Vehicle insurance image processing method, apparatus, server, and system
A vehicle insurance image processing method includes: acquiring a vehicle insurance image; processing the vehicle insurance image by using a preset image classification algorithm, to determine at least one category label of the vehicle insurance image; and storing, based on classification, the vehicle insurance image in a corresponding usage scenario according to the at least one category label.
1. A vehicle insurance image processing method, comprising:
acquiring a vehicle insurance image;
processing the vehicle insurance image by using a preset image classification algorithm, to determine at least one category label of the vehicle insurance image; and
storing, based on classification, the vehicle insurance image in a corresponding usage scenario according to the at least one category label,
wherein the using the preset image classification algorithm comprises:
using a multi-task deep convolutional neural network classification model which has a shared convolution layer.
2. The vehicle insurance image processing method of claim 1 , wherein the storing, based on classification, comprises:
storing classification results of the vehicle insurance image by using a relational database.
3. The vehicle insurance image processing method of claim 2 , further comprising:
detecting the vehicle insurance image by using a selected optical character recognition algorithm, to recognize text information in the vehicle insurance image; and
storing the text information and the vehicle insurance image in an associated manner.
4. The vehicle insurance image processing method of claim 3 , wherein after the text information in the vehicle insurance image is recognized, the method further comprises:
detecting whether key information of a preset type in the text information is complete, and recording detection results of the key information.
5. A vehicle insurance image processing apparatus, comprising:
a processor; and
a memory storing instructions executable by the processor,
wherein, when executing the instructions, the processor performs:
acquiring a vehicle insurance image;
processing the vehicle insurance image by using a preset image classification algorithm, to determine at least one category label of the vehicle insurance image; and
storing, based on classification, the vehicle insurance image in a corresponding usage scenario according to the at least one category label,
wherein the using the preset image classification algorithm comprises:
using a multi-task deep convolutional neural network classification model which has a shared convolution layer.
6. The vehicle insurance image processing apparatus of claim 5 , wherein, when executing the instructions, the processor further performs:
using a relational database to store classification results of the vehicle insurance image.
7. The vehicle insurance image processing apparatus of claim 6 , wherein, when executing the instructions, the processor further performs:
detecting the vehicle insurance image by using a selected optical character recognition algorithm, to recognize text information in the vehicle insurance image; and
storing the text information and the vehicle insurance image in an associated manner.
8. The vehicle insurance image processing apparatus of claim 7 , wherein, when executing the instructions, the processor further performs:
detecting whether key information of a preset type in the text information is complete, and recording detection results of the key information.
9. The vehicle insurance image processing apparatus of claim 5 , being a server.
10. A vehicle insurance image system, comprising:
a collected-image storage unit;
an algorithm server; and
a vehicle insurance image database,
wherein the algorithm server comprises at least one processor and a memory for storing instructions executable by the processor, and when executing the instructions, the at least one processor performs:
acquiring a vehicle insurance image from the collected-image storage unit;
processing the vehicle insurance image by using a preset image classification algorithm, to determine at least one category label of the vehicle insurance image; and
storing the vehicle insurance image, according to the at least one category label, in a storage zone of a corresponding usage scenario in the vehicle insurance image database,
wherein the using the preset image classification algorithm comprises:
using a multi-task deep convolutional neural network classification model which has a shared convolution layer.