Traffic light state recognizing method and apparatus, computer device and readable medium

1. A traffic light state recognizing method, wherein the method comprises:

obtaining a first image collected by a long-focus camera, a second image collected by a short-focus camera, and positioning information and travelling direction of the autonomous vehicle collected by a positioning sensor at a target moment; the long-focus camera and short-focus camera are respectively disposed on a roof of the autonomous vehicle;

according to the positioning information and traveling direction of the autonomous vehicle at the target moment, obtaining, from a high-precision map server, location information of traffic lights within a range of preset distance threshold ahead in the traveling direction of the autonomous vehicle at the target moment;

recognizing the state of the traffic lights at the target moment, according to the location information of the traffic lights at the target moment, and one image of the first image and second image, wherein the second image is used for recognizing the state of the traffic lights in a situation where the state of the traffic lights is not recognized using the first image.

2. The method according to claim 1 , wherein the obtaining positioning information of the autonomous vehicle collected by a positioning sensor at a target moment specifically comprises:

if it is detected that the positioning sensor does not collect the positioning information of the autonomous vehicle at the target moment, obtaining the positioning information of the autonomous vehicle collected by the positioning sensor at a previous moment and a later moment of the target moment respectively;

calculating the positioning information of the autonomous vehicle at the target moment in an interpolation manner according to the positioning information of the autonomous vehicle at the previous moment and the later moment of the target moment.

3. The method according to claim 1 , wherein the step of, according to the positioning information and traveling direction of the autonomous vehicle at the target moment, obtaining, from a high-precision map server, location information of traffic lights within a range of preset distance threshold ahead in the traveling direction of the autonomous vehicle at the target moment specifically comprises:

sending to the high-precision map server a traffic light location request carrying the positioning information and travelling direction of the autonomous vehicle at the target moment, so that the high-precision map server obtains location information of traffic lights within a range of preset distance threshold away from the autonomous vehicle ahead in the traveling direction of the autonomous vehicle, according to the positioning information and traveling direction of the autonomous vehicle in the traffic light location request;

receiving the location information of traffic lights within the range of preset distance threshold ahead in the traveling direction of the autonomous vehicle sent by the high-precision map server.

4. The method according to claim 1 , wherein the recognizing the state of traffic lights at the target moment, according to the location information of the traffic lights at the target moment, and one image of the first image and second image specifically comprises:

obtaining first location information of the traffic lights at the target moment in the first image according to the location information of the traffic light at the target moment;

judging whether the first location information is valid;

if the first location information is valid, obtaining from the first image a first region of interest for recognizing the state of the traffic lights according to the first location information;

performing detection for the first region of interest to recognize the state of the traffic lights at the target moment.

5. The method according to claim 4 , wherein if the first location information is invalid, the method further comprises:

obtaining second location information of the traffic lights at the target moment in the second image according to the location information of the traffic lights at the target moment;

obtaining from the second image a second region of interest for recognizing the state of the traffic lights according to the second location information;

performing detection for the second region of interest to recognize the state of the traffic lights at the target moment.

6. The method according to claim 5 , wherein the obtaining first location information of the traffic lights at the target moment in the first image according to the location information of the traffic light at the target moment specifically comprises:

according to the location information of the traffic lights at the target moment, and in conjunction with an internal parameter of the long-focus camera and an external rotation transformation expression of the long-focus camera relative to the positioning sensor, obtaining the first location information of the traffic lights in the first image using the following equation:I long =K l ×[R|T] l ×P w , wherein I long is the first location information of the traffic lights in the first image, K l is the internal parameter of the long-focus camera, [R|T] l is the external rotation transformation expression of the long-focus camera relative to the positioning sensor, and P W is the location information of the traffic lights at the target moment;

the obtaining second location information of the traffic lights at the target moment in the second image according to the location information of the traffic lights at the target moment specifically comprises:

according to the location information of the traffic lights at the target moment, and in conjunction with an internal parameter of the short-focus camera and an external rotation transformation expression of the short-focus camera relative to the positioning sensor, obtaining the second location information of the traffic lights in the second image using the following equation: I snort =K s ×[R|T] S ×P W , wherein I short is the second location information of the traffic lights in the second image, K s is the internal parameter of the short-focus camera, [R|T] s is the external rotation transformation expression of the short-focus camera relative to the positioning sensor, and P W is the location information of the traffic lights at the target moment.

