The present application relates to the technical field of data processing, in particular to a street view recognition model training methodview, comprising the following steps: obtaining a pair of training samples used in the current round of training; inputting each of the acquired at least one pair of training samples into an entry corresponding to a branch of a neural network having two branches, and determining a first feature and a second feature using the neural network as the input pair of training samples, respectively; matching a comparison structure obtained by comparing the first feature and the second feature with the labeling value of the corresponding training sample pair; training the neural network in this round according to the matching result; after multiple rounds of training of the neural network, obtaining a street view recognition model. By using street view recognition model training method provided by the application scheme, a street view recognition model withhigh recognition accuracy can be obtained. The present application also provides a training system of a street view recognition model, a street view recognition method, a system and an electronic device.