CN-121971851-A - Object motion detection method, device, electronic equipment and storage medium

Abstract

The method comprises the steps of obtaining a model corresponding to a scene where an object is located, wherein the model is provided with collision grids, obtaining a blocking direction corresponding to the collision grids of the model, wherein the blocking direction is used for indicating a moving direction of the object when the object is blocked by the corresponding collision grids in a moving state, carrying out voxelization on the collision grids based on the blocking direction corresponding to the collision grids to obtain voxel data corresponding to the scene, each voxel in the voxel data is provided with a type identifier, the type identifier is determined based on the blocking direction of the collision grids corresponding to the voxels, obtaining a moving route of the object in the scene, detecting the moving process of the object in the scene based on the voxel data and the moving route, and obtaining a detection result, wherein the detection result is used for indicating whether the object is blocked in the moving process along the moving route. The present disclosure extends the applicable scenarios of motion detection of objects.

Inventors

• WANG CHEN

Assignees

• 星臻科技(上海)有限公司

Dates

Publication Date

20260505

Application Date

20251210

Claims (14)

1. 1. A method of object motion detection, the method comprising: Obtaining a model corresponding to a scene where an object is located, wherein the model is provided with a collision grid; Acquiring a blocking direction corresponding to the collision grid of the model, wherein the blocking direction is used for indicating the movement direction of the object when the object is blocked by the corresponding collision grid in a movement state; Voxel processing is carried out on the collision grid based on the blocking direction corresponding to the collision grid, voxel data corresponding to the scene is obtained, each voxel in the voxel data is provided with a type identifier, and the type identifier indicates blocking direction information determined based on the blocking direction of the collision grid corresponding to the voxel; Acquiring a motion route of the object in the scene; And detecting the movement process of the object in the scene based on the voxel data and the movement route to obtain a detection result, wherein the detection result is used for indicating whether the object is blocked in the movement process along the movement route.
2. 2. The method according to claim 1, wherein detecting the motion process of the object in the scene based on the voxel data and the motion route to obtain a detection result comprises: determining a starting position and an expected ending position of the object based on the movement route; Initializing a current position based on the starting position; determining a target position in the movement route based on the current position, wherein the target position is a position to which the object is expected to move through single movement along the movement route; Based on the voxels corresponding to the current position and the voxels corresponding to the target position, obtaining a single motion result, wherein the single motion result is used for indicating whether the object can move to the target position or not; and obtaining the detection result based on the single motion result.
3. 3. The method of claim 2, wherein the obtaining the detection result based on the single motion result comprises: obtaining a detection result indicating that the single movement is blocked in the process of moving along the movement route under the condition that the single movement result indicates that the single movement cannot move to the target position; If the single motion result indicates that the target position is moved to the expected end position, updating the current position based on the target position, updating the target position based on the current position and the motion direction, and repeatedly executing the voxels corresponding to the current position and the voxels corresponding to the target position to obtain a single motion result and the single motion result, thereby obtaining the detection result; And if the target position is the expected end position, obtaining a detection result indicating that the movement along the movement route is not blocked in the process of moving.
4. 4. The method of claim 1, wherein the type is identified as staining information, and wherein if different voxels in the voxel data have different staining information, collision grids corresponding to the different voxels have different blocking directions.
5. 5. The method of claim 4, wherein the voxel blocks the object bi-directionally if the staining information of the voxel indicates a first color, wherein the voxel blocks the object uni-directionally if the staining information of the voxel indicates a second color, and wherein the voxel does not block the object if the staining information of the voxel indicates a third color.
6. 6. The method of claim 2, wherein the type identifier is staining information, and the obtaining a single motion result based on the voxel corresponding to the current position and the voxel corresponding to the target position includes: If the height difference between the voxels corresponding to the current position and the voxels corresponding to the target position is smaller than a preset threshold value, obtaining a single motion result indicating motion to the target position; and if the height difference between the voxel corresponding to the current position and the voxel corresponding to the target position is greater than or equal to the preset threshold value, obtaining the single motion result based on the first dyeing information of the voxel corresponding to the current position and the second dyeing information of the voxel corresponding to the target position.
7. 7. The method of claim 6, wherein the voxel bi-directionally blocks the object if the staining information of the voxel indicates a first color, wherein the voxel uni-directionally blocks the object if the staining information of the voxel indicates a second color, and wherein the voxel does not block the object if the staining information of the voxel indicates a third color; the obtaining the single motion result based on the first staining information of the voxels corresponding to the current position and the second staining information of the voxels corresponding to the target position includes: if the first dyeing information indicates the first color, the second color or the third color, and the second dyeing information indicates the second color or the third color, obtaining a single movement result indicating movement to the target position; and if the second dyeing information indicates the first color, obtaining a single movement result indicating that the target position cannot be moved.
8. 8. The method according to claim 2, wherein the obtaining a single motion result based on the voxel corresponding to the current position and the voxel corresponding to the target position comprises: and if the voxel corresponding to the target position does not exist, obtaining a single motion result indicating to move to the target position.
9. 9. The method of claim 7, wherein the obtaining a single motion result based on the voxel corresponding to the current location and the voxel corresponding to the target location comprises: And if the voxel corresponding to the current position does not exist, but the voxel corresponding to the target position exists, obtaining the single motion result based on the second dyeing information of the voxel corresponding to the target position.
10. 10. The method of claim 9, wherein the obtaining the single motion result based on the second staining information of the voxel corresponding to the target location comprises: If the second dyeing information indicates the third color, obtaining a single movement result indicating movement to the target position; And if the second dyeing information indicates the first color or the second color, obtaining a single movement result indicating that the target position cannot be moved.
11. 11. An object motion detection apparatus, the apparatus comprising: the model acquisition module is configured to execute a model corresponding to a scene where an acquisition object is located, and the model is provided with a collision grid; The system comprises a model, a voxel data acquisition module, a voxel data processing module and a model generation module, wherein the model is used for acquiring a blocking direction corresponding to a collision grid of the model, and the blocking direction is used for indicating a motion direction of an object when the object is blocked by the corresponding collision grid in a motion state; And detecting the motion process of the object in the scene based on the voxel data and the motion route to obtain a detection result, wherein the detection result is used for indicating whether the object is blocked in the motion process along the motion route.
12. 12. An electronic device, comprising: A processor; A memory for storing the processor-executable instructions; wherein the processor is configured to execute the instructions to implement the object motion detection method of any one of claims 1 to 10.
13. 13. A computer readable storage medium, characterized in that the electronic device performs the object motion detection method according to any one of claims 1 to 10 when instructions in the computer readable storage medium are executed by a processor of the electronic device.
14. 14. A computer program product, characterized in that the computer program product comprises a computer program stored in a readable storage medium, from which at least one processor of a computer device reads and executes the computer program, causing the device to perform the object motion detection method according to any one of claims 1 to 10.

