US 12,223,589 B2Granted

Method of generating a smooth image from point cloud data

OwnerYoldas Askaninventor

Life of this patent

Estimating the term…

Loading patent…

Abstract

A method is explained for processing an array of pixels in a point cloud. Local error bars are calculated for each pixel distance value. A difference is calculated between distance values of the pixel being processed and neighboring pixels with distance values within the error bars. If the difference is outside the error bars, the distance value of the pixel being processed is changed by a small fraction while remaining inside the error bars; if the difference is within the error bars, the pixel value is replaced by a weighted average. The neighboring pixels with distance values within the error bars of the pixel are counted and if a predetermined threshold is met, the counted values are averaged and replace the pixel value, but if not met, the pixel value is unchanged. If loop exit criteria have been met, the loop is terminated and if not, looping begins again.

Claims8

Dependent claims

A system for generating a processed point cloud data set comprising:

a scanner operably configured to scan a surface of an object or a portion thereof with a light source so as to generate a first point cloud data set, the first point cloud data set representing a noisy 3D model of the object and including first distance values corresponding to points on the surface of the object;
and
a processor operably configured to execute instructions for:
receiving the first point cloud data set;
generating a difference between a distance value, among the distance values, in a pixel of the first point cloud dataset being processed and each neighboring pixel or a most suitable neighboring pixel distance value;
generating second distance values from the first point cloud data set, at least one of the second distance values is based on at least two neighboring points;
generating a second point cloud data set based on the second distance values which are based on the distance value of the at least two neighboring points;
generating said second distance values from the first point cloud data set, at least one of the second distance values is an average distance value or a weighted average distance value of the at least two neighboring points;
performing one or more iterations to generate distance values so as to generate the second point cloud data set representing a noise free and smoothed representation of the noisy 3D model of the scanned object, each of the iteratively generated distance values is based on the distance value of at least two neighboring points of each of the distance values of the previous iteration;
and
outputting the processed point cloud data set to any one of a monitor for display, or a file for storage, or a printer.

8 more elements

A method for generating a processed point cloud data set comprising:

scanning, with a scanner, an object to generate a first point cloud data set, the first point cloud data set representing a noisy 3D model of the scanned object and including first distance values corresponding to points on the surface of the object;
and
receiving, by a processor, the first point cloud data set;
generating a difference between a distance value, among the distance values, in a pixel of the first point cloud dataset being processed and each neighboring pixel or a most suitable neighboring pixel distance value;
generating, by the processor, second distance values from the first point cloud data set, at least one of the second distance values is based on at least two neighboring points;
generating, by the processor, second point cloud data set based on the second distance values which are based on the distance value of the at least two neighboring points;
generating, by the processor, second distance values from the first point cloud data set, at least one of the said second distance values is an average distance value or a weighted average distance value of the at least two neighboring points;
performing, by the processor, one or more iterations to generate distance values so as to generate the second point cloud data set representing a noise free and smoothed representation of the noisy 3D model of the scanned object, each of the iteratively generated distance values is based on the distance value of at least two neighboring points of each of the distance values of the previous iteration;
and
outputting, by the processor, the processed point cloud data set to any one of a monitor for display, or a file for storage, or a printer.

7 more elements

Dependencies are inferred from the claim text.

Facts

Application: 16/006,534Filed 12 Jun 2018 · continuation in part
Priority: 25 Jun 2012
Published: US 2019/0043250 A17 Feb 2019
Granted: 11 Feb 2025
Term ends: Not estimatedThe source records don’t support an estimate.

Classification4 codes

G06T 17/00
Three-dimensional [3d] modelling for computer graphics
G06T 2200/08
Indexing scheme for image data processing or generation, in general: involving all processing steps from image acquisition to 3D model generation
G06T 2210/56
Indexing scheme for image generation or computer graphics: particle system, point based geometry or rendering
G06T
Image data processing or generation, in general

Family

Application13/532,691Parent of this continuation-in-part · filed 25 Jun 2012
Application14/166,840Parent of this continuation · filed 28 Jan 2014
US 8,705,110 B2Parent of this continuation-in-part · filed 22 Apr 2014
Application15/043,492Parent of this continuation-in-part · filed 12 Feb 2016
US 9,300,841 B2Parent of this continuation · filed 29 Mar 2016
US 10,032,255 B2Parent of this continuation-in-part · filed 24 Jul 2018
US 2019/0043250 A1Pre-grant publication · published 7 Feb 2019
Application15/043,492Parent of this continuation
Application14/166,840Parent of this continuation-in-part
US 2013/0342877 A1Published application
US 2014/0313527 A1Published application
US 2016/0292829 A1Published application
US 12,223,589 B2This patent

From the US filing history. Foreign family members aren’t listed.

Timeline

Citations

Open elsewhere

Building an agent or pulling many patents? Use the API or MCP server instead of scraping pages. Get an API key