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

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...

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 constructing a fully convolutional neural network , to obtain a color attribute semantic classification network feature model with good classification accuracy in complex scenes, thereby improving the accuracy of color semantic classification of color images in complex and changeable environments

Images