Cuboid Colour Image Segmentation Using Intuitive Distance Measure

In this paper, an improved algorithm for cuboid image segmentation is proposed. To address the two main limitations of the recently proposed cuboid segmentation algorithm, the improved algorithm substitutes colour quantization in HCL colour space with infinity norm distance in RGB colour space along...

Full description

Saved in:
Bibliographic Details
Main Authors: Tania, Sheikh, Murshed, Manzur, Teng, Shyh Wei, Karmakar, Gour
Format: Conference Proceeding
Language:English
Subjects:
Classification algorithms
Cuboid segmentation
Image color analysis
Image retrieval
Image segmentation
infinity norm
Measurement
Partitioning algorithms
Quantization (signal)
segmentation quality metric
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this paper, an improved algorithm for cuboid image segmentation is proposed. To address the two main limitations of the recently proposed cuboid segmentation algorithm, the improved algorithm substitutes colour quantization in HCL colour space with infinity norm distance in RGB colour space along with a different way to impose area thresholding. We also propose a new metric to evaluate the quality of segmentation. Experimental results show that the proposed cuboid segmentation algorithm significantly outperforms the existing cuboid segmentation algorithm in terms of quality of segmentation.
ISSN:2151-2205
DOI:10.1109/IVCNZ.2018.8634676