Fractally-Colorimetric Approach to Evaluating the Quality of Gamified Images

The thesis examines a new methodological approach called the fractally-colorimetric method, aimed at improving the evaluation of gamified images. The research focuses on analyzing current methods for assessing color and contrast, as well as identifying their limitations and problems encountered in their application. The analysis revealed that existing methods, such as histogram stretching, local contrast, and edge filtering, have certain drawbacks, such as increased noise, artifacts in high-contrast areas, and incorrect edge correction. To address these issues, a new approach has been developed that combines fractal color analysis with the Color Difference Evaluation 2000 (CIEDE2000) methodology. The fractally-colorimetric method involves using fractal algorithms to analyze and correct the color characteristics of the image. The proposed approach utilizes fractal models to identify and correct color anomalies, allowing for a more accurate and stable quality assessment. Fractal analysis helps to reveal recurring structures and patterns in images, which improves the accuracy of color correction and contrast.