenario key point information set, first scenario key point information corresponding to a first scenario key point whose first life value on a single mobile device is greater than a first preset life value threshold, wherein the first life value represents a probability that the first scenario key point is a static scenario key point;

determining, in the first scenario key point information, second scenario key point information corresponding to a second scenario key point whose second life value on a plurality of mobile devices is greater than a second preset life value threshold; and

constructing the database based on the second scenario key point information when the second scenario key point information does not match preset scenario key point information in a database.

2. The database construction method of claim 1 , wherein the second scenario key point information does not match the preset scenario key point information when the second scenario key point information does not exist in the database, wherein the database construction method further comprises adding the second scenario key point information to the database, and wherein the second scenario key point information comprises target descriptor information about target natural condition information.

3. The database construction method of claim 1 , wherein the second scenario key point information does not match the preset scenario key point information when the second scenario key point information exists in the database and when the second scenario key point information does not comprise target descriptor information about target natural condition information, wherein the database construction method further comprises adding the target descriptor information to the preset second scenario key point information in the database.

4. The database construction method of claim 2 , wherein before adding the second scenario key point information to the database, the database construction method further comprises:

determining three-dimensional (3D) coordinates of the second scenario key point; and

determining that the second scenario key point information does not exist in the database when a difference between 3D coordinates of each preset scenario key point in the database and the 3D coordinates of the second scenario key point is greater than a first preset threshold or determining that the second scenario key point information exists in the database when a difference between 3D coordinates of any one preset scenario key point in the database and the 3D coordinates of the second scenario key point is less than the first preset threshold.

5. The database construction method of claim 3 , wherein before adding the target descriptor information to the preset second scenario key point information, the database construction method further comprises:

determining at least one piece of descriptor information of a preset second scenario key point in the database, wherein a distance between a first descriptor corresponding to the at least one piece of descriptor information and a second descriptor corresponding to the target descriptor information is less than a preset distance threshold; and

determining that the preset second scenario key point information in the database does not comprise the target descriptor information when the at least one piece of descriptor information is not determined.

6. The database construction method of claim 1 , further comprising:

selecting an image based on a preset distance spacing; and

determining that the image belongs to the target image set when a difference between a coincidence degree of the image and a preset coincidence degree threshold is within a preset precision range, wherein the preset distance spacing is d k+1 , wherein the preset distance spacing is based on the following equation:

d k+1 =d k +d k (α*−α),

wherein α* is the preset coincidence degree threshold, wherein d k is a distance spacing for image selection at a previous moment, and wherein α is a coincidence degree of the image when the image is selected based on the distance spacing d k .

7. The database construction method of claim 1 , further comprising:

selecting an image based on a preset angle spacing; and

determining that the image belongs to the target image set when a difference between a coincidence degree of the image and a preset coincidence degree threshold is within a preset precision range, wherein the preset angle spacing is θ k+1 , wherein the preset angle spacing is based on the following equation:

θ k+1 =θ k +θ k (α*−α),

wherein α* is the preset coincidence degree threshold, wherein θ k is an angle spacing for image selection at a previous moment, and wherein α is a coincidence degree of the image when the image is selected based on the angle spacing θ k .

8. The database construction method of claim 1 , wherein the scenario key point information comprises descriptor information corresponding to the natural condition information, and wherein the database construction method further comprises:

processing a target image to obtain a scenario key point;

constituting the scenario key point information using the scenario key point, the target image, and natural condition information corresponding to the target image; and

repeating the processing and the constituting until the scenario key point information set is constituted.

9. The database construction method of claim 1 , wherein after constructing the database, the database construction method further comprises:

determining third scenario key point information in the constructed database; and

deleting the third scenario key point information from the database when a feature number controlling score (FNCS) of a third scenario key point corresponding to the third scenario key point information is less than a preset FNCS threshold, wherein the FNCS is based on the following equation:

FNCS

=

m

i

M

⁢

log

⁢

N

′

n

i

⁢

wherein

⁢

⁢

m

i

M

is a probability that the third scenario key point is used during positioning, and wherein

N

n

i

is a percentage of a quantity of descriptors of the third scenario key point in a total quantity of descriptors in an image to which the third scenario key point belongs.

