Application of 3D Image Reconstruction on Landscape Architecture in Environmental Design System

Computer software assistance is widely used in the field of design due to its fast response and convenience in current environmental design. Combining with semantic segmentation technology for efficient planning of three-dimensional landscape environment can effectively improve the visual effect of landscape environment planning and design. The traditional method first calculates the edge location of the environment image layout and denoises the image layout information, but ignores the filtering processing, resulting in incomplete image layout information. Aiming at the above problems, this article constructs a model of environmental image character understanding and semantic expression based on deep learning (DL), and applies it to CAD modeling of environmental design to improve the accuracy of landscape architecture 3D image reconstruction. The research results indicate that this method has good test results in landscape design of 3D images, and has obvious advantages over traditional modeling methods in operational efficiency, which provides an important theoretical basis for the analysis of landscape architecture distribution rationality. The proposed method can effectively improve the integrity of three-dimensional environmental image layout, and provide theoretical and technical basis for the construction of environmental design system. © 2024, CAD Solutions, LLC. All rights reserved.