Image quality assessment based on metrics using quantum algorithms
Abstract
This paper analyzes the mathematical foundations, practical calculation algorithms, and responses to filtered images of three primary no-reference image quality metrics —Blind/Reference less Image Spatial Quality Evaluator , Natural Image Quality Evaluator, and Blind Image Quality Index. Classical Gaussian and quantum-inspired filtering methods were applied to noisy images, with quality indicators for each method evaluated through Blind/Reference less Image Spatial Quality Evaluator, Natural Image Quality Evaluato, and Blind Image Quality Index. Experiments were conducted in Python environment on a set of real images. Results indicate that quantum filtering can significantly improve image quality in certain cases. The paper also mathematically investigates the conditions under which each metric provides optimal assessment.