After classifying the LiDAR point cloud using deep learning, can I remove specific objects?

2 views (last 30 days)
Hi, everyone.
I am new to MATLAB.
This page classifies point cloud data by object.
Is it possible to delete only certain types of objects from the point cloud data?

Answers (1)

Manish
Manish on 29 Nov 2024
Hi Motoki,
I understand that you want to remove a certain type of object from the point cloud after classification using deep learning.
You can achieve this by identifying the label you wish to eliminate.
Then, select all the points corresponding to that label and remove them from the point cloud. Extract the points you want to keep and use the select function to create a new point cloud with only those points.
Here is the sample code below:
% Read the test data and perform semantic segmentation
I = read(imdsTest);
predictedResult = pcsemanticseg(I, net, "Classes", classNames);
% index for "car" in classNames
carLabel = find(strcmp(classNames, "car"));
% points to keep (i.e., not cars)
pointsToKeep = predictedResult ~= carLabel;
filteredPtCloud = select(I, pointsToKeep);% filtered point cloud
figure;
pcshow(filteredPtCloud);
Refer the below documentation links for better understanding:
Hope it Helps!

Categories

Find more on Labeling, Segmentation, and Detection in Help Center and File Exchange

Products


Release

R2022a

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!