Related Experiment Video
Updated: Mar 19, 2026
07:30
Optimization for Sequencing and Analysis of Degraded FFPE-RNA Samples
Published on: June 8, 2020
12.9K
Full-Reference Image Quality Assessment with Linear Combination of Genetically Selected Quality Measures
1Department of Computer and Control Engineering, Rzeszow University of Technology, Rzeszow, Poland.
Plos One
|June 25, 2016
Summary
Developing algorithms for automatic image quality assessment (IQA) is crucial. This study proposes a novel approach combining multiple IQA methods to achieve human-consistent image quality evaluation.
Area of Science:
- Computer Vision
- Image Processing
- Signal Processing
Background:
- Image information can be distorted through electronic storage and communication.
- Automatic image quality assessment (IQA) algorithms are needed to align with human perception.
- Existing IQA methods may not consistently reflect subjective quality.
Purpose of the Study:
- To develop a robust image quality assessment approach.
- To create a joint model that integrates multiple IQA algorithms.
- To ensure automated quality evaluation aligns with human judgment.
Main Methods:
- Proposed an IQA approach using a weighted sum of multiple IQA measures.
- Defined an optimization problem to minimize the error between objective and subjective scores.
- Employed a genetic algorithm to determine optimal weights and select relevant IQA measures.
- Evaluated the joint model on four large image benchmarks.
Main Results:
- The proposed joint multimeasure IQA approach demonstrated superior performance.
- Outperformed existing state-of-the-art full-reference IQA methods in evaluations.
- Achieved image quality assessment consistent with human observers.
Conclusions:
- Combining multiple IQA measures through optimization significantly enhances assessment accuracy.
- The genetic algorithm effectively optimizes the aggregation of diverse IQA techniques.
- The developed approach offers a more reliable and human-aligned method for image quality evaluation.
