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

Sample extraction method and image classification method based on hole convolution and residual link

An extraction method and sample technology, applied in the field of image processing, can solve the problems of poor self-adaptability and inability to obtain sufficient robust training samples, etc., and achieve the effect of strong self-adaptability and cognitive ability

Active Publication Date: 2020-04-14
CHANGAN UNIV
View PDF6 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a sample extraction and image classification method based on atrous convolution and residual linking, which is used to solve the poor adaptive ability of the current deep learning classification network in the prior art, and the inability to obtain sufficient and robust training samples, etc. question

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
  • Sample extraction method and image classification method based on hole convolution and residual link
  • Sample extraction method and image classification method based on hole convolution and residual link
  • Sample extraction method and image classification method based on hole convolution and residual link

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0035] In the process of biological vision forming scene recognition, only the central fovea of ​​the human retina contains a large number of rod cells that can be clearly imaged, while the peripheral area can only be fuzzy imaged by cone cells, resulting in the need for micro-fibrillation (microsaccade, Also known as "micro-saccade", "micro-saccade"), the neural mechanism is to scan the scene by unconsciously and quickly turning the eyeballs to form an image of the entire scene. And in the process of microfibrillation, the blurred vision in the peripheral area of ​​the retina provides probabilistic regional cognition, forming a priori reference, and on this basis, combined with the clear vision of the fovea, a complete and clear scene cognition is formed.

[0036] A sample extraction method, comprising the steps of:

[0037] Step 1: Get the original image Oimg, shrink Oimg n times, where, L represents the length of Oimg, W represents the width of Oimg, obtains an image pyra...

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 belongs to the technical field of image processing. The invention discloses a sample extraction method and an image classification method based on hole convolution and residual link. According to the sample extraction method, an image pyramid of an original image is constructed, then convolution is carried out on the image pyramid, an identification result is extracted, a region identification image with a category label is finally obtained through methods of binarization, fuzzy clustering and the like, and finally a homologous reliable classification sample is obtained through image mapping. According to the image classification method based on cavity convolution and residual link, a completely self-adaptive deep learning classification network is constructed by a method ofsimulating clear vision in the central fovea, different types of unique feature modes are learned from homologous reliable samples of the images, and completely self-adaptive classification is carriedout on the images according to the learned feature modes.

Description

technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to a sample extraction and an image classification method based on atrous convolution and residual linking. Background technique [0002] Image classification is the most basic technical means in the hot machine vision and artificial intelligence. As the most basic image analysis method, image classification can reasonably combine various information from complex and redundant image data, so as to integrate information with certain characteristics into the same category and complete the classification. This technology can provide fast, accurate, and non-redundant prior references for a wide range of practical applications such as target recognition, detection, and tracking, and is the basis for various image analysis tasks. In particular, the image classification technology based on deep learning, based on the powerful cognitive advantages of the neural network...

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): G06K9/62G06K9/46
CPCG06V10/56G06F18/241
Inventor 丛铭韩玲席江波顾俊凯丁明涛
Owner CHANGAN UNIV
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

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com