Indoor Environment 3D Space Design Based on 3D Modeling and Image Processing

Abstract

Aim: 3D vision technology-based research is suggested to model indoor environmental sensory design because machine vision-based indoor environmental sensory design models aren’t very good at getting rid of objects. Methodology: First, the three-dimensional spatial points are built inside the room. Then, they use the Euclidean distance to figure out the vector relationships between these points, which results in the matching of three-dimensional spatial points. Subsequent to the above processes, the author utilizes the Elastic Fusion algorithm to construct an indoor environmental sensory design model. This concludes the modeling of indoor environmental sensory design based on 3D vision. Experimental analysis: Ten spatial points from the same scene within the database are selected, and both models are used to independently calculate the depth values of these ten spatial points. The experimental results demonstrate that the depth values of the ten spatial points obtained by the proposed model are all 0, meeting the removal requirements. Conclusion: In contrast, when using the machine vision-based indoor environmental perception design model, all seven out of the seven spatial points yield depth values greater than 0, failing to meet the removal requirements. This substantiates that the proposed model exhibits high accuracy in object removal.