Artificial Neural Network Model for Indoor Decoration Intelligence Calculation and Automation Design

Area of Science:

• Computer Science
• Artificial Intelligence
• Architecture

Background:

• The indoor decoration industry is evolving towards mechanization, specialization, and intelligence.
• Previous research provides a foundation for developing advanced design tools.

Purpose of the Study:

• To propose an artificial neural network (ANN) model for intelligent indoor decoration calculation and automation.
• To develop an automated design system leveraging ANNs and digital image processing.

Main Methods:

• Utilizing digital image processing for automatic identification and preprocessing of apartment floor plans.
• Establishing an ANN-based automated design model trained on an indoor decoration dataset.
• Analyzing ANN network structure and training processes.

Main Results:

• The ANN model demonstrated a significant reduction in Mean Absolute Error (MAE) and Mean Squared Error (MSE) with increased training.
• The model achieved a small, stabilized error, indicating high accuracy.
• ANN algorithms generated multiple layout schemes within one minute, proving efficient and reasonable.

Conclusions:

• ANN-based automation design is superior for indoor decoration.
• The proposed model offers efficient, functional, and user-centric design solutions, saving time and resources.
• The system provides users with diverse, optimized layout choices meeting all functional requirements.