10. A network device, comprising:

a processor; and

a memory coupled to the processor and configured to store instructions that, when executed by the processor, cause the network device to be configured to:

determine a target image set that meets a preset image coincidence degree requirement, wherein the target image set comprises at least one image, and wherein each of the at least one image corresponds to a type of natural condition information;

obtain a scenario key point information set based on the target image set and the natural condition information;

determine, in the scenario key point information set, first scenario key point information corresponding to a first scenario key point whose first life value on a single mobile device is greater than a first preset life value threshold, wherein the first life value represents a probability that the first scenario key point is a static scenario key point;

determine, in the first scenario key point information, second scenario key point information corresponding to a second scenario key point whose second life value on a plurality of mobile devices is greater than a second preset life value threshold; and

construct a database based on the second scenario key point information when the second scenario key point information does not match preset scenario key point information in the database.

11. The network device of claim 10 , wherein the second scenario key point information does not match the preset scenario key point information when the second scenario key point information does not exist in the database, wherein the instructions further cause the processor to be configured to add the second scenario key point information to the database, and wherein the second scenario key point information comprises target descriptor information about target natural condition information.

12. The network device of claim 10 , wherein the second scenario key point information does not match the preset scenario key point information when the second scenario key point information exists in the database, and the second scenario key point information does not comprise target descriptor information about target natural condition information, wherein the instructions further cause the processor to be configured to add the target descriptor information to the preset second scenario key point information in the database.

13. The network device of claim 11 , wherein the instructions further cause the processor to be configured to:

determine three dimensional (3D) coordinates of the second scenario key point; and

determine that the second scenario key point information does not exist in the database when a difference between 3D coordinates of each preset scenario key point in the database and the 3D coordinates of the second scenario key point is greater than a first preset threshold, or determine that the second scenario key point information exists in the database when a difference between 3D coordinates of any one preset scenario key point in the database and the 3D coordinates of the second scenario key point is less than the first preset threshold.

14. The network device of claim 12 , wherein the instructions further cause the processor to be configured to:

determine at least one piece of descriptor information of a preset second scenario key point in the database, wherein a distance between a first descriptor corresponding to the at least one piece of descriptor information and a second descriptor corresponding to the target descriptor information is less than a preset distance threshold; and

determine that the preset second scenario key point information in the database does not comprise the target descriptor information when the network device does not determine the at least one piece of descriptor information.

15. The network device of claim 10 , wherein the instructions further cause the processor to be configured to:

select an image based on a preset distance spacing; and

determine that the image belongs to the target image set when a difference between a coincidence degree of the image and a preset coincidence degree threshold is within a preset precision range, wherein the preset distance spacing is d k+1 , wherein the preset distance spacing is based on the following equation:

d k+1 =d k +d k (α*−α),

wherein α* is the preset coincidence degree threshold, wherein d k is a distance spacing for image selection at a previous moment, and wherein α is a coincidence degree of the image when the image is selected based on the distance spacing d k .

16. The network device of claim 10 , wherein the instructions further cause the processor to be configured to:

select an image based on a preset angle spacing; and

determine that the image belongs to the target image set when a difference between a coincidence degree of the image and a preset coincidence degree threshold is within a preset precision range, wherein the preset angle spacing is θ k+1 , wherein the preset angle spacing is based on the following equation:

θ k+1 =θ k +θ k (α*−α),

wherein α* is the preset coincidence degree threshold, wherein θ k is an angle spacing for image selection at a previous moment, and wherein α is a coincidence degree of the image when the image is selected based on the angle spacing θ k .

17. The network device of claim 10 , wherein the scenario key point information comprises descriptor information corresponding to the natural condition information, and wherein the instructions further cause the processor to be configured to:

process a target image to obtain a scenario key point;

constitute the scenario key point information using the scenario key point, the target image, and natural condition information corresponding to the target image; and

repeat processing and the constituting until the scenario key point information set is constituted.