fit cuboid for for sequence of pcd data

2 views (last 30 days)
Pavan Kumar B N
Pavan Kumar B N on 21 Jun 2021
Commented: Elad Kivelevitch on 22 Jun 2021
Hello All,
I am using this openExample('lidar/CuboidModelExample') example for fitting cuboid to non-ground objects. Can anyone please advice how to use it for a pcap file (streaming data)?
fileName = 'test.pcap';
deviceModel = 'VLP16';
veloReader = velodyneFileReader(fileName, deviceModel);
% Read a scan of lidar data
ptCloud = readFrame(veloReader) %#ok<NOPTS>
roi = [0 40 -20 30 -7 7];
in = findPointsInROI(ptCloud,roi);
ptCloudIn = select(ptCloud,in);
hcluster = figure;
panel = uipanel('Parent',hcluster,'BackgroundColor',[0 0 0]);
ax = axes('Parent',panel,'Color',[0 0 0]);
pcshow(ptCloudIn,'MarkerSize',30,'Parent',ax)
title('Input Point Cloud')
maxDistance = 0.3;
referenceVector = [0 0 1];
[~,inliers,outliers] = pcfitplane(ptCloudIn,maxDistance,referenceVector);
ptCloudWithoutGround = select(ptCloudIn,outliers,'OutputSize','full');
hSegment = figure;
panel = uipanel('Parent',hSegment,'BackgroundColor',[0 0 0]);
ax = axes('Parent',panel,'Color',[0 0 0]);
pcshowpair(ptCloudIn,ptCloudWithoutGround,'Parent',ax)
legend('Ground Region','Non-Ground Region','TextColor', [1 1 1])
title('Segmented Ground Plane')
distThreshold = 1;
[labels,numClusters] = pcsegdist(ptCloudWithoutGround,distThreshold);
labelColorIndex = labels;
hCuboid = figure;
panel = uipanel('Parent',hCuboid,'BackgroundColor',[0 0 0]);
ax = axes('Parent',panel,'Color',[0 0 0]);
pcshow(ptCloudIn.Location,labelColorIndex,'Parent',ax)
title('Fitting Bounding Boxes')
hold on
for i = 1:numClusters
idx = find(labels == i);
model = pcfitcuboid(ptCloudWithoutGround,idx);
plot(model)
end
  1 Comment
Elad Kivelevitch
Elad Kivelevitch on 22 Jun 2021
Removed Sensor Fusion and Tracking Toolbox from the list of products as it is not relevant.
For your informatio the example (and all lidar processing) is from Lidar Toolbox.

Sign in to comment.

Answers (0)

Community Treasure Hunt

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

Start Hunting!