Research on Garden Space Layout Optimization Algorithm Based on Deep Learning

This paper studies a garden space layout optimization algorithm based on deep learning, aiming to improve the scientific and optimization efficiency of garden design. The system uses GAN as the layout generation model and combines MOGA to achieve layout optimization. The system architecture includes data acquisition and processing, layout generation and optimization decision modules. The experiment was verified on real garden design data and public terrain database. The results show that compared with traditional heuristic algorithms and rule-based optimization methods, the optimization effect of deep learning models is significantly improved. The ecological suitability score of the optimization scheme increased by 25.7<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">%</sup>, the aesthetic score increased by an average of 18.9%, and the calculation time was reduced by 21.4%. The simulation test further verified the adaptability and stability of the model under different scene scales and complexities. This study provides an effective new method for the intelligentization of garden design, and will further explore the fusion of multimodal data and the development of dynamic optimization systems in the future.