IPOL Journal · Image Processing On Line
- published
- 2011-06-22
- reference
- Jose-Luis Lisani, Antoni Buades, and Jean-Michel Morel, Image Color Cube Dimensional Filtering and Visualization, Image Processing On Line, 1 (2011), pp. 57–69. https://doi.org/10.5201/ipol.2011.blm-cdf
Communicated by Julie Delon
Demo edited by Jose-Luis Lisani
Abstract
We call RGB cube the three dimensional representation of the colors of a digital image in RGB space. It has been observed that 2D structures predominate in the color cube. They are qualitatively explained by two main causes: image blur and texture. In this article a filtering algorithm, the Local Linear Projection (LLP2), is used to smooth the 3D color distribution and to reveal the underlying 2D structures of the color clusters.
Download
- full text manuscript: PDF low-res. (522.8K) PDF (6.7M) [?]
- source code: TAR/GZ
Supplementary Materials
History
- this article was converted to PDF on 2015-10-23
- the original version was published on 2011-06-22: manuscript
- Note from the editor: The source code was updated on July 5, 2011 to fix a bug that occurs on images whose colors are sparsely distributed in RGB space.
Note from the editor: The source code was updated on January 21, 2021 to fix a bug that occurs on gray scale images. These images must not be filtered by the algorithm. The previous version of the code can be found here.
Note from the editor: the manuscript of the article was modified on 2022-01-01 to include information about its editors. The original version of the manuscript is available here.