7. A computer device for an autonomous vehicle, wherein the device comprises:

one or more processors;

a storage for storing one or more programs;

the one or more programs, when executed by said one or more processors, enable said one or more processors to implement a traffic light state recognizing method, wherein the method comprises:

obtaining a first image collected by a long-focus camera, a second image collected by a short-focus camera, and positioning information and travelling direction of the autonomous vehicle collected by a positioning sensor at a target moment; the long-focus camera and short-focus camera are respectively disposed on a roof of the autonomous vehicle;

according to the positioning information and traveling direction of the autonomous vehicle at the target moment, obtaining, from a high-precision map server, location information of traffic lights within a range of preset distance threshold ahead in the traveling direction of the autonomous vehicle at the target moment;

recognizing the state of the traffic lights at the target moment, according to the location information of the traffic lights at the target moment, and one image of the first image and second image, wherein the second image is used for recognizing the state of the traffic lights in a situation where the state of the traffic lights is not recognized using the first image.

8. The computer device according to claim 7 , wherein the obtaining positioning information of the autonomous vehicle collected by a positioning sensor at a target moment specifically comprises:

if it is detected that the positioning sensor does not collect the positioning information of the autonomous vehicle at the target moment, obtaining the positioning information of the autonomous vehicle collected by the positioning sensor at a previous moment and a later moment of the target moment respectively;

calculating the positioning information of the autonomous vehicle at the target moment in an interpolation manner according to the positioning information of the autonomous vehicle at the previous moment and the later moment of the target moment.

9. The computer device according to claim 7 , wherein the step of, according to the positioning information and traveling direction of the autonomous vehicle at the target moment, obtaining, from a high-precision map server, location information of traffic lights within a range of preset distance threshold ahead in the traveling direction of the autonomous vehicle at the target moment specifically comprises:

sending to the high-precision map server a traffic light location request carrying the positioning information and travelling direction of the autonomous vehicle at the target moment, so that the high-precision map server obtains location information of traffic lights within a range of preset distance threshold away from the autonomous vehicle ahead in the traveling direction of the autonomous vehicle, according to the positioning information and traveling direction of the autonomous vehicle in the traffic light location request;

receiving the location information of traffic lights within the range of preset distance threshold ahead in the traveling direction of the autonomous vehicle sent by the high-precision map server.

10. The computer device according to claim 7 , wherein the recognizing the state of traffic lights at the target moment, according to the location information of the traffic lights at the target moment, and one image of the first image and second image specifically comprises:

obtaining first location information of the traffic lights at the target moment in the first image according to the location information of the traffic light at the target moment;

judging whether the first location information is valid;

if the first location information is valid, obtaining from the first image a first region of interest for recognizing the state of the traffic lights according to the first location information;

performing detection for the first region of interest to recognize the state of the traffic lights at the target moment.

11. The computer device according to claim 10 , wherein if the first location information is invalid, the method further comprises:

obtaining second location information of the traffic lights at the target moment in the second image according to the location information of the traffic lights at the target moment;

obtaining from the second image a second region of interest for recognizing the state of the traffic lights according to the second location information;

performing detection for the second region of interest to recognize the state of the traffic lights at the target moment.

12. The computer device according to claim 11 , wherein the obtaining first location information of the traffic lights at the target moment in the first image according to the location information of the traffic light at the target moment specifically comprises:

according to the location information of the traffic lights at the target moment, and in conjunction with an internal parameter of the long-focus camera and an external rotation transformation expression of the long-focus camera relative to the positioning sensor, obtaining the first location information of the traffic lights in the first image using the following equation:I long =K l ×[R|T] l ×P w , wherein I long is the first location information of the traffic lights in the first image, K l is the internal parameter of the long-focus camera, [R|T] l is the external rotation transformation expression of the long-focus camera relative to the positioning sensor, and P W is the location information of the traffic lights at the target moment;

the obtaining second location information of the traffic lights at the target moment in the second image according to the location information of the traffic lights at the target moment specifically comprises:

according to the location information of the traffic lights at the target moment, and in conjunction with an internal parameter of the short-focus camera and an external rotation transformation expression of the short-focus camera relative to the positioning sensor, obtaining the second location information of the traffic lights in the second image using the following equation:I short =K s ×[R|T] S ×P W , wherein I short is the second location information of the traffic lights in the second image, K s is the internal parameter of the short-focus camera, [R|T] s is the external rotation transformation expression of the short-focus camera relative to the positioning sensor, and P W is the location information of the traffic lights at the target moment.

