Urban flyer recognition method based on micro deep neural network search

The invention relates to an urban flyer recognition method based on micro deep neural network search, belongs to the technical field of intelligent recognition, and solves the problem of low recognition accuracy of small target flyers in the prior art. The method comprises the following steps: cleaning an image containing target flyers to obtain a plurality of groups of data and labels corresponding to different target flyers so as to obtain a data set; based on the data set, searching and training by using a micro deep neural network structure to obtain a corresponding micro deep neural network; introducing an expansion convolution feature pyramid into the obtained differentiable deep neural network to obtain an optimized differentiable deep neural network; and identifying the urban flyer by using the optimized micro deep neural network, and labeling a corresponding label for the urban flyer. The small target flyer can be easily separated from the complex environment background, and the recognition accuracy