IMAGE PROCESSING METHOD AND APPARATUS

1 . An image processing method for use in a mobile device, comprising:

acquiring a video stream of a vehicle by a camera of the mobile device according to a user instruction;

obtaining an image corresponding to a frame in the video stream;

determining whether the image meets a predetermined criterion by inputting the image into a classification model, the classification model comprising a first convolutional neural network for use in the mobile device;

in response to the image meeting the predetermined criterion, adding at least one of a target box or target segmentation information to the image by inputting the image into a target detection and segmentation model, the at least one of the target box or the target segmentation information corresponding to at least one of a vehicle part or vehicle damage of the vehicle, the target detection and segmentation model comprising a second convolutional neural network for use in the mobile device; and

displaying to a user the at least one of the target box or the target segmentation information.

2 . The image processing method of claim 1 , further comprising:

in response to the image not meeting the predetermined criterion, prompting the user based on a classification result from the classification model.

3 . The image processing method of claim 1 , wherein the classification model classifies the image based on at least one of: whether the image is blurred, whether the image includes vehicle damage, whether a light intensity is sufficient, whether a shooting angle is skewed, or whether a shooting distance is appropriate.

4 . The image processing method of claim 1 , further comprising:

presenting a shooting flow to the user before the video stream of the vehicle is acquired by the camera.

5 . The image processing method of claim 1 , further comprising:

prompting the user based on the at least one of the target box or the target segmentation information.

6 . The image processing method of claim 5 , wherein the prompting the user based on the at least one of the target box or the target segmentation information comprises:

prompting the user to move forward or backward based on the at least one of the target box or the target segmentation information.

7 . The image processing method of claim 5 , wherein the prompting the user based on the at least one of the target box or the target segmentation information comprises:

prompting the user to shoot based on the at least one of the target box or the target segmentation information, to obtain a damage assessment photo corresponding to the image of the frame.

8 . The image processing method of claim 7 , further comprising:

uploading the damage assessment photo to a server.

9 . The image processing method of claim 7 , further comprising:

obtaining, based on the video stream, an association between the image and a first image, wherein the first image is an image of a first frame before the frame in the video stream.

10 . The image processing method of claim 9 , wherein the association comprises at least one of: an optical flow, a mapping matrix, or a position and angle transformation relation between the image and the first image.

11 . The image processing method of claim 9 , further comprising:

uploading the association to the server.

12 . The image processing method of claim 1 , wherein the first convolutional neural network and the second convolutional neural network are a same convolutional neural network shared by the classification model and the target detection and segmentation model.

13 . A mobile device, comprising:

a memory storing instructions; and

a processor configured to execute the instructions to:

acquire a video stream of a vehicle by a camera according to a user instruction;

obtain an image corresponding to a frame in the video stream;

determine whether the image meets a predetermined criterion by inputting the image into a classification model, the classification model comprising a first convolutional neural network for use in the mobile device;

if the image meets the predetermined criterion, add at least one of a target box or target segmentation information to the image by inputting the image into a target detection and segmentation model, the at least one of the target box or the target segmentation information corresponding to at least one of a vehicle part or vehicle damage of the vehicle, the target detection and segmentation model comprising a second convolutional neural network for use in the mobile device; and

display to a user the at least one of the target box or the target segmentation information.

14 . The mobile device of claim 13 , wherein the processor is further configured to execute the instructions to:

in response to the image not meeting the predetermined criterion, prompt the user based on a classification result from the classification model.

15 . The mobile device of claim 13 , wherein the classification model classifies the image based on at least one of: whether the image is blurred, whether the image includes vehicle damage, whether a light intensity is sufficient, whether a shooting angle is skewed, or whether a shooting distance is appropriate.

16 . The mobile device of claim 13 , wherein the processor is further configured to execute the instructions to:

present a shooting flow to the user before the video stream of the vehicle is acquired by the camera.

17 . The mobile device of claim 13 , wherein the processor is further configured to execute the instructions to:

prompt the user based on the at least one of the target box or the target segmentation information.

18 . The mobile device of claim 17 , wherein the prompting the user based on the at least one of the target box or the target segmentation information comprises:

prompting the user to move forward or backward based on the at least one of the target box or the target segmentation information.

19 . The mobile device of claim 17 , wherein the prompting the user based on the at least one of the target box or the target segmentation information comprises:

prompting the user to shoot based on the at least one of the target box or the target segmentation information, to obtain a damage assessment photo corresponding to the image of the frame.

20 . The mobile device of claim 19 , wherein the processor is further configured to execute the instructions to:

upload the damage assessment photo to a server.

21 . The mobile device of claim 19 , wherein the processor is further configured to execute the instructions to:

obtain, based on the video stream, an association between the image and a first image, wherein the first image is an image of a first frame before the frame in the video stream.

22 . The mobile device of claim 21 , wherein the association comprises at least one of: an optical flow, a mapping matrix, or a position and angle transformation relation between the image and the first image.

23 . The mobile device of claim 21 , wherein the processor is further configured to execute the instructions to:

upload the association to the server.

24 . The mobile device of claim 13 , wherein the first convolutional neural network and the second convolutional neural network are a same convolutional neural network shared by the classification model and the target detection and segmentation model.

25 . A non-transitory computer-readable medium storing instructions that, when executed by a processor of a mobile device, cause the mobile device to perform an image processing method, the method comprising:

acquiring a video stream of a vehicle by a camera according to a user instruction;

obtaining an image corresponding to a frame in the video stream;

determining whether the image meets a predetermined criterion by inputting the image into a classification model, the classification model comprising a first convolutional neural network;

if the image meets the predetermined criterion, adding at least one of a target box or target segmentation information to the image by inputting the image into a target detection and segmentation model, the at least one of the target box or the target segmentation information corresponding to at least one of a vehicle part or vehicle damage of the vehicle, the target detection and segmentation model comprising a second convolutional neural network; and

displaying to a user the at least one of the target box or the target segmentation information.