Urban virtual environment landscape design and system based on PSO-BP neural network

Area of Science:

• Computer Science
• Urban Planning
• Digital Art

Background:

• Conventional landscape design is challenged by virtual landscapes powered by neural networks.
• Virtual landscapes shift design from the physical to the digital realm, altering creation and expression methods.
• The immersive and interactive nature of virtual landscapes offers new possibilities for scenic experiences.

Purpose of the Study:

• To investigate the impact of neural network technologies on conventional landscape design.
• To explore the potential of virtual landscapes in urban development and cultural heritage.
• To assess the effectiveness of a PSO-BP neural network in improving urban environment landscape design.

Main Methods:

• Utilizing PSO-BP neural network for urban environment landscape design.
• Conducting experimental research and data analysis on landscape improvements.
• Employing user satisfaction surveys before and after design interventions.

Main Results:

• Before neural network application, 57% were satisfied and 17% dissatisfied with Yanta East Garden.
• After improvement, satisfaction rose to 67% with only 5% dissatisfaction, a 10% increase.
• User feedback highlighted improved historical ambiance and cultural expression through landscape elements.

Conclusions:

• Neural network-based methods are highly suitable for urban landscape design enhancement.
• Virtual landscapes offer innovative approaches for digital cities and heritage sites.
• This research bridges the gap between digital technology, urban planning, and cultural preservation.