Smoothing and Clustering Guided Image Decolorization

Fang Li, Yuanming Zhu


In this paper, we propose a new image decolorization method based on image clustering and weight optimization. First, we smooth the color image and cluster it into several classes and get the class centers. Each center can represent a distinctive color in the image. Then the class centers are sorted according to their brightness measured by Euclidean norm. By assuming that the decolorized grayscale image is a linear combination of the three channels of the color image, we propose an optimization problem by forcing the sorted class centers to correspond to specified grayscale values satisfying uniform distribution. Numerically, the problem is solved by quadratic programming. Experiments on two popular data sets demonstrate that the proposed method is competitive with the state-of-the-art decolorization method.

clustering; contrast; decolorization; optimization; smoothing

Full Text:


DOI: 10.5566/ias.2348

Copyright (c) 2021 Image Analysis & Stereology

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

Image Analysis & Stereology
EISSN 1854-5165 (Electronic version)
ISSN 1580-3139 (Printed version)