Landscape Planning and Image Analysis Based on Multipopulation Coevolution Particle Swarm Radial Basis Function...

Area of Science:

• Urban planning and design
• Artificial intelligence in landscape architecture
• Computational intelligence for environmental design

Background:

• Urban landscape planning significantly impacts urban development and living environments.
• Current landscape planning faces issues like low data reuse, design-reality discrepancies, and professional shortages.
• Artificial neural networks offer potential solutions to these landscape planning challenges.

Purpose of the Study:

• To propose and evaluate an artificial neural network algorithm for garden planning and design.
• To enhance the accuracy and applicability of urban landscape planning methods.
• To address limitations in current urban landscape design practices.

Main Methods:

• Development of a multipopulation coevolution particle swarm radial basis function neural network algorithm.
• Simulation experiments for evaluating garden planning and design accuracy.
• Plant configuration simulation to test urban planning and design adjustments.

Main Results:

• The proposed algorithm achieved a prediction error of less than 5% in simulation experiments.
• Demonstrated good accuracy and generalization ability in performance evaluations.
• Effectively evaluated urban planning and design, proposing relevant adjustment schemes.

Conclusions:

• The multipopulation coevolution particle swarm radial basis function neural network algorithm shows high accuracy and generalization for urban landscape planning.
• The method provides effective solutions for garden planning and urban design, aligning with practical needs.
• This approach can improve data reuse and bridge the gap between design schemes and actual urban situations.