Evaluating Aesthetic Preferences for Indoor Landscapes from a Cybersecurity Perspective Using Machine

Introduction: With the acceleration of modern urbanization, indoor landscape design has not only become an important element to improve the quality of life, but also closely related to network security. However, how to scientifically evaluate the aesthetic preferences of indoor landscapes while ensuring network security has become a question worth exploring. Objectives: In response to the subjectivity and low efficiency of traditional aesthetic evaluation methods, this study proposes a deep learning based indoor landscape aesthetic quality evaluation method. Methods: This method combines convolutional neural networks and graph neural networks to extract and analyze global and local aesthetic features of indoor images, taking into account the impact of indoor layout on network security. Results: The results showed that the aesthetic evaluation accuracy of this method on indoor landscape datasets reached 97.74%, an increase of 7.54 percentage points compared to traditional methods. Compared with other aesthetic evaluation schemes, this method achieved a 14.21% higher aesthetic score and a 10.6 point improvement in functional evaluation. Conclusions: The conclusion indicates that this method can effectively improve the objectivity and efficiency of indoor landscape aesthetic evaluation, providing a novel evaluation tool for indoor landscape design under the premise of ensuring network security.