Vision-based detection and classification of traffic lights
The present disclosure is directed to an autonomous vehicle having a vehicle control system. The vehicle control system includes an image processing system. The image processing system receives an image that includes a plurality of image portions. The image processing system also calculates a score for each image portion. The score indicates a level of confidence that a given image portion represents an illuminated component of a traffic light. The image processing system further identifies one or more candidate portions from among the plurality of image portions. Additionally, the image processing system determines that a particular candidate portion represents an illuminated component of a traffic light using a classifier. Further, the image processing system provides instructions to control the autonomous vehicle based on the particular candidate portion representing an illuminated component of a traffic light.
1 . A computer system, comprising:
at least one processor; and
data storage, wherein the data storage contains instructions executable by the at least one processor to perform functions comprising:
receiving an image captured by an image-capture device coupled to a vehicle, wherein the image is representative of an area in a direction of travel of the vehicle;
determining a plurality of candidate traffic lights in the image, wherein determining the plurality of candidate traffic lights in the image comprises excluding one or more objects in the image based on one or more object characteristics;
determining a speed at which the vehicle is travelling; and
providing vehicle control instructions to alter at least one of the determined speed at which the vehicle is travelling or a direction in which the vehicle is travelling, wherein the vehicle control instructions are based on the determined speed at which the vehicle is travelling and at least one of the plurality of candidate traffic lights in the image.
2 . The computer system of claim 1 , wherein the functions further comprise:
determining a nearest traffic light from among the plurality of candidate traffic lights.
3 . The computer system of claim 2 , wherein determining the nearest traffic light from among the plurality of candidate traffic lights comprises determining which of the plurality of candidate traffic lights appears largest in the image.
4 . The computer system of claim 3 , wherein the vehicle control instructions are based on the nearest traffic light.
5 . The computer system of claim 4 , wherein the vehicle control instructions are based on a state of the nearest traffic light, wherein the state is represented by at least a color.
6 . The computer system of claim 1 , wherein the area in the direction of travel of the vehicle comprises an intersection.
7 . The computer system of claim 6 , wherein the vehicle control instructions are based on a state of the at least one of the plurality of candidate traffic lights in the image signaling traffic to stop, wherein the functions further comprise:
determining, based on the determined speed at which the vehicle is travelling, a distance from the intersection at which to begin decreasing the determined speed of the vehicle.
8 . A method, comprising:
receiving an image captured by an image-capture device coupled to a vehicle, wherein the image is representative of an area in a direction of travel of the vehicle;
determining a plurality of candidate traffic lights in the image, wherein determining the plurality of candidate traffic lights in the image comprises excluding one or more objects in the image based on one or more object characteristics;
determining a speed at which the vehicle is travelling; and
providing vehicle control instructions to alter at least one of the determined speed at which the vehicle is travelling or a direction in which the vehicle is travelling, wherein the vehicle control instructions are based on the determined speed at which the vehicle is travelling and at least one of the plurality of candidate traffic lights in the image.
9 . The method of claim 8 , further comprising:
determining a nearest traffic light from among the plurality of candidate traffic lights.
10 . The method of claim 9 , wherein determining the nearest traffic light from among the plurality of candidate traffic lights comprises determining which of the plurality of candidate traffic lights appears largest in the image.
11 . The method of claim 10 , wherein the vehicle control instructions are based on the nearest traffic light.
12 . The method of claim 11 , wherein the vehicle control instructions are based on a state of the nearest traffic light, wherein the state is represented by at least a color.
13 . The method of claim 8 , wherein the area in the direction of travel of the vehicle comprises an intersection.
14 . The method of claim 13 , wherein the vehicle control instructions are based on a state of the at least one of the plurality of candidate traffic lights in the image signaling traffic to stop, further comprising:
determining, based on the determined speed at which the vehicle is travelling, a distance from the intersection at which to begin decreasing the determined speed of the vehicle.
15 . A non-transitory computer-readable medium having stored thereon instructions that, upon execution by at least one processor of a computing device, cause the computing device to perform functions comprising:
receiving an image captured by an image-capture device coupled to a vehicle, wherein the image is representative of an area in a direction of travel of the vehicle;
determining a plurality of candidate traffic lights in the image, wherein determining the plurality of candidate traffic lights in the image comprises excluding one or more objects in the image based on one or more object characteristics;
determining a speed at which the vehicle is travelling; and
providing vehicle control instructions to alter at least one of the determined speed at which the vehicle is travelling or a direction in which the vehicle is travelling, wherein the vehicle control instructions are based on the determined speed at which the vehicle is travelling and at least one of the plurality of candidate traffic lights in the image.
16 . The non-transitory computer-readable medium of claim 15 , wherein the functions further comprise:
determining a nearest traffic light from among the plurality of candidate traffic lights.
17 . The non-transitory computer-readable medium of claim 16 , wherein determining the nearest traffic light from among the plurality of candidate traffic lights comprises determining which of the plurality of candidate traffic lights appears largest in the image.
18 . The non-transitory computer-readable medium of claim 17 , wherein the vehicle control instructions are based on the nearest traffic light.
19 . The non-transitory computer-readable medium of claim 18 , wherein the vehicle control instructions are based on a state of the nearest traffic light, wherein the state is represented by at least a color.
20 . The non-transitory computer-readable medium of claim 15 , wherein the vehicle control instructions are based on a state of the at least one of the plurality of candidate traffic lights in the image signaling traffic to stop, wherein the area in the direction of travel of the vehicle comprises an intersection, and wherein the functions further comprise:
determining, based on the determined speed at which the vehicle is travelling, a distance from the intersection at which to begin decreasing the determined speed of the vehicle.