Related Experiment Video
Updated: Sep 29, 2025
13:44
Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
Published on: August 30, 2013
43.1K
Image Retrieval Using Different Distance Methods and Color Difference Histogram Descriptor for Human Healthcare
Himani Chugh1, Sheifali Gupta1, Meenu Garg1
1Chitkara University Institute of Engineering and Technology, Chitkara University, Chandigarh, Punjab, India.
Journal of Healthcare Engineering
|March 24, 2022
Summary
This study introduces a novel image retrieval method using the color difference histogram (CDH) descriptor in the L*a*b* color space. It effectively extracts similar images from large medical datasets, optimizing retrieval accuracy.
Area of Science:
- Computer Science
- Image Processing
- Medical Informatics
Background:
- The rapid expansion of multimedia technology necessitates efficient methods for managing and retrieving large image datasets.
- Image retrieval systems are crucial for applications ranging from general multimedia use to specialized fields like medical imaging.
Purpose of the Study:
- To develop and evaluate an image retrieval system that extracts similar images based on their features from extensive image datasets.
- To specifically investigate the effectiveness of the color difference histogram (CDH) descriptor for content-based image retrieval (CBIR).
Main Methods:
- The proposed method involves searching a query image within a dataset and utilizing the color difference histogram (CDH) descriptor.
- CDH calculates color differences between distinct labels within the L*a*b* color space.
- Image features are extracted using various distance metrics, and system performance is evaluated using precision, recall, and F-measure.
Main Results:
- The study demonstrates the application of the CDH descriptor for retrieving similar images from medical image datasets.
- Comparative analysis using the F-measure identifies the optimal distance method for image retrieval within this framework.
Conclusions:
- The color difference histogram (CDH) descriptor is a viable tool for content-based image retrieval, particularly in specialized domains like medical imaging.
- The choice of distance metric significantly impacts the performance of image retrieval systems, highlighting the need for careful selection and comparative analysis.
