CNN-driven Art Design Decision Support System Based on Big Data

Abstract

The aim of this article is to introduce an innovative approach to the visual interpretation of artistic scenes, leveraging the capabilities of Convolutional Neural Networks (CNN) to address the limitations of conventional rendering methods. Towards this objective, we employed advanced deep learning strategies to autonomously analyze and categorize artistic scene imagery by designing and refining CNN architectures. In our experimentation, we deliberately chose illustrative animal and plant images as test subjects to assess the algorithm's proficiency holistically. The findings reveal that our approach offers notable benefits in terms of both rendering swiftness and visual fidelity. In contrast to established practices, our optimized model has achieved a substantial decrease in rendering duration alongside a marked enhancement in visual quality, yielding sharper visuals and more intricate details. In conclusion, CNN-driven visual analysis techniques for artistic scenes have demonstrated considerable worth in elevating rendering efficiency and excellence, presenting fresh insights and utilities for scholars and practitioners alike in related domains.