The behavior that the crowded areas tourism is quite sudden and randomicity

key to reflect the characteristics of susceptible to keep out

there is monitoring blind Angle

the traditional approach to monitoring recognition ability

lack of independent behavior is affected by the block

on some suspected behavior lack of accurate identification

puts forward a kind of based on Internet of things crowded areas tourism trampling risk image monitoring method

analyzes the Internet monitoring platform

to extract the crowded areas tourism trampling risk Hu moment invariant features of the image and the input of RBF neural network learning and classification

crowded areas for various types of tourism trampling behavior

will test samples in the input of RBF neural network model is established to test

and the RBF network output corresponding test sample trample risk image is the tourism concentration areas

so as to realize tourism cluster trample image monitoring risk.The simulation results show that the proposed method in crowded areas tourism trampling risk monitor image

accuracy and efficiency than traditional methods have greatly improved.