Method and apparatus for detecting braking behavior of front vehicle of autonomous vehicle

1. A method for detecting a braking behavior of a front vehicle of an autonomous vehicle, an image acquisition device being mounted on the autonomous vehicle, the method comprising:

extracting a vehicle image from a vehicle area in an image captured by the image acquisition device mounted on the autonomous vehicle;

converting a color space of the vehicle image to generate a first vehicle image;

setting a pixel value of a pixel point in the first vehicle image meeting any condition in a preset condition group to a first preset value to generate a second vehicle image;

analyzing the second vehicle image to determine a candidate vehicle light area group; and

detecting, based on the candidate vehicle light area group, whether a vehicle indicated by the vehicle image is braking, and generating a detection result.

2. The method according to claim 1 , wherein the converting a color space of the vehicle image comprises:

converting the color space of the vehicle image into a Hue-Saturation-Value color space.

3. The method according to claim 2 , wherein the pixel value of the pixel point comprises a hue, a saturation, and a value, and the preset condition group comprises at least one of: the hue is not within a preset interval, the saturation is less than a first threshold, or the value is less than a second threshold.

4. The method according to claim 1 , wherein the analyzing the second vehicle image to determine a candidate vehicle light area group comprises:

analyzing a connected area of the second vehicle image to obtain the candidate vehicle light area group.

5. The method according to claim 1 , wherein the detecting, based on the candidate vehicle light area group, whether a vehicle indicated by the vehicle image is braking comprises:

for each candidate vehicle light area in the candidate vehicle light area group, determining a coordinate of a center point of the candidate vehicle light area and a size of the candidate vehicle light area on the basis of the coordinate of the pixel point included in the candidate vehicle light area, and classifying the candidate vehicle light area into a vehicle light area set if the candidate vehicle light area meets a first preset condition, wherein the coordinate of the pixel point includes an X coordinate value and a Y coordinate value, and the first preset condition comprises that the Y coordinate value of the center point is not lower than a product of a height value of the vehicle image and a second preset value, and a ratio of the size of the candidate vehicle light area to the size of the vehicle image is greater than a third threshold;

determining whether a first target vehicle light area and a second target vehicle light area exist in the vehicle light area set, wherein both the ratio of the size of the first target vehicle light area to the size of the vehicle image and the ratio of the size of the second target vehicle light area to the size of the vehicle image are greater than a fourth threshold, an absolute value of a difference between the Y coordinate value of the center point of the first target vehicle light area and the Y coordinate value of the center point of the second target vehicle light area is less than a fifth threshold, and the X coordinate value of the center point of the first target vehicle light area and the X coordinate value of the center point of the second target vehicle light area meet one, but a different one of: not greater than a product of a width value of the vehicle image and a third preset value, or not less than a product of the width value and a fourth preset value; and

determining that the vehicle is not braking if the first target vehicle light area and the second target vehicle light area do not exist in the vehicle light area set.

6. The method according to claim 5 , wherein the detecting, based on the candidate vehicle light area group, whether a vehicle indicated by the vehicle image is braking further comprises:

determining a midpoint of a line connecting the center point of the first target vehicle light area and the center point of the second target vehicle light area if the first target vehicle light area and the second target vehicle light area exist in the vehicle light area set; and

searching for a third target vehicle light area vertically upward from the midpoint in the second vehicle image, and determining that the vehicle is braking if the third target vehicle light area is found.

7. An apparatus for detecting a braking behavior of a front vehicle of an autonomous vehicle, an image acquisition device being mounted on the autonomous vehicle, the apparatus comprising:

at least one processor; and

a memory storing instructions, the instructions when executed by the at least one processor, cause the at least one processor to perform operations, the operations comprising:

extracting a vehicle image from a vehicle area in an image captured by the image acquisition device mounted on the autonomous vehicle;

converting a color space of the vehicle image to generate a first vehicle image;

setting a pixel value of a pixel point in the first vehicle image meeting any condition in a preset condition group to a first preset value to generate a second vehicle image;

analyzing the second vehicle image to determine a candidate vehicle light area group; and

detecting, based on the candidate vehicle light area group, whether a vehicle indicated by the vehicle image is braking, and generating a detection result.

8. The apparatus according to claim 7 , wherein the converting a color space of the vehicle image comprises:

converting the color space of the vehicle image into a Hue-Saturation-Value color space.

9. The apparatus according to claim 8 , wherein the pixel value of the pixel point comprises a hue, a saturation, and a value, and the preset condition group comprises at least one of: the hue is not within a preset interval, the saturation is less than a first threshold, or the value is less than a second threshold.

10. The apparatus according to claim 7 , wherein the analyzing the second vehicle image to determine a candidate vehicle light area group comprises:

analyzing a connected area of the second vehicle image to obtain the candidate vehicle light area group.

11. The apparatus according to claim 7 , wherein the detecting, based on the candidate vehicle light area group, whether a vehicle indicated by the vehicle image is braking comprises:

for each candidate vehicle light area in the candidate vehicle light area group, determining a coordinate of a center point of the candidate vehicle light area and a size of the candidate vehicle light area on the basis of the coordinate of the pixel point included in the candidate vehicle light area, and classifying the candidate vehicle light area into a vehicle light area set if the candidate vehicle light area meets a first preset condition, wherein the coordinate of the pixel point includes an X coordinate value and a Y coordinate value, and the first preset condition comprises that the Y coordinate value of the center point is not lower than a product of a height value of the vehicle image and a second preset value, and a ratio of the size of the candidate vehicle light area to the size of the vehicle image is greater than a third threshold;

determining whether a first target vehicle light area and a second target vehicle light area exist in the vehicle light area set, wherein both the ratio of the size of the first target vehicle light area to the size of the vehicle image and the ratio of the size of the second target vehicle light area to the size of the vehicle image are greater than a fourth threshold, an absolute value of a difference between the Y coordinate value of the center point of the first target vehicle light area and the Y coordinate value of the center point of the second target vehicle light area is less than a fifth threshold, and the X coordinate value of the center point of the first target vehicle light area and the X coordinate value of the center point of the second target vehicle light area meet one, but a different one of: not greater than a product of a width value of the vehicle image and a third preset value, or not less than a product of the width value and a fourth preset value; and

determining that the vehicle is not braking if the first target vehicle light area and the second target vehicle light area do not exist in the vehicle light area set.

12. The apparatus according to claim 11 , wherein the detecting, based on the candidate vehicle light area group, whether a vehicle indicated by the vehicle image is braking further comprises:

determining a midpoint of a line connecting the center point of the first target vehicle light area and the center point of the second target vehicle light area if the first target vehicle light area and the second target vehicle light area exist in the vehicle light area set; and

searching for a third target vehicle light area vertically upward from the midpoint in the second vehicle image, and determining that the vehicle is braking if the third target vehicle light area is found.

13. A non-transitory computer-readable storage medium storing a computer program, the computer program when executed by one or more processors, causes the one or more processors to perform operations, the operations comprising:

extracting a vehicle image from a vehicle area in an image captured by the image acquisition device mounted on the autonomous vehicle;

converting a color space of the vehicle image to generate a first vehicle image;

setting a pixel value of a pixel point in the first vehicle image meeting any condition in a preset condition group to a first preset value to generate a second vehicle image;

analyzing the second vehicle image to determine a candidate vehicle light area group; and

detecting, based on the candidate vehicle light area group, whether a vehicle indicated by the vehicle image is braking, and generating a detection result.