Classification identification method based on convolutional neural network and multi-modal fusion

The invention discloses a classification identification method based on a convolutional neural network and multi-modal fusion. According to the implementation scheme, the method comprises the following steps: (1) processing original echoes of a target to obtain distance-Doppler-amplitude data including the target; (2) acquiring a 4 * 16 distance-Doppler image around the target, and serially connecting and fusing Doppler data of different modals to respectively form a one-dimensional data set and a two-dimensional data set; (3) constructing a multi-modal fusion model based on a convolutional neural network, and carrying out classification and recognition by using one-dimensional and two-dimensional convolutional neural networks; (4) testing the network through forward propagation and backward propagation training, and optimizing the network model by using a gradient descent algorithm; and (5) performing decision-level fusion on classification results obtained by the two modules through a natural Bayesian algor