Method and apparatus for providing road congestion cause

1 . A system for providing a road congestion cause, the system comprising at least one vehicle and a server, the server comprising at least one processor and at least one memory, wherein the at least one memory stores instructions, and the at least one processor is coupled to the at least one memory and is configured to execute the instructions to:

determine a congestion point in a road section;

determine a plurality of vehicles located in the road section of a first predetermined length centered on the congestion point;

send an indication message to the plurality of vehicles, wherein the indication message indicates to the plurality of vehicles to photograph the congestion point using a vehicle-mounted camera and to upload photograph data to the server;

receive the photograph data from the plurality of vehicles;

determine the road congestion cause based on the photograph data by inputting the photograph data into a first model, wherein the first model is a pre-trained image recognition model, and determining the road congestion cause based on an output of the first model, wherein the output of the first model includes an element identifier for each recognized element in the photograph data from a plurality of predetermined elements for each of a plurality of photographs in the photograph data, wherein each predetermined element corresponds to a training sample image in a training sample set, and wherein each element identifier corresponds to a label value of a training sample image in the training sample set; and

send, to the at least one vehicle, data indicative of the road congestion cause and/or the congestion point for display; and

display the data indicative of the road congestion cause and/or the congestion point on a navigation map displayed on a display screen of the at least one vehicle,

wherein the first model is used to recognize whether the photograph data has at least one of a plurality of predetermined features, the plurality of predetermined features are a set of at least one predetermined feature respectively corresponding to a plurality of predetermined congestion causes, and the at least one processor is configured to execute the instructions to:

determine the road congestion cause based on a congestion cause ranking of the plurality of predetermined congestion causes based on the element identifiers output by the image recognition model for the plurality of photographs in the photograph data,

wherein the congestion cause ranking is based on a percentage or ratio of a) a quantity of elements corresponding to each congestion cause in the output predetermined elements to b) all elements of the congestion cause.

2 . The system according to claim 1 , wherein the at least one processor is configured to execute the instructions to:

indicate to at least one vehicle located in a road section of a second predetermined length centered on the congestion point to photograph the congestion point using a front-facing camera, a side-mounted camera, and a rear-facing camera, wherein the second predetermined length is shorter than the first predetermined length.

3 . The system according to claim 2 , wherein the at least one processor is configured to execute the instructions to:

indicate to at least one vehicle located in front of the congestion point to photograph the congestion point using a rear-facing camera.

4 . The system according to claim 1 , wherein the at least one processor is configured to execute the instructions to:

indicate to at least one vehicle located behind the congestion point to photograph the congestion point using a front-facing camera.

5 . The system according to claim 1 , wherein the road congestion cause is a traffic accident cause, and the at least one processor is configured to execute the instructions to:

input the photograph data into a second model;

perform a traffic accident responsibility identification based on an output of the second model; and

issue the traffic accident responsibility identification.

6 . An apparatus for providing a road congestion cause, comprising at least one processor and at least one memory, wherein the at least one memory stores instructions, and the at least one processor is coupled to the at least one memory and is configured to execute the instructions to:

receive an indication message from a server;

photograph a congestion point in a road section based on the indication message using a vehicle-mounted camera;

send, to the server based on the indication message, photograph data that is of the congestion point and that is obtained through photographing;

receive, from the server, data indicative of a congestion cause and the congestion point for display, the road congestion cause having been determined based on the photograph data by inputting the photograph data into a first model, wherein the first model is a pre-trained image recognition model, and determining the road congestion cause based on an output of the first model, wherein the output of the first model includes an element identifier for each recognized element in the photograph data from a plurality of predetermined elements for each of a plurality of photographs in the photograph data, wherein each predetermined element corresponds to a training sample image in a training sample set, and wherein each element identifier corresponds to a label value of a training sample image in the training sample set; and

display the data indicative of the congestion cause and/or the congestion point on a navigation map displayed on a display screen of the apparatus,

wherein the first model is used to recognize whether the photograph data has at least one of a plurality of predetermined features, the plurality of predetermined features are a set of at least one predetermined feature respectively corresponding to a plurality of predetermined congestion causes, and the at least one processor is configured to execute the instructions to:

determine the road congestion cause based on a congestion cause ranking of the plurality of predetermined congestion causes based on the element identifiers output by the image recognition model for the plurality of photographs in the photograph data,

wherein the congestion cause ranking is based on a percentage of a) a quantity of elements corresponding to each congestion cause in the output predetermined elements to b) all elements of the congestion cause.

7 . A method for providing a road congestion cause, wherein the method is executed by a server and comprises:

determining a congestion point in a road section;

determining a plurality of vehicles located in the road section of a first predetermined length centered on the congestion point;

sending an indication message to the plurality of vehicles, wherein the indication message indicates to the plurality of vehicles to photograph the congestion using a vehicle-mounted camera and to upload photograph data to the server;

receiving the photograph data from the plurality of vehicles; and

determining the road congestion cause based on the photograph data by inputting the photograph data into a first model, wherein the first model is a pre-trained image recognition model, and determining the road congestion cause based on an output of the first model, wherein the output of the first model includes an element identifier for each recognized element in the photograph data from a plurality of predetermined elements for each of a plurality of photographs in the photograph data, wherein each predetermined element corresponds to a training sample image in a training sample set, and wherein each element identifier corresponds to a label value of a training sample image in the training sample set;

sending data indicative of the road congestion cause and/or the congestion point for display to at least one vehicle of the plurality of vehicles; and

displaying the data indicative of the road congestion cause and/or the congestion point on a navigation map displayed on a display screen of the at least one vehicle,

wherein the first model is used to recognize whether the photograph data has at least one of a plurality of predetermined features, the plurality of predetermined features are a set of at least one predetermined feature respectively corresponding to a plurality of predetermined congestion causes, and the method further comprises:

determining the road congestion cause based on a congestion cause ranking of the plurality of predetermined congestion causes based on the element identifiers output by the image recognition model for the plurality of photographs in the photograph data,

wherein the congestion cause ranking is based on a percentage of a) a quantity of elements corresponding to each congestion cause in the output predetermined elements to b) all elements of the congestion cause.

8 . The method according to claim 7 , further comprising:

indicating to at least one vehicle located in a road section of a second predetermined length centered on the congestion point to photograph the congestion point using a front-facing camera, a side-mounted camera, and a rear-facing camera, wherein the second predetermined length is shorter than the first predetermined length.

9 . The method according to claim 7 , further comprising:

indicating to at least one vehicle located in front of the congestion point to photograph the congestion point using a rear-facing camera.

10 . The method according to claim 7 , further comprising:

indicating to at least one vehicle located behind the congestion point to photograph the congestion point using a front-facing camera.

11 . The method according to claim 7 , wherein the road congestion cause is a traffic accident cause, and the method further comprises:

inputting the photograph data into a second model;

performing a traffic accident responsibility identification based on an output of the second model; and

issuing the traffic accident responsibility identification.