Vehicle re-identification method based on few sample attention

[0047] In order to make the technical means, creative features, goals and effects achieved by the present invention easy to understand, the present invention will be further described below in conjunction with specific embodiments.

[0048] like Figure 1-Figure 2 As shown, the vehicle re-identification method based on few-shot attention consists of feature extraction module, few-shot attention module and re-identification module. Its network flow chart is as follows figure 1 As shown, its feature is to extract the characteristics of the input vehicle, generate the vehicle attention map through FSAM, and then compare it with the target domain feature map optimized by FIM, which reduces the dependence on a large amount of data and makes the re-identification network more efficient. The ability to transform is stronger; the specific steps are as follows:

[0049] 1. Network construction

[0050] Step 1: The feature extraction network of this patent mainly adopts convolutional ne