Description

Object motion detection method, device, electronic equipment and storage medium Technical Field The disclosure relates to the field of computer technology, and in particular, to an object motion detection method, an object motion detection device, electronic equipment and a storage medium. Background The related art may generate voxels based on a collision grid of an obstacle of a scene, and perform motion detection of an object using the voxels, i.e., if a voxel blocking is encountered during a motion of the object in the scene from a start point to an end point along a predetermined motion route, it may be determined that forward progress cannot be continued, such voxels having no information about blocking directions, and the object cannot go forward through the voxels from any direction. However, the related art does not address the modeling requirement for unidirectional traffic and the object motion detection requirement in unidirectional traffic scenes, for example, obstacles in some scenes allow objects to pass from one direction but cannot pass from other directions, and the current related art cannot perform motion detection of objects based on voxels in such scenes. Disclosure of Invention The present disclosure provides an object motion detection method, apparatus, electronic device, and storage medium to solve the problems in the related art. The technical scheme of the present disclosure is as follows: according to a first aspect of embodiments of the present disclosure, there is provided an object motion detection method, the method comprising: Obtaining a model corresponding to a scene where an object is located, wherein the model is provided with a collision grid; Acquiring a blocking direction corresponding to the collision grid of the model, wherein the blocking direction is used for indicating the movement direction of the object when the object is blocked by the corresponding collision grid in a movement state; Voxel processing is carried out on the collision grid based on the blocking direction corresponding to the collision grid, voxel data corresponding to the scene is obtained, each voxel in the voxel data is provided with a type identifier, and the type identifier indicates blocking direction information determined based on the blocking direction of the collision grid corresponding to the voxel; Acquiring a motion route of the object in the scene; And detecting the movement process of the object in the scene based on the voxel data and the movement route to obtain a detection result, wherein the detection result is used for indicating whether the object is blocked in the movement process along the movement route. In an exemplary embodiment, the detecting the motion process of the object in the scene based on the voxel data and the motion route to obtain a detection result includes: determining a starting position and an expected ending position of the object based on the movement route; Initializing a current position based on the starting position; determining a target position in the movement route based on the current position, wherein the target position is a position to which the object is expected to move through single movement along the movement route; Based on the voxels corresponding to the current position and the voxels corresponding to the target position, obtaining a single motion result, wherein the single motion result is used for indicating whether the object can move to the target position or not; and obtaining the detection result based on the single motion result. In an exemplary embodiment, the obtaining the detection result based on the single motion result includes: obtaining a detection result indicating that the single movement is blocked in the process of moving along the movement route under the condition that the single movement result indicates that the single movement cannot move to the target position; If the single motion result indicates that the target position is moved to the expected end position, updating the current position based on the target position, updating the target position based on the current position and the motion direction, and repeatedly executing the voxels corresponding to the current position and the voxels corresponding to the target position to obtain a single motion result and the single motion result, thereby obtaining the detection result; And if the target position is the expected end position, obtaining a detection result indicating that the movement along the movement route is not blocked in the process of moving. In an exemplary embodiment, the type is identified as staining information, and if different voxels in the voxel data have different staining information, collision grids corresponding to the different voxels have different blocking directions. In an exemplary embodiment, the voxel blocks the object bi-directionally if the staining information of the voxel indicates a first color, the voxel blocks the object