La calidad de la apariencia

0
347

The article “The Quality of Appearance” by Garrett M. Johnson, published in 2005, examines the inherent complexity of the perception of image quality and how traditional color appearance models cannot fully encompass these aspects due to the spatially complex nature of images. While models such as CIECAM02 have advanced in the accurate prediction of the appearance of simple color stimuli, the apparent need for models that can handle the spatial complexity of images led to the development of the image appearance model called iCAM.

The iCAM model is extended to include spatial phenomena and additional attributes that describe the appearance of an image, such as contrast, graininess, and sharpness. These attributes are critical for perceiving and evaluating image quality. The incorporation of these aspects allows the iCAM model not only to predict the perceived appearance of an image but also the quality differences between images through prior spatial processing.

The article explains the relevance of various image appearance attributes and provides a comparison with traditional color appearance models. It provides examples of how traditional evaluations of attributes such as contrast and sharpness can be misleading if the complex spatial contexts of real images are not considered. For example, a perception of graininess can vary significantly depending on the image content and cannot be fully evaluated through uniform patterns.

Furthermore, the importance of chromatic adaptation and spatial frequency in image perception is discussed. Chromatic adaptation, well studied in color research, suggests that the human visual system adjusts to viewing conditions, which has direct implications on how image quality is perceived. Similarly, spatial frequency adaptation alters the sensitivity of the visual system depending on the frequencies present in the scene, affecting the perception of sharpness and contrast depending on the image content.

In summary, Johnson’s article emphasizes that to accurately assess image quality, a model that integrates multiple image appearance attributes and allows adaptation according to the image content is necessary. The iCAM model, by providing spatial filters and specific adaptations based on the frequencies and content of the image, manages to predict quality differences more effectively than traditional methods. These predictions are important not only for industries related to color science and digital imaging but also for any application that depends on high-quality images, where an accurate and contextual representation of appearance and quality is crucial.

Review on the publisher’s website