NormalEstimationIntegralImage¶
What is this?¶

Compute normals for an organized point cloud using integral images.
Subscribing Topics¶
~input
(sensor_msgs/PointCloud2
)Input point cloud. (point type is
pcl::PointXYZRGB
)
Publishing Topics¶
~output
(sensor_msgs/PointