Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Attention mechanism-based image classification method of pulse convolutional neural network

A technology of convolutional neural network and classification method, applied in the field of image classification of impulse convolutional neural network, can solve problems such as poor image classification effect, achieve good network classification effect, fast training and recognition speed, and save computing cost. Effect

Pending Publication Date: 2020-10-30
XI'AN POLYTECHNIC UNIVERSITY
View PDF5 Cites 17 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a kind of image classification method based on the pulse convolutional neural network of attention mechanism, which solves the problem that the picture classification effect in the prior art is not good

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
  • Attention mechanism-based image classification method of pulse convolutional neural network
  • Attention mechanism-based image classification method of pulse convolutional neural network
  • Attention mechanism-based image classification method of pulse convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0062] Execute step 1, using the MNIST dataset as the dataset;

[0063] The MNIST dataset comes from the National Institute of Standards and Technology (NIST). The training set consisted of handwritten digits from 250 different people, 50% of whom were high school students and 50% from Census Bureau workers. The test set is also the same proportion of handwritten digital data. The MNIST data set classifies and recognizes 10 numbers from 0 to 9. The training set contains 60,000 pictures, and the test set contains 10,000 pictures.

[0064] Execute steps 2 to 4:

[0065] In the Gaussian difference temporal coding layer, σ 1 and σ 2 are 1 and 2 respectively, and the Gaussian kernel window size is 7 in this embodiment, that is, the value range of x and y is [-3,3],

[0066] The size of the two-dimensional convolution kernel template kernel is set to 17×17;

[0067] In the first impulse convolutional attention layer and the second impulse convolutional attention layer, the init...

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 discloses an attention mechanism-based image classification method of a pulse convolutional neural network, which is characterized by comprising the following specific steps: 1, downloading a data set; 2, preprocessing the data set to obtain a preprocessed data set; 3, establishing a pulse convolutional neural network model, wherein the pulse convolutional neural network model adoptsa leakage integral ignition neuron model; 4, training the preprocessed data set by adopting the pulse convolutional neural network model to obtain a trained pulse convolutional neural network model;and step 5, inputting to-be-classified pictures, and classifying the to-be-classified pictures by adopting the trained pulse convolutional neural network model to obtain a classification result. According to the invention, the problem of poor picture classification effect in the prior art is solved.

Description

technical field [0001] The invention belongs to the technical field of image classification, and relates to an image classification method based on an attention mechanism-based impulse convolutional neural network. Background technique [0002] The visual attention mechanism is a brain signal processing mechanism unique to human vision. When the brain performs visual tasks, it will always give priority to obtaining information that it considers useful, and discard the secondary content directly. The attention mechanism has the ability to make the neural network only focus on selecting specific feature inputs. Applying the attention mechanism can improve the efficiency and accuracy of neural network information processing. At present, the attention mechanism is widely used in the field of deep learning and has achieved good results. [0003] Image classification is a basic research problem in the field of computer vision, and the research on image classification has wide a...

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): G06F16/55G06N3/04G06N3/08
CPCG06F16/55G06N3/049G06N3/08G06N3/045Y02D10/00
Inventor 赵雪青张军军
Owner XI'AN POLYTECHNIC UNIVERSITY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products