Object collision prediction method and apparatus
This application provides a collision detection method and related apparatus. An image taken by a photographing unit may be used to predict whether a collision with a to-be-detected target will occur. In a current collision prediction method, a type of the to-be-detected target needs to be determined first based on the image taken by the photographing unit, which requires consuming of a large amount of computing power. In the collision prediction method provided in this application, a change trend of a distance between the to-be-detected target and a vehicle in which the apparatus is located may be determined based on the distances between the to-be-detected target and the vehicle at different moments, to predict a collision between the to-be-detected target and the vehicle. This method can improve efficiency in collision prediction and reduce energy consumption in predicting collision.
1. An object collision prediction method, wherein the method is applied to a computing device, the computing device is located in an object comprising a photographing unit, and the method comprises:
adjusting the photographing unit so that images taken by the photographing unit include a front portion of the object;
controlling the photographing unit to shoot a first image and a second image at a first moment and a second moment respectively, wherein the first image and the second image each comprise a to-be-detected target and the front portion of the object, and the second moment is later than the first moment;
measuring a first distance between the object and the to-be-detected target in the first image and a second distance between the object and the to-be-detected target in the second image; and
predicting, based on the first distance and the second distance, whether the object collides with the to-be-detected target.
2. The method according to claim 1 , wherein when the second distance is less than the first distance, the method further comprises:
obtaining a relative velocity between the to-be-detected target and the object based on a difference between the second distance and the first distance and a difference between the second moment and the first moment; and
predicting, based on the relative velocity and the second distance, a time in which the object collides with the to-be-detected target.
3. The method according to claim 1 , wherein the predicting, based on the first distance and the second distance, whether the object collides with the to-be-detected target comprises:
calculating a location difference between the object in the second image and the object in the first image; and
predicting, based on the first distance and a sum of the second distance and the location difference, whether the object collides with the to-be-detected target.
4. The method according to claim 1 , wherein the measuring a first distance between the object and the to-be-detected target in the first image and a second distance between the object and the to-be-detected target in the second image comprises:
obtaining two-dimensional borders of the to-be-detected target in the first image and the second image; and
measuring the first distance between the object and the two-dimensional borders of the to-be-detected target in the first image and the second distance between the object and the two-dimensional border of the to-be-detected target in the second image.
5. The method according to claim 4 , wherein when a distance between each pixel of a lower edge of the two-dimensional border of the to-be-detected target and the object is different, a shortest distance between a pixel comprised in the lower edge of the two-dimensional border of the to-be-detected target and the object is used as a distance between the object and the two-dimensional border of the to-be-detected target in the image.
6. The method according to claim 1 , wherein
the object is a vehicle, and the vehicle and the to-be-detected target are located in a same lane.
7. The method according to claim 6 , wherein the method further comprises:
identifying lane lines in the first image and the second image, wherein the lane lines comprise a first lane line and a second lane line, the first lane line and the second lane line are adjacent lane lines, and the vehicle and the to-be-detected target are located between the first lane line and the second lane line;
calculating a transverse velocity of the to-be-detected target based on a ratio of a distance between the to-be-detected target and the first lane line to a size of the to-be-detected target in the first image and a ratio of a distance between the to-be-detected target and the first lane line to a size of the to-be-detected target in the second image; and
predicting, based on the ratio of the distance between the to-be-detected target and the first lane line to the size of the to-be-detected target in the second image and the transverse velocity of the to-be-detected target, a time in which the to-be-detected target leaves the current lane.
8. A computing device, wherein the computing device is located in an object comprising a photographing unit, and the computing device comprises:
a processor; and
a computer-readable storage medium coupled to the processor and storing programming instructions for execution by the processor, wherein the programming instructions instruct the processor to control the photographing unit to shoot a first image and a second image at a first moment and a second moment respectively, wherein the first image and the second image each comprise a to-be-detected target and the object, and the second moment is later than the first moment;
measure a first distance between the object and the to-be-detected target in the first image and a second distance between the object and the to-be-detected target in the second image; and
predict, based on the first distance and the second distance, whether the object collides with the to-be-detected target.
9. The computing device according to claim 8 , wherein the programming instructions instruct the processor to:
when the second distance is less than the first distance, obtain a relative velocity between the to-be-detected target and the object based on a difference between the second distance and the first distance and a difference between the second moment and the first moment; and
predict, based on the relative velocity and the second distance, a time in which the object collides with the to-be-detected target.
