A Deep Neural Network Method Based on Spatial Fusion Pooling

Abstract

The invention relates to a deep neural network method based on space fusion pooling, which is used for image classification, including: collecting various types of images, and marking the image categories as image label information; dividing image sets; dividing collected images For the training set, verification set and test set, the training set is used to train the convolutional neural network; the deep neural network structure applied to image classification is designed, including the number of convolutional layers used and the number of layers of the spatial fusion pooling layer, and the design The number of filters in the convolutional layer, design the fusion function form and spatial sliding step size and pooling function and pooling window size and step size in the space fusion pooling layer, design the convolution filter for feature fusion Structure, design the number of iterations of the network training cycle and the final convergence conditions of the network, and initialize the network parameters; input the training data batches into the network for calculation and training.

Description

technical field [0001] The invention relates to a method for high-performance image recognition classification and object recognition in the field of computer vision, in particular to a method for image recognition classification and object recognition using a deep learning method. Background technique [0002] In recent years, deep learning technology has been widely used in multiple tasks such as image classification, semantic segmentation and object detection, and automatic driving in the field of computer vision. As an important implementation method in deep learning technology, deep convolutional neural network has achieved remarkable results in many tasks. [0003] Deep convolutional neural networks are often composed of multi-layer convolutional layers and pooling layers. The convolutional layer contains filter parameters for feature extraction, and the pooling layer is used to maintain the translation invariance of the neural network and reduce the impact of data dis...

AI Technical Summary

Problems solved by technology

[0011] The purpose of the present invention is to solve the problem that the pooling layer of the existing deep convolutional neural network cannot effectively extract deep-level features, and propose a deep convolutional neural network method based on spatial fusion pooling suitable for image classification. The channel (spatial) information between channels extracts more representative features, and at the same time, through the fusion of spatial information, the number of feature channels is reduced to further improve the efficiency of the neural network

Images