Color image reconstruction method and device and image classification method and device

A color image and image technology, applied in the field of color image reconstruction method and device, image classification method and device, can solve the problem that the reconstructed image cannot approach the image to be classified, etc., so as to improve the classification accuracy, improve the accuracy, avoid the The effect of bias

Active Publication Date: 2020-04-28
EZHOU INST OF IND TECH HUAZHONG UNIV OF SCI & TECH +1
View PDF8 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The present invention provides a color image reconstruction method and device, an image classification method and device, to solve or partially sol

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
  • Color image reconstruction method and device and image classification method and device
  • Color image reconstruction method and device and image classification method and device
  • Color image reconstruction method and device and image classification method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0045] In order to make those skilled in the technical field of the present invention understand the present invention more clearly, the technical solutions of the present invention will be described in detail below through specific embodiments in conjunction with the accompanying drawings.

[0046] The solution in the prior art is to propose the generated network model as a generator for image reconstruction based on the already trained Deep Convolutional Generative Adversarial Network (DCGAN) model. When the generator and the discriminator in the deep convolution generation confrontation network model have sufficient data representation capabilities, the goal is to make the noise z randomly obtained from the latent space generate an image G(z) through the generation network, and use it several times The gradient descent method makes the generated reconstructed image approximate to the image x to be classified, that is, it needs to find a noise z in the latent space, so that t...

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 a color image reconstruction method and device, and an image classification method and device, and the method comprises the steps of extracting a generation network model and adiscrimination network model through the training of a tensor-based deep convolution generative adversarial network, obtaining a generator based on the generation network model, and obtaining a decoder based on the discrimination network model; converting a color image to be classified into a color image code by using the decoder, wherein the similarity between the spatial distribution of the color image coding and the spatial distribution of the noise is higher than a preset threshold, and the noise is obtained from a potential space; repeatedly inputting the color image codes into the generator to generate a plurality of target color images; evaluating and scoring the plurality of target color images by using the discrimination network model; and taking the target color image of which the score meets the requirement as the reconstructed color image.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a color image reconstruction method and device, and an image classification method and device. Background technique [0002] In recent years, in the field of Cyber-Physical Social Systems (CPSS), with the development of IoT devices and the deepening of related research, the scale of data has become larger and the types of data have become more and more complex. Applications are also becoming more and more common. Image classification, as a typical image application, has been widely studied by people, and related technologies emerge in endlessly. However, most of the existing image classification technologies are based on neural network models, and the models are generally linear. When , even small noises can be amplified several times after the output, so that the classification result output by the classifier is wrong, and even the attacker can specify the output of th...

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): G06T9/00
CPCG06T9/001
Inventor 杨天若朱宇翔冯君张顺利任志安
Owner EZHOU INST OF IND TECH HUAZHONG UNIV OF SCI & 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