<abstract xmlns="http://www.w3.org/1999/xhtml">

<p>In this paper, HOG features are used to describe positive and negative sample images, and visual communication elements are extracted by distinguishable mean shift algorithm, and the visual communication elements finally obtained can represent the characteristics of this category and also have a good differentiation to other categories. On the basis of visual communication elements, the set generator, relationship-based discriminator and wireframe rendering discriminator are used to construct the layout generation model of visual communication elements, and the graphic design based on the generation model is empirically analyzed by means of experimental simulation analysis. The results show that after the number of iterations reaches 10,000, the fluctuation range of the generator loss value (0.237~0.154) is very unstable, indicating that the effect of the graphic design based on the generative model has a small real difference with the original image. The change of the discriminator loss value shows a trend of increasing and then decreasing, and its maximum and minimum values are 0.954 and 0.487, respectively, indicating that the input data is not trained in place, and the simple planar design, which is more biased, leads to the effect of the discriminator is not obvious. By reducing the manual operation in the design process, saving the designer’s drawing performance time, and promoting healthy development of graphic design, this study is significant.</p>
</abstract>

In this paper, HOG features are used to describe positive and negative sample images, and visual communication elements are extracted by distinguishable mean shift algorithm, and the visual communication elements finally obtained can represent the characteristics of this category and also have a good differentiation to other categories. On the basis of visual communication elements, the set generator, relationship-based discriminator and wireframe rendering discriminator are used to construct the layout generation model of visual communication elements, and the graphic design based on the generation model is empirically analyzed by means of experimental simulation analysis. The results show that after the number of iterations reaches 10,000, the fluctuation range of the generator loss value (0.237~0.154) is very unstable, indicating that the effect of the graphic design based on the generative model has a small real difference with the original image. The change of the discriminator loss value shows a trend of increasing and then decreasing, and its maximum and minimum values are 0.954 and 0.487, respectively, indicating that the input data is not trained in place, and the simple planar design, which is more biased, leads to the effect of the discriminator is not obvious. By reducing the manual operation in the design process, saving the designer’s drawing performance time, and promoting healthy development of graphic design, this study is significant.

Expression and Application of Visual Communication Elements in Graphic Design under the Background of Internet+

Applied Mathematics and Nonlinear Sciences

This work is licensed under the Creative Commons Attribution 4.0 International License.

In this paper, HOG features are used to describe positive and negative sample images, and visual communication elements are extracted by distinguishable mean...

Expression and Application of Visual Communication Elements in Graphic Design under the Background of Internet+

Publié en ligne: 31 janv. 2024

Pages: -

Reçu: 18 déc. 2023

Accepté: 27 déc. 2023

DOI: https://doi.org/10.2478/amns-2024-0148

Mots clés
HOG features, Mean shift algorithm, Generator, Discriminator, Visual communication elements

© 2024 Guiwei Bai et al., published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Expression and Application of Visual Communication Elements in Graphic Design under the Background of Internet+

Publié en ligne: 31 janv. 2024

Pages: -

Reçu: 18 déc. 2023

Accepté: 27 déc. 2023

DOI: https://doi.org/10.2478/amns-2024-0148

Mots clésHOG features, Mean shift algorithm, Generator, Discriminator, Visual communication elements

© 2024 Guiwei Bai et al., published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Mots clés
HOG features, Mean shift algorithm, Generator, Discriminator, Visual communication elements