This repo contains Kmeans and HSV pixel range Color Recognition methods.

to cluster domianant color of a given box, input is box coordinate(x,y,w,h), return is color name

Advantage: accurate

Disadvantage: Time-consuming

Reference: K-Means Clustering in OpenCV

set up a HSV color look up table first. then compare each pixel of roi image with the look up table, then the largest distribution of color is the dominant.

1.node-colour-extractor: Extract colour palettes from images

2.Color Thief: Grabs the dominant color or a representative color palette from an image.

3.Color Extractor:a library and a CLI tool to extract the dominant colors of the main object of an image

4.How We Tackled Color Identification?--a blog

5.Image Background Removal

6.Opencv: Interactive Foreground Extraction using GrabCut Algorithm