Image quality assessment: the integral CTS index for holistic similarity evaluation of images (ISII)
摘要
This manuscript introduces the Integral CTS Image Similarity Index (ISII), a new integral metric for assessing the similarity between two digital images. ISII combines color, structure, and texture analyses into a single numerical parameter, enabling a holistic comparison of the visual characteristics of images. ISII is the most sensitive parameter for evaluating similarity between two images, capable of detecting a change as small as a single grayscale level in one channel of an individual pixel in the observed image. Existing metrics typically focus on individual aspects without considering their interrelationships and influences. ISII overcomes this limitation by integrating the analysis of all key visual elements into a unified similarity indicator, making it suitable for numerous applications in the field of computer vision, image retrieval, duplicate content detection, and similar tasks.