Vehicle color classification model training method, device and vehicle color recognition method
A technology of color classification and model training, applied in the field of image recognition, can solve problems such as difficult color detection areas, difficult recognition by human eyes, vehicle color is easily affected by scene factors such as light and dust, and achieves strong adaptability and high color recognition Accurate, Adaptable and Robust Effects
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0043] see figure 1 , figure 1 It is a flow chart of a vehicle color classification model training method provided by an embodiment of the present invention; including:
[0044] S11. Input several vehicle images into the pre-trained deep neural network to extract the vehicle color feature set;
[0045] S12. Using a clustering algorithm to perform a clustering operation on the vehicle color feature set to obtain several subsets under different scene categories;
[0046] S13. Input the vehicle images under the scene category corresponding to the subset into the corresponding vehicle color classification sub-model, so as to train the vehicle color classification sub-model;
[0047] S14. Link the output layers of all the vehicle color classification sub-models to the input layer of the vehicle color classification model to train the vehicle color classification model; wherein, both the vehicle color classification sub-model and the vehicle color classification model are Designed ...
Embodiment 2
[0060] see Figure 4 , Figure 4 It is a schematic structural diagram of a vehicle color classification model training device provided by an embodiment of the present invention; including:
[0061] The vehicle color feature set extraction module 11 is used to input several vehicle images into the pre-trained deep neural network to extract the vehicle color feature set;
[0062] The clustering operation module 12 is used to perform a clustering operation on the vehicle color feature set by using a clustering algorithm to obtain several subsets under different scene categories;
[0063] Vehicle color classification sub-model training module 13, for inputting the vehicle image under the scene category corresponding to the subset into the corresponding vehicle color classification sub-model, to train the vehicle color classification sub-model;
[0064] Vehicle color classification model training module 14, is used for linking the output layer of all described vehicle color class...
Embodiment 3
[0077] see Figure 5 , Figure 5 It is a flow chart of a vehicle color recognition method provided by an embodiment of the present invention; including:
[0078] S21. Obtain the image to be recognized of the vehicle;
[0079] S22. Input the image to be recognized into the pre-trained vehicle color classification model; wherein, the training method of the vehicle color classification model is the vehicle color classification model training method as described in the first embodiment above;
[0080] S23. Obtain output data of the vehicle color classification model as a result of the vehicle color recognition.
[0081] The vehicle color recognition method described in the embodiment of the present invention adopts the vehicle color classification model training model trained in the vehicle color classification model training method described in Embodiment 1, which can cope with changes in vehicle body color in different scenarios and improve the accuracy of vehicle color recogn...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More