Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Sign point hole filling method based on neural network in tri-D scanning point cloud

A technology of 3D scanning and neural network, which is applied to the field of filling holes in 3D scanning point clouds based on neural networks. , The point cloud features are well represented and the speed is fast

Inactive Publication Date: 2009-11-18
海安江理工技术转移中心有限公司 +1
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This type of method has a wide range of applications, and it has better results for nonlinear data, noisy data, and data with unclear pattern characteristics, but it has shortcomings: many operations, and each hole must be artificially selected. Sample point set; the density of points in the repaired area is inconsistent with the surrounding; for some feature areas with large curvature changes, the feature performance is not good enough

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
  • Sign point hole filling method based on neural network in tri-D scanning point cloud
  • Sign point hole filling method based on neural network in tri-D scanning point cloud
  • Sign point hole filling method based on neural network in tri-D scanning point cloud

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0035] A point cloud hole filling method for 3D scanning based on neural network:

[0036] Step 1: For the hole formed by the marker points pasted on the surface of the head of Venus, in the 3D scanning point cloud, when obtaining the sample point set around the hole, according to the 3D coordinates of the marker points (x b ,y b ,z b ), take the data points in the cube centered on the marker point as the sample point P for hole filling s (s=0,1,...,t), t is the number of sample points, and the 8 vertices of the cube are (x b -r,y b -r,z b -r), (x b -r,y b -r,z b +r), (x b -r,y b + r, z b -r), (x b -r,y b + r, z b +r), (x b +r,y b -r,z b -r), (x b +r,y b -r,z b +r), (x b +r,y b + r, z b -r), (x b +r,y b + r, z b +r), where r is the selected domain value, in order to obtain a suitable sample point set, take 1.2 to 1.5 times the radius of the marker point;

[0037] Step 2: According to the sample point data P s , training the neural network for filling...

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 point cloud hole filling method based on neural network-based three-dimensional scanning, which can ensure that the filled hole data is continuous with the surrounding data and has better point cloud feature representation. The invention has the advantage of simple method. The invention is mainly used in the application occasion of filling holes of various complex curved shapes generated by marker points in a three-dimensional scanning system. Using the neural network method in the present invention, the network for filling the hole can be obtained, and then according to the density of the boundary points of the hole, sampling points are taken in the hole area, and the point of filling the hole is further adjusted according to the curvature of the point, so as to achieve smooth filling of the hole.

Description

technical field [0001] The invention relates to a method for repairing three-dimensional graphics, in particular to a method for filling holes in marker points based on neural networks in a three-dimensional scanning point cloud. Background technique [0002] Reverse Engineering (Reverse Engineering, RE) technology is a new technology that appeared in the field of advanced manufacturing in the late 1980s. It generally includes four basic links: three-dimensional shape detection and conversion (obtaining physical data), data preprocessing (point Cloud processing, recognition, multi-view splicing), CAD model establishment (surface reconstruction), CAM part forming, the basic flow chart is as follows figure 1 shown. In the process of 3D shape detection and conversion, the 3D digital scanner is used to quickly scan and measure the surface of the physical model. On the premise of satisfying the discrete sampling speed and data quality, the 3D discrete data of the product is obta...

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 Patents(China)
IPC IPC(8): G06T11/00G06N3/02
Inventor 达飞鹏谷继兵盖绍彦朱正键杨伟光
Owner 海安江理工技术转移中心有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products