Image processing method and device based on convolutional neural network

The invention provides an image processing method and device based on a convolutional neural network, and the method comprises the steps: executing a first convolution layer operation according to preset convolution operation strategy information, and judging whether the information of a second convolution layer operation to be executed meets the following conditions: the convolution operation size is 1*1, and the stepping length value s is greater than 1; under the condition that the conditions are met, screening out an effective value part for correspondingly executing the second convolutionlayer operation in the first convolution layer operation result according to the stepping length value before executing the second convolution layer operation; and executing the second convolution layer 1*1 convolution operation on the effective value part according to the stepping length value of 1. Therefore, in the process of executing convolution operation for image processing, the corresponding effective value is sc