Training method and training device of convolutional neural network model

Abstract

The invention discloses a training method and a training device of a convolutional neural network (CNN) model, and belongs to the field of image recognition. The training method comprises the steps of respectively carrying out a convolution operation, a maximum pooling operation and a horizontal pooling operation on a training image so as to acquire a second feature image; determining a feature vector according to the second feature image; carrying out processing on the feature vector so as to acquire a category probability vector; calculating a category error according to the category probability vector and the initial category; adjusting model parameters based on the category error; and continuing the model parameter adjusting process based on the adjusted model parameters, and using model parameters at the moment when the number of iterations reaches a preset number of times as model parameters of the well trained CNN model. According to the invention, the convolution operation and the maximum pooling operation are carried out on the training image on different levels of convolution layers, and then the horizontal pooling operation is carried out. The horizontal pooling operation can extract a feature image marking a horizontal direction feature of the image from the feature image, so that the well trained CNN model is ensured to recognize images of any size, and the application range of the well trained CNN model in image recognition is expanded.

Description

technical field [0001] The invention relates to the field of image recognition, in particular to a training method and device for a convolutional neural network model. Background technique [0002] In the field of image recognition, a CNN (Convolutional Neural Network, Convolutional Neural Network) model is often used to determine the category of the image to be recognized. Before using the CNN model to identify the category of the image to be recognized, the CNN model needs to be trained first. [0003] When training a CNN model, it is usually implemented in the following way: First, initialize the model parameters of the CNN model to be trained, which include the initial convolution kernel of each convolutional layer, the initial bias matrix of each convolutional layer, and the fully connected layer The initial weight matrix of and the initial bias vector of the fully connected layer. Next, from each pre-selected training image, a region to be processed with a fixed heig...

AI Technical Summary

Problems solved by technology

[0005] Since in the process of training the CNN model, it is necessary to obtain the fixed height and fixed width of the area to be processed from the pre-selected training images, therefore, the trained CNN model can only recognize images with fixed height and fixed width, resulting in poor training. The CNN model has certain limitations when recognizing images, resulting in limited scope of application

Images