Color image color meaning classification method of based on fully convolutional network

A fully convolutional network, color image technology, applied in the field of color semantic classification of color images, to achieve the effect of improving the training speed, improving the accuracy of color semantic classification, and increasing the type and quantity

Inactive Publication Date: 2017-08-04
HEFEI UNIV OF TECH
View PDF4 Cites 23 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0009] In order to solve the deficiencies in the prior art, the present invention provides a color image color semantic classification method based on a fully convolutional network, aiming at solving the color image pixel-level color attribute semantic classification problem, by construct

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 color meaning classification method of based on fully convolutional network
  • Color image color meaning classification method of based on fully convolutional network
  • Color image color meaning classification method of based on fully convolutional network

Examples

Experimental program
Comparison scheme
Effect test

Example Embodiment

[0048] Such as figure 1 As shown, in this embodiment, a method for color semantic classification of color images based on a full convolutional network is performed according to the following steps:

[0049] Step 1. Construct a full convolutional network for pixel-level color semantic classification of color images I(x,y,k) of any size; a full convolutional network consists of a convolutional layer, a pooling layer, and a deconvolutional layer ,

[0050] Such as figure 2 As shown, in this embodiment, the full convolutional network includes five stages of convolutional pooling operations: the first and second stages each include two convolutional layers and one pooling layer; each of the third, fourth, and fifth stages It includes three convolutional layers and one pooling layer. The full convolutional network has 13 convolutional layers, five pooling layers, and three deconvolutional layers. The arbitrary size of the color image means: the size of the color image sent to the net...

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 color meaning classification method of based on a fully convolutional network. The method includes 1. constructing a fully convolutional network; 2 obtaining a color image data set with pixel level labeling; 3. training the fully convolutional network by means of the color image data set to obtain the feature model for performing pixel level color meaning classification on color images in any size; 4. performing the pixel level color meaning classification on any color image by the feature model to evaluate the classification precision of the feature model; and 5. optimizing the network classification result by a whole connection condition random field method to obtain the color class label of each pixel in the image, and transforming the color class labels into the corresponding color spaces to display the pixel level color meaning classification result according to the mapping relationship between the class labels and the color spaces. The method provided can realize the pixel level color meaning classification of the color images, and effectively improves the classification precision of the color image color meaning in the complex and changeable environment.

Description

Technical field [0001] The invention belongs to the field of computer / machine vision, image processing and analysis, and specifically is a color image color semantic classification method based on a full convolution network. Background technique [0002] In computer vision, color is an important attribute of images and an important way for humans to perceive image information. By assigning image color category labels, it can be further used in image retrieval, image annotation, color blindness assistance, visual tracking, language human-computer interaction and other fields. Therefore, good color semantic classification results are helpful for further image processing and image analysis. [0003] The existing image color semantic classification methods include methods based on statistical models and methods based on deep learning. [0004] Methods based on statistical models are mainly based on color stimuli, such as color semantic classification by perceiving the three color appea...

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): G06K9/62G06N3/04G06N3/08G06T7/90
CPCG06N3/08G06T2207/10024G06N3/045G06F18/241
Inventor 张骏熊高敏高隽张旭东
Owner HEFEI 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