Color feature extraction method and system

A technology of color features and extraction methods, applied in the field of extracting color features to form vectors for feature matching and color feature extraction, can solve the problems of ignoring image spatial distribution information, not comprehensively considering other features of image color, and complex content, so as to avoid matching. Low efficiency, accurate feature description, and the effect of avoiding interference

Pending Publication Date: 2020-08-28
深圳市图郅创新科技有限公司
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Due to the complex content and variety of color images, and the different feature emphases of different images, the feature extraction of color images is very challenging.
[0003] At present, most of the image color features are based on statistical features of image color, using methods such as color histogram, without comprehensive consideration of other features of image color, ignoring the spatial distribution information of the image

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 feature extraction method and system
  • Color feature extraction method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0030] Such as figure 1 Shown is the method flowchart of color feature extraction of the present invention, and concrete color feature extraction method is as follows:

[0031] S10 acquires images and preprocesses them.

[0032] Get the image to be processed, convert the image to HSV color space, and divide the image into multiple cells.

[0033] Specifically, the image is first converted to the HSV color space, scaled to a fixed size of 320*320, and divided into 32*32 cells with a size of 10*10. Scaling the image to a specified size is beneficial to divide it into multiple cells in equal parts. After the image is divided into multiple cells, the color feature extraction of a single cell is performed, compared to the overall processing of the entire image , processing image cells one by one can extract color features more accurately. The division method in this embodiment is only an example, and is not intended to limit the present invention.

[0034] S11 quantizes the ima...

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 relates to a color feature extraction method and system, and the method specifically comprises the steps: obtaining an image to be processed, converting the image into an HSV color space, and dividing the image into a plurality of cells; quantifying the color of the image; constructing color feature descriptors, judging whether the image in each cell has a main color or not, whereinthe images in the cells have main colors; extracting and describing the features of the main colors, wherein the images in the cells do not have the main colors; extracting and describing the featuresof mosaic conditions with random color distribution, and combining the features of the main colors and the features of the mosaic conditions to obtain color feature descriptors; and extracting colorfeature dimensions, and performing dimension removal processing on the color features. Whether the main color exists in the pattern or not is comprehensively considered, and dimension removal processing is performed on the image so that the extracted color features can be compared on the same scale.

Description

technical field [0001] The present invention relates to the field of image data processing, in particular to a color feature extraction method and system, in particular to a method and system for extracting color feature vectors for feature matching. Background technique [0002] At present, color image feature extraction has become a very important research topic in the field of image processing technology. The feature extraction of color images has a wide range of applications, such as target recognition and detection, image clustering, big data retrieval and other fields. Due to the complex content and variety of color images, the feature emphases of different images are different, which makes the feature extraction of color images very challenging. [0003] At present, image color feature extraction is mostly based on statistical features of image color, using methods such as color histogram, without comprehensive consideration of other features of image color, ignoring...

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): G06K9/46G06T7/90
CPCG06T7/90G06T2207/10004G06T2207/10016G06V10/56
Inventor 欧阳铖黄伟杰
Owner 深圳市图郅创新科技有限公司
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