Method and device for vehicle identification

1. A vehicle identification method, comprising:

obtaining a first vehicle image and a second vehicle image, wherein the first vehicle image comprises a first vehicle, and the second vehicle image comprises a second vehicle;

extracting a first color feature and a first vehicle type feature of the first vehicle, and extracting a second color feature and a second vehicle type feature of the second vehicle based on a convolutional neural network model;

combining the first color feature and the first vehicle type feature into a first vehicle feature, and combining the second color feature and the second vehicle type feature into a second vehicle feature;

calculating a similarity parameter between the first vehicle feature and the second vehicle feature; and

determining whether the first vehicle is the same one as the second vehicle according to the similarity parameter;

wherein the step of extracting a first color feature and a first vehicle type feature of the first vehicle, and a second color feature and a second vehicle type feature of the second vehicle based on a convolutional neural network model, comprises:

pre-training a first convolutional neural network model through a preset image database;

training the first convolution neural network model through vehicle images of a plurality of colors;

inputting the first vehicle image and the second vehicle image into the first convolution neural network model respectively, wherein the first convolution neural network model comprises a multi-layer convolution layer and a multi-layer full-connection layer, wherein an output of a previous layer is an input of a current layer; and

extracting the first color feature from the first vehicle image and extracting the second color feature from the second vehicle image, after the processing through the convolution layer and the full-connection layer.

2. The method according to claim 1 , wherein the step of extracting a first color feature and a first vehicle type feature of the first vehicle and a second color feature and a second vehicle type feature of the second vehicle based on a convolutional neural network model further comprises:

pre-training a second convolutional neural network model through a preset image database;

training the second convolution neural network model through vehicle images of a plurality of visual angles;

inputting the first vehicle image and the second vehicle image into the second convolution neural network model respectively, wherein the second convolution neural network model comprises a multi-layer convolution layer and a multi-layer full-connection layer, and an output of a previous layer is an input of a current layer; and

extracting the first vehicle type feature from the first vehicle image, and extracting the second vehicle type feature from the second vehicle image after the processing through the convolution layer and the full-connection layer.

3. The method according to claim 1 , wherein the step of combining the first color feature and the first vehicle type feature into a first vehicle feature and combining the second color feature and the second vehicle type feature into a second vehicle feature comprises:

obtaining a color weight of the first color feature and the second color feature, and obtaining a vehicle type weight of the first vehicle type feature and the second vehicle type feature, wherein the vehicle type weight is larger than the color weight;

summating a multiplication of the first color feature and the color weight with a multiplication of the first vehicle type feature and the vehicle type weight to obtain the first vehicle feature; and

summating a multiplication of the second color feature and the color weight with a multiplication of the second vehicle type feature and the vehicle type weight to obtain the second vehicle feature.

4. The method according to claim 1 , wherein the step of determining whether the first vehicle is the same one as the second vehicle according to the similarity parameter comprises:

obtaining a preset threshold value;

determining that the first vehicle is the same one as the second vehicle when the similarity parameter is larger than or equal to the preset threshold value; or

determining that the first vehicle is not the same one as the second vehicle when the similarity parameter is smaller than the preset threshold value.

5. A terminal device comprising a memory, a processor, and a computer readable instruction stored in the memory and executable by the processor, wherein when executing the computer readable instruction, the processor implements following steps of:

obtaining a first vehicle image and a second vehicle image, wherein the first vehicle image comprises a first vehicle, and the second vehicle image comprises a second vehicle;

extracting a first color feature and a first vehicle type feature of the first vehicle, and extracting a second color feature and a second vehicle type feature of the second vehicle based on a convolutional neural network model;

combining the first color feature and the first vehicle type feature into a first vehicle feature, and combining the second color feature and the second vehicle type feature into a second vehicle feature;

calculating a similarity parameter between the first vehicle feature and the second vehicle feature; and

determining whether the first vehicle is the same one as the second vehicle according to the similarity parameter;

wherein the step of extracting a first color feature and a first vehicle type feature of the first vehicle and a second color feature and a second vehicle type feature of the second vehicle based on a convolutional neural network model comprises:

pre-training a first convolutional neural network model through a preset image database;

training the first convolution neural network model through vehicle images of a plurality of colors;

inputting the first vehicle image and the second vehicle image into the first convolution neural network model respectively, wherein the first convolution neural network model comprises a multi-layer convolution layer and a multi-layer full-connection layer, and an output of a previous layer is an input of a current layer; and

extracting the first color feature from the first vehicle image, and extracting the second color feature from the second vehicle image after the processing through the convolution layer and the full-connection layer.

6. The terminal device according to claim 5 , wherein the step of extracting a first color feature and a first vehicle type feature of the first vehicle and a second color feature and a second vehicle type feature of the second vehicle based on a convolutional neural network model further comprises:

pre-training a second convolutional neural network model through a preset image database;

training the second convolution neural network model through vehicle images of a plurality of visual angles;

inputting the first vehicle image and the second vehicle image into the second convolution neural network model respectively, wherein the second convolution neural network model comprises a multi-layer convolution layer and a multi-layer full-connection layer, and an output of a previous layer is an input of a current layer; and

extracting the first vehicle type feature from the first vehicle image, and extracting the second vehicle type feature from the second vehicle image after the processing through the convolution layer and the full-connection layer.

7. The terminal device according to claim 5 , wherein the step of combining the first color feature and the first vehicle type feature into a first vehicle feature and combining the second color feature and the second vehicle type feature into a second vehicle feature comprises:

obtaining a color weight of the first color feature and the second color feature, and obtaining a vehicle type weight of the first vehicle type feature and the second vehicle type feature, wherein the vehicle type weight is larger than the color weight;

summating a multiplication of the first color feature and the color weight with a multiplication of the first vehicle type feature and the vehicle type weight to obtain the first vehicle feature; and

summating a multiplication of the second color feature and the color weight with a multiplication of the second vehicle type feature and the vehicle type weight to obtain the second vehicle feature.

8. The terminal device according to claim 5 , wherein the step of determining whether the first vehicle is the same one as the second vehicle according to the similarity parameter comprises:

obtaining a preset threshold value;

determining that the first vehicle is the same one as the second vehicle when the similarity parameter is larger than or equal to the preset threshold value; or

determining that the first vehicle is not the same one as the second vehicle when the similarity parameter is smaller than the preset threshold value.