Supercharge Your Innovation With Domain-Expert AI Agents!

Deep convolutional neural network method based on parallel convolution units

A deep convolution, neural network technology, applied in the fields of computer vision, image processing, and pattern recognition, can solve problems such as network performance degradation, and achieve the effect of strong generalization ability, good performance, and simple program.

Inactive Publication Date: 2018-03-06
TIANJIN UNIV
View PDF0 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0009] This patent application provides a deep neural network method based on parallel convolution units to solve the problem of network performance degradation after traditional convolution is simplified in the prior art. This method can effectively extract features and improve network classification performance

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Deep convolutional neural network method based on parallel convolution units
  • Deep convolutional neural network method based on parallel convolution units
  • Deep convolutional neural network method based on parallel convolution units

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0028] The most critical idea of ​​the present invention is that: the present invention first constructs the parallel convolution unit, and constructs a deep convolutional neural network based on this convolutional unit, trains the convolutional neural network to obtain a good classifier, and uses this classifier Image classification has the advantage of high accuracy. This patent can be applied to image classification tasks, but is not limited to this task. The convolutional neural network can be applied to many tasks of deep learning.

[0029] This patent provides a method for improving the performance of a deep convolutional neural network based on parallel convolutional units. The convolutional neural network system mainly includes two stages: a training stage and a testing stage. The present invention is applied to both stages at the same time.

[0030] The mathematical representation of the traditional convolution filter is W×H×M×N, where W is the width of the filter, H...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention relates to a deep convolutional neural network method based on parallel convolution units. A training stage of the deep convolutional neural network method comprises the steps of 1) collecting data; 2) defining each parallel convolution unit to be formed by arranging a simplified convolution and a variant of the simplified convolution in a parallel mode, wherein the variant of the simplified convolution is formed by exchanging two filters adopted in the original simplified convolution in position, setting the structure of the deep convolutional neural network, wherein the settingcomprises the number and the combination mode of the parallel convolution units, setting the number and size of feature maps of each convolution layer, and setting the pooling mode and the size of apooling window; 3) performing initialization; 4) performing forward calculation; and 5) performing back propagation.

Description

technical field [0001] The invention relates to the fields of pattern recognition, computer vision and image processing, in particular to methods for image classification and object recognition based on deep convolutional neural networks. Background technique [0002] In recent years, deep learning models based on convolutional neural network ideas have been widely used in many tasks such as image classification, target recognition, and object detection in the field of computer vision, and have achieved remarkable results. [0003] Convolutional Neural Networks (CNNs, Convolutional Neural Networks) provide an end-to-end learning model, usually consisting of several convolutional layers, activation layers and pooling layers alternately. Among them, the convolution layer performs feature extraction through the convolution kernel and input features, and can learn hierarchical features. The shallow convolutional layer mainly extracts information such as edges and directions, wh...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06N3/04G06N3/08
CPCG06N3/084G06N3/045
Inventor 庞彦伟侯聪聪
Owner TIANJIN UNIV
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com