Application of big data and artificial intelligence in visual communication art design

Area of Science:

• Digital Art and Design
• Computer Science Applications
• Artificial Intelligence in Creative Industries

Background:

• The integration of artificial intelligence (AI) and big data is transforming visual communication (VISCOM) art.
• Novel technologies are needed to enhance interdisciplinary artistic expression between art and technology.
• Current challenges involve creating art tailored to specific scenes and audiences.

Purpose of the Study:

• To investigate and apply big data and AI methods to VISCOM art.
• To develop and test the STING algorithm for multi-resolution information clustering in VISCOM.
• To utilize convolutional neural networks (CNNs) for identifying objects and scenes in VISCOM art.

Main Methods:

• Proposed the STING algorithm for efficient big data clustering in VISCOM.
• Employed convolutional neural networks (CNNs) for AI-driven identification of visual elements.
• Applied STING and CNNs to diverse VISCOM design projects including logos, text, scenes, packaging, and posters.

Main Results:

• Achieved an average clustering accuracy above 82% using the STING algorithm.
• Demonstrated scene element recognition accuracy exceeding 80% with CNNs.
• Attained facial recognition accuracy above 80% and expert-rated reliability and practicality scores averaging around 80 points.

Conclusions:

• Artificial intelligence and big data applications are effective and reliable for VISCOM art design.
• The STING algorithm and CNNs successfully enhance the clustering and identification of design elements.
• This approach enables the creation of art with characteristics tailored to different scenes and demographics.