A Study of Visual Attention Elements with Experiment Analysis Based on Composition

The elements in the image that can attract people's visual attention are visual attention elements. Composition is a kind of visual perception feature in the study of audio-visual fusion information processing. The balance of composition depends on the distribution of visual attention elements in the image. Aiming at the problems of poor pertinence of existing visual attention models, unsystematic classification, unscientific selection of visual attention elements and insufficient consideration of human visual perception, this paper proposes a method of visual attention elements screening based on composition. Firstly, a material library based on composition is constructed, and PAD and harmony are marked. Then, a subjective assessment experiment is designed to screen out visual attention elements based on human perception. Finally, the experimental data were analyzed, the elements of visual attention based on composition were screened out, classified and summarized.