BP classification algorithm based on improved bat algorithm

A bat algorithm and bat technology, applied in computing, manufacturing computing systems, computer components and other directions, to achieve fast convergence speed, enhance local search capabilities, and improve classification accuracy.

Active Publication Date: 2021-08-06
CHANGCHUN UNIV OF TECH
View PDF6 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] The object of the present invention is to provide a kind of weight and the threshold value that trains neural network with improved bat algorithm (WG-BA), this method can overcome the shortcoming that error backpropagation network convergence time is long, easily falls into local optimum, and then Significantly improved image classification accuracy to address issues mentioned in the background story above

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
  • BP classification algorithm based on improved bat algorithm
  • BP classification algorithm based on improved bat algorithm
  • BP classification algorithm based on improved bat algorithm

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0025] Step 1: Input the original image, perform LBP processing on the original image, extract feature vectors of information such as inclusions, plaques, cracks, pitting, rolling scale, scratches, etc., and input them to the network.

[0026] Step 2: Initialize the network, set the number of layers of the network, and the number of nodes in each layer of the network.

[0027] Step 3: assign values ​​to the initial parameters of the present invention, including initial position, speed, loudness, frequency, maximum number of iterations, etc.

[0028] Step 4: Calculate the weight experience factor, the position of the bat is determined by X i (t+1)=X i (t)+V i (t+1), the velocity is determined by the equation V ik (t+1)=ω·V ik (t)+(X ik (t)-P k (t)) f i (t) Determine and update loudness and pulse rate.

[0029] Step 5: Record the global optimal position and local optimal position of the current population, and update the speed update using the formula, and obtain the new...

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 improved bat algorithm for training the weight and the threshold value of a neural network, and the accuracy of image classification is greatly improved. The method comprises the following algorithm steps: 1, inputting an original image, and processing the original image; 2, initializing the network; 3, assigning the initial parameters of the method; 4, calculating a weight empirical factor, moving the bat by using an equation, and updating the loudness and the pulse rate; 5, recording the global optimal position and the local optimal position of the current population, updating the speed by using a formula, and obtaining the position of the bat of the population according to the formula; 6, enabling the optimal solution X to respectively correspond to the weight and the threshold of the network, and outputting a result; and 7, judging whether the maximum number of iterations is reached or not, if so, outputting a result, and if not, returning to the step 4. Compared with other algorithms, the method has the advantages of higher convergence speed, higher development capability and higher stability.

Description

Technical field: [0001] The invention relates to the technical field of steel strip surface detection, in particular to an algorithm for matching images and identifying six defects including inclusions, plaques, cracks, pitting, rolling scale and scratches. Background technique: [0002] In recent years, with the rapid development of image recognition technology, machine vision technology is gradually penetrating into all aspects of production and processing. Among them, the surface quality inspection of strip steel is a very suitable part of its application. [0003] Steel plates are prone to pitting, cracking, pits, scratches and other problems during the production process. Therefore, its surface defect detection has been widely concerned by steel plate manufacturers. [0004] Using image recognition technology can solve the classification problem of steel plate defects. Image recognition and technology can be divided into two steps: image feature extraction and image c...

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
IPC IPC(8): G06K9/62G06N3/04G06N3/08
CPCG06N3/084G06N3/044G06F18/241Y02P90/30
Inventor 岳晓峰卢禹成高学亮马国元张守鑫郜军涛于显宁张明志
Owner CHANGCHUN UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap