No abstract

Although an image can consist of myriad different colors, due to the limitations of the human visual system and the mechanism of selective visual attention, only a limited number of them are prominent, that is, they stand out or are noticeable at first sight. In this article, a framework for building a model for extracting image prominent colors based on machine learning is presented. The model is learned on human‐extracted themes of prominent colors and uses numerous features, which were defined based on the properties of human visual system. For the purpose of this study, we constructed a database of images with their associated human‐extracted themes of prominent colors, which are open to the public and available to other researchers. The analysis of observers' data shows a high interobserver agreement on prominent color categories as well as high diversity of prominent colors. According to our results, the most influential factors on the perception of prominent colors are associated with color coverage (which should be adjusted with a saliency map), color properties—lightness and chroma, and diversity of colors. To the best of our knowledge, this is the first study on image prominent colors and first attempt to extract them with a model trained on the real data. The presented model has a high practical importance, since it can be used for extracting image colors in different scenarios, for example, for automatic color design, image categorization, as a descriptor in content‐based image retrieval, and image content analysis frameworks.

The colour represents an essential element of visual and graphic communications. It plays an important role in the perception of visual design and it is significant for all participants in the process of planning, developing and promoting graphic products. Designers are interested in a psychological and presentational aspect of colours, while to the technologists the colour represents one of the most important quality attributes. The process of choosing colours that are harmonious, usable and efficient is complex. In addition, many designers have inadequate background knowledge of colour theory, which could help them with the selection of colours. As a result, designers usually spend a great deal of time and expend significant effort in choosing appropriate colour combinations. In this article, the importance of colour harmony and its application when extracting colours, rating and generating colour schemes is presented.

In a world with different readers with varying needs, the idea that readers might be given the option of text formatting their own text in advance seems very appealing. In this research, differences between reading speed, reading comprehension, and reading comfort were compared with the pre-set and self-set texts. The respondents read paragraphs and identified illogical words contained therein, while reading two comparable blocks of self-set and pre-set texts. The differences in the number of paragraphs read and the mistakes made in a limited timeframe were compared using a web test that was based on the Tinker's test. The results showed that the respondents encounter different difficulties while reading digital text. We statistically proved that people with dyslexia chose larger type sizes more frequently than people without dyslexia, whereas no such statistically significant trend was observed for all other variables (typeface, tracking, and leading). We did not observe any significant differences between the reading speed associated with pre-set and self-set texts; however, the reading speed was higher in the group of people without dyslexia. A significant difference was observed in reading comprehension, because reading comprehension was better in the case of self-set text used.

An image can contain myriad different colours, but only a few of them are noticeable at first sight, hence in some way are defining the image. Therefore, the question arises which colours are prominent and what are the main factors that affect their prominence. In this paper, we present a webbased application that supports an interactive selection of colours and can be used to gather prominent colours for a set of images. The resulted database of images and their corresponding colours can be further used for investigating which image features or colour properties contribute to their prominence and for validating the models for automatic extraction of prominent colours from the images. First, the perception of prominent colours is addressed, and applied perspective of this knowledge is given. In the remainder of the paper, an overview of application architecture is presented, and a more detailed description of application's settings and usage is given. Although the main purpose of our application is to gather prominent colours of images based on the observer's opinion, the application could also be used for conducting other psychophysical experiments. The application supports three different modes for selecting the prominent colours: selection of basic hues, selection from ColorChecker Classic target and selection from custom defined patches.