Content-based image retrieval using block truncation coding based on edge quantization

In this paper, we propose an effective image retrieval approach using block truncation coding compressed data stream based on edge-based quantization (EQBTC). First, an image is compressed into corresponding quantisers and a bitmap image by EQBTC. Then, the quantisers are used for colour feature ext...

Full description

Saved in:
Bibliographic Details
Published inConnection science Vol. 32; no. 4; pp. 431 - 448
Main Authors Chen, Yan-Hong, Chang, Ching-Chun, Hsu, Cheng-Yi
Format Journal Article
LanguageEnglish
Published Abingdon Taylor & Francis 01.10.2020
Taylor & Francis Ltd
Subjects
block truncation technique
Color
colour histogram feature
Contest-based image retrieval (CBIR)
Data transmission
edge detection
edge structure feature
Feature extraction
Histograms
Image coding
Image management
Image retrieval
Measurement
Online AccessGet full text

Cover

More Information
Summary:In this paper, we propose an effective image retrieval approach using block truncation coding compressed data stream based on edge-based quantization (EQBTC). First, an image is compressed into corresponding quantisers and a bitmap image by EQBTC. Then, the quantisers are used for colour feature extraction, whereby the bitmap image and grey image are used for luminance and edge feature extraction. Subsequently, two image features, the colour histogram feature (CHF) and the overall structure feature (OSF), are computed to measure the similarity between two images using a specific distance metric computation. The results presented in this paper demonstrate that the proposed model is superior to the block truncation coding image retrieval scheme and some earlier proposed methods.
ISSN:0954-0091
1360-0494
DOI:10.1080/09540091.2020.1753174