pcdenoise

Remove noise from 3-D point cloud

Syntax

ptCloudOut = pcdenoise(ptCloudIn)

[ptCloudOut,inlierIndices,outlierIndices] = pcdenoise(ptCloudIn)

[ptCloudOut,___] = pcdenoise(___Name=Value)

Description

ptCloudOut = pcdenoise(ptCloudIn) returns a filtered point cloud that removes outliers.

[ptCloudOut,inlierIndices,outlierIndices] = pcdenoise(ptCloudIn) additionally returns the linear indices to the points that are identified as inliers and outliers.

[ptCloudOut,___] = pcdenoise(___Name=Value) specifies a name-value argument in addition to any combination of arguments from previous syntaxes. For example, Threshold=1.0 sets the outlier threshold to 1.0.

Examples

Remove Outliers from Noisy Point Cloud

Create a plane point cloud.

gv = 0:0.01:1;
[X,Y] = meshgrid(gv,gv);
ptCloud = pointCloud([X(:),Y(:),0.5*ones(numel(X),1)]);

figure
pcshow(ptCloud);
title('Original Data');

Add uniformly distributed random noise.

noise = rand(500, 3);
ptCloudA = pointCloud([ptCloud.Location; noise]);

figure
pcshow(ptCloudA);
title('Noisy Data');

Remove outliers.

ptCloudB = pcdenoise(ptCloudA);

figure;
pcshow(ptCloudB);
title('Denoised Data');

Preserve Organized Structure of Point Cloud After Denoising

Load an organized point cloud from a saved MAT-file.

ld = load('drivingLidarPoints.mat');
orgPtCloud = ld.ptCloud;

Remove noise from the point cloud.

orgPtCloudOut = pcdenoise(orgPtCloud,PreserveStructure=true);

Input Arguments

ptCloudIn — Point cloud

pointCloud object

Point cloud, specified as a pointCloud object.

Name-Value Arguments

Specify optional pairs of arguments as Name1=Value1,...,NameN=ValueN, where Name is the argument name and Value is the corresponding value. Name-value arguments must appear after other arguments, but the order of the pairs does not matter.

Before R2021a, use commas to separate each name and value, and enclose Name in quotes.

Example: Threshold=1.0 sets the threshold to 1.0.

NumNeighbors — Number of nearest neighbor points

4 (default) | positive integer

Number of nearest neighbor points, specified as the comma-separated pair consisting of 'NumNeighbors' and a positive integer in pixels. The value is used to estimate the mean of the average distance to neighbors of all points. Decreasing this value makes the filter more sensitive to noise. Increasing this value increases the number of computations.

Data Types: single | double

Threshold — Outlier threshold

1.0 (default) | positive scalar

Outlier threshold, specified as the comma-separated pair consisting of 'Threshold' and a positive scalar. By default, the threshold is one standard deviation from the mean of the average distance to neighbors of all points. A point is considered to be an outlier if the average distance to its k-nearest neighbors is above the specified threshold.

Data Types: single | double

PreserveStructure — Preserve the organized structure of a point cloud

false (default) | true

Preserve the organized structure of a point cloud as an M-by-N-by-3 location matrix, specified as true or false. The table describes the point cloud structure according to the value of PreserveStructure.

PreserveStructure	The function returns
true	An organized, denoised, point cloud. The Location property that describes the structure of the point cloud, contains an M-by-N-by-3 matrix. Points that are not selected in the denoised point cloud are filled with NaN, and the corresponding color is set to [0 0 0]. To return an organized point cloud, the input must be an organized point cloud.
false	An unorganized, denoised, point cloud. The Location property that describes the structure of the point cloud, contains an M-by-3 matrix.

Output Arguments

ptCloudOut — Filtered point cloud

pointCloud object

Filtered point cloud, returned as a pointCloud object.

inlierIndices — Linear index of inlier points

1-by-N vector

Linear index of inlier points, returned as a 1-by-N vector.

Data Types: uint32

outlierIndices — Linear index of outlier points

1-by-N vector

Linear index of outlier points, returned as a 1-by-N vector.

Data Types: uint32

References

[1] Rusu, R. B., Z. C. Marton, N. Blodow, M. Dolha, and M. Beetz. “Towards 3D Point Cloud Based Object Maps for Household Environments”. Robotics and Autonomous Systems Journal. 2008.

Extended Capabilities

C/C++ Code Generation
Generate C and C++ code using MATLAB® Coder™.

GPU Code Generation

Generate CUDA® code for NVIDIA® GPUs using GPU Coder™.

• The PreserveStructure name-value argument is not supported for GPU code generation.

Version History

Introduced in R2015a