13. A non-transitory computer readable medium on which a computer program is stored, wherein the program, when executed by a processor, implements a traffic light state recognizing method, wherein the method comprises:

obtaining a first image collected by a long-focus camera, a second image collected by a short-focus camera, and positioning information and travelling direction of the autonomous vehicle collected by a positioning sensor at a target moment; the long-focus camera and short-focus camera are respectively disposed on a roof of the autonomous vehicle;

according to the positioning information and traveling direction of the autonomous vehicle at the target moment, obtaining, from a high-precision map server, location information of traffic lights within a range of preset distance threshold ahead in the traveling direction of the autonomous vehicle at the target moment;

recognizing the state of the traffic lights at the target moment, according to the location information of the traffic lights at the target moment, and one image of the first image and second image, wherein the second image is used for recognizing the state of the traffic lights in a situation where the state of the traffic lights is not recognized using the first image.

14. The non-transitory computer readable medium according to claim 13 , wherein the obtaining positioning information of the autonomous vehicle collected by a positioning sensor at a target moment specifically comprises:

if it is detected that the positioning sensor does not collect the positioning information of the autonomous vehicle at the target moment, obtaining the positioning information of the autonomous vehicle collected by the positioning sensor at a previous moment and a later moment of the target moment respectively;

calculating the positioning information of the autonomous vehicle at the target moment in an interpolation manner according to the positioning information of the autonomous vehicle at the previous moment and the later moment of the target moment.

15. The non-transitory computer readable medium according to claim 13 , wherein the step of, according to the positioning information and traveling direction of the autonomous vehicle at the target moment, obtaining, from a high-precision map server, location information of traffic lights within a range of preset distance threshold ahead in the traveling direction of the autonomous vehicle at the target moment specifically comprises:

sending to the high-precision map server a traffic light location request carrying the positioning information and travelling direction of the autonomous vehicle at the target moment, so that the high-precision map server obtains location information of traffic lights within a range of preset distance threshold away from the autonomous vehicle ahead in the traveling direction of the autonomous vehicle, according to the positioning information and traveling direction of the autonomous vehicle in the traffic light location request;

receiving the location information of traffic lights within the range of preset distance threshold ahead in the traveling direction of the autonomous vehicle sent by the high-precision map server.

16. The non-transitory computer readable medium according to claim 13 , wherein the recognizing the state of traffic lights at the target moment, according to the location information of the traffic lights at the target moment, and one image of the first image and second image specifically comprises:

obtaining first location information of the traffic lights at the target moment in the first image according to the location information of the traffic light at the target moment;

judging whether the first location information is valid;

if the first location information is valid, obtaining from the first image a first region of interest for recognizing the state of the traffic lights according to the first location information;

performing detection for the first region of interest to recognize the state of the traffic lights at the target moment.

17. The non-transitory computer readable medium according to claim 16 , wherein if the first location information is invalid, the method further comprises:

obtaining second location information of the traffic lights at the target moment in the second image according to the location information of the traffic lights at the target moment;

obtaining from the second image a second region of interest for recognizing the state of the traffic lights according to the second location information;

performing detection for the second region of interest to recognize the state of the traffic lights at the target moment.

18. The non-transitory computer readable medium according to claim 17 , wherein the obtaining first location information of the traffic lights at the target moment in the first image according to the location information of the traffic light at the target moment specifically comprises:

according to the location information of the traffic lights at the target moment, and in conjunction with an internal parameter of the long-focus camera and an external rotation transformation expression of the long-focus camera relative to the positioning sensor, obtaining the first location information of the traffic lights in the first image using the following equation:I long =K l ×[R|T] l ×P W , wherein I long is the first location information of the traffic lights in the first image, K l is the internal parameter of the long-focus camera, [R|T] l is the external rotation transformation expression of the long-focus camera relative to the positioning sensor, and P W is the location information of the traffic lights at the target moment;

the obtaining second location information of the traffic lights at the target moment in the second image according to the location information of the traffic lights at the target moment specifically comprises:

according to the location information of the traffic lights at the target moment, and in conjunction with an internal parameter of the short-focus camera and an external rotation transformation expression of the short-focus camera relative to the positioning sensor, obtaining the second location information of the traffic lights in the second image using the following equation:I short =K s ×[R|T] S ×P w , wherein I short is the second location information of the traffic lights in the second image, K s is the internal parameter of the short-focus camera, [R|T] s is the external rotation transformation expression of the short-focus camera relative to the positioning sensor, and P W is the location information of the traffic lights at the target moment.