Wall reconstruction method and device

A wall and algorithm technology, applied in the field of wall reconstruction methods and devices, can solve the problems of strict requirements for drawing methods and field definitions, etc.

Pending Publication Date: 2019-05-21
DONG YI RI SHENG HOME DECORATION GRP
View PDF7 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

These methods either require manual assistance, or have relatively strict requirements on drawing methods and field definitions. At the same time, any traditional algorithm can only deal with very clear and structured raster images. For example, most of the image styles exported from CAD are white backgrounds. , a picture with a black foreground

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Wall reconstruction method and device
  • Wall reconstruction method and device
  • Wall reconstruction method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0041] like figure 1 As shown, the present embodiment proposes a wall reconstruction method, the method comprising:

[0042] S101. Acquire a grid floor plan, and perform semantic segmentation on the grid floor map to obtain a semantic segmentation result.

[0043] Specifically, such as figure 2 Shown is the acquired grid floor plan. In this embodiment, a neural network structure is used to carry out semantic segmentation processing on the architectural grid floor plan, preferably using a U-shaped convolutional neural network. By cleaning and screening the training data required by the network, configuring the network learning and parameter generation, and finally generating semantic segmentation results, such as image 3 shown. Of course, other types of neural network structures based on deep learning methods can also realize pixel-level representation and learning of raster floor plans. But in terms of overall effect, it is still best to use U-shaped convolutional neur...

corresponding Embodiment 1

[0058] Corresponding to Embodiment 1, this embodiment also proposes a wall reconstruction device, which includes a processor configured with operating instructions executable by the processor to perform the following operations:

[0059] Obtain the grid floor plan;

[0060] Semantic segmentation is performed on the grid floor plan to obtain a semantic segmentation result;

[0061] Perform image post-processing on the semantic segmentation result to obtain wall nodes;

[0062] Obtaining an effective and correct connection result according to the wall nodes;

[0063] The walls are reconstructed according to the effective and correct connection results.

[0064] Specifically, this embodiment uses a neural network structure to perform semantic segmentation processing on the building grid floor plan, preferably using a U-shaped convolutional neural network. By cleaning and screening the training data required by the network, configuring the network learning and parameter generat...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention provides a wall reconstruction method. The method comprises: acquiring a grid house type graph; performing semantic segmentation on the grid house type image to obtain a semantic segmentation result; Performing image post-processing on the semantic segmentation result to obtain wall nodes; Obtaining an effective correct connection result according to the wall nodes; And reconstructing the wall body according to the effective correct connection result. The invention further provides a wall body reconstruction device correspondingly. The method comprises the following steps: firstly, obtaining wall body related pixels through semantic segmentation, and then performing image post-processing to obtain wall body nodes; and finally, the wall nodes are connected to obtain wall data,so that identification and reconstruction of the house type map wall are rapidly realized. And quick end-to-end intelligent home decoration service is realized.

Description

technical field [0001] The present application relates to the field of home decoration, and in particular to a wall reconstruction method and device. Background technique [0002] In the link of home decoration design, with the development of CAD / CAM technology, designers can use computer-aided drawing software to draw architectural floor plans, and then use related software to carry out 3D reconstruction and other designs, which brings great benefits to the three-dimensional realization of architectural design. Great convenience. However, in these design processes based on vector graphics, designers need to perform a lot of manual processing on the design drawings, which is time-consuming and labor-intensive. Moreover, this kind of software is generally used by architectural designers, and the participation value provided by ordinary people is of little significance. we know, vector Figure 1 It generally requires designers to spend a certain amount of time to complete th...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06T17/10G06K9/34G06K9/46G06N3/04
Inventor 朱燕朱程
Owner DONG YI RI SHENG HOME DECORATION GRP
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap