Optimization method and device of convolutional neural network (CNN) and computer storage medium

Abstract

The embodiment of the invention discloses an optimization method and device of a convolutional neural network (CNN) and a computer storage medium. The method includes: constructing the convolutional neural network, wherein the convolutional neural network includes at least four network layers of an image input layer, at least one convolutional layer, at least one pooling layer and at least one fully connected layer; decreasing the number of convolution kernels in the CNN when the number of to-be-detected objects is less than a preset threshold value; dividing an input image of the image inputlayer into at least one memory data segment, which is stored by utilizing continuous memory, according to a set border determination strategy, and utilizing a set continuous memory copy function to carry out data copying on each memory data segment; and merging original parameters in a batch normalization (BN) layer and parameters of the convolutional layer or the fully connected layer according to the set merging strategy, and using merged parameters as new parameters of the batch normalization layer, wherein the batch normalization layer is after the convolutional layer or the fully connected layer. Calculation consumption in carrying out detection through the CNN is reduced.

Description

technical field [0001] The present invention relates to the technical field of image processing, in particular to a convolutional neural network (CNN, Convolutional Neural Network) optimization method, device and computer storage medium. Background technique [0002] With the development of computer technology and signal processing technology, more and more home appliances can be controlled according to the user's voice or gesture in addition to the traditional button operation control. [0003] To control home appliances through gestures, it is necessary to detect objects such as hands. In the current related hand detection schemes, the hand detection scheme based on deep convolutional neural network (CNN, Convolutional NeuralNetwork) is usually adopted. Convolutional neural network for hand detection on first-view RGB images. Since this scheme sets a deep neural network layer and needs to perform full convolution processing on the entire image during the detection proces...

AI Technical Summary

Problems solved by technology

Since this scheme sets a deep neural network layer and needs to perform full convolution processing on the entire image during the detection process, the detection accuracy of this scheme is high, but at the same time it increases the calculation cost, which also increases the calculation cost of the scheme. time spent during detection

Images