10. The computing device according to claim 8 , wherein the programming instructions instruct the processor to:
calculate a location difference between the object in the second image and the object in the first image; and
predict, based on the first distance and a sum of the second distance and the location difference, whether the object collides with the to-be-detected target.
11. The computing device according to claim 8 , wherein the programming instructions instruct the processor to:
obtain two-dimensional borders of the to-be-detected target in the first image and the second image; and
measure the first distance between the object and the two-dimensional border of the to-be-detected target in the first image and the second distance between the object and the two-dimensional border of the to-be-detected target in the second image.
12. The computing device according to claim 8 , wherein
the object is a vehicle, and the vehicle and the to-be-detected target are located in a same lane.
13. The computing device according to claim 12 , wherein the programming instructions instruct the processor to:
identify lane lines in the first image and the second image, wherein the lane lines comprise a first lane line and a second lane line, the first lane line and the second lane line are adjacent lane lines, and the vehicle and the to-be-detected target are located between the first lane line and the second lane line; and
calculate a transverse velocity of the to-be-detected target based on a ratio of a distance between the to-be-detected target and the first lane line to a size of the to-be-detected target in the first image and a ratio of a distance between the to-be-detected target and the first lane line to a size of the to-be-detected target in the second image; and
predict, based on the ratio of the distance between the to-be-detected target and the first lane line to the size of the to-be-detected target in the second image and the transverse velocity of the to-be-detected target, a time in which the to-be-detected target leaves the current lane.
14. A non-transitory computer readable medium comprising computer program codes stored thereon, executable by one or more processors for an object collision prediction, the computer program codes including:
instructions for controlling the photographing unit to shoot a first image and a second image at a first moment and a second moment respectively, wherein the first image and the second image each comprise a to-be-detected target and the object, and the second moment is later than the first moment;
instructions for measuring a first distance between the object and the to-be-detected target in the first image and a second distance between the object and the to-be-detected target in the second image; and
instructions for predicting, based on the first distance and the second distance, whether the object collides with the to-be-detected target.
15. The non-transitory computer readable medium according to claim 14 , the computer program codes including:
instructions for obtaining a relative velocity between the to-be-detected target and the object based on a difference between the second distance and the first distance and a difference between the second moment and the first moment when the second distance is less than the first distance; and
instructions for predicting, based on the relative velocity and the second distance, a time in which the object collides with the to-be-detected target.
16. The non-transitory computer readable medium according to claim 14 , the instructions for predicting, based on the first distance and the second distance, whether the object collides with the to-be-detected target comprises:
instructions for calculating a location difference between the object in the second image and the object in the first image; and
instructions for predicting, based on the first distance and a sum of the second distance and the location difference, whether the object collides with the to-be-detected target.
17. The non-transitory computer readable medium according to claim 14 , the instructions for measuring a first distance between the object and the to-be-detected target in the first image and a second distance between the object and the to-be-detected target in the second image comprises:
instructions for obtaining two-dimensional borders of the to-be-detected target in the first image and the second image; and
instructions for measuring the first distance between the object and the two dimensional borders of the to-be-detected target in the first image and the second distance between the object and the two-dimensional border of the to-be-detected target in the second image.
18. The non-transitory computer readable medium according to claim 17 , wherein when a distance between each pixel of a lower edge of the two-dimensional border of the to-be-detected target and the object is different, a shortest distance between a pixel comprised in the lower edge of the two-dimensional border of the to-be-detected target and the object is used as a distance between the object and the two-dimensional border of the to-be-detected target in the image.
19. The non-transitory computer readable medium according to claim 14 , wherein the object is a vehicle, and the vehicle and the to-be-detected target are located in a same lane.
20. The non-transitory computer readable medium according to claim 19 , the computer program codes including:
instructions for identifying lane lines in the first image and the second image, wherein the lane lines comprise a first lane line and a second lane line, the first lane line and the second lane line are adjacent lane lines, and the vehicle and the to-be-detected target are located between the first lane line and the second lane line;
instructions for calculating a transverse velocity of the to-be-detected target based on a ratio of a distance between the to-be-detected target and the first lane line to a size of the to-be-detected target in the first image and a ratio of a distance between the to-be-detected target and the first lane line to a size of the to-be-detected target in the second image; and
instructions for predicting, based on the ratio of the distance between the to-be detected target and the first lane line to the size of the to-be-detected target in the second image and the transverse velocity of the to-be-detected target, a time in which the to-be-detected target leaves the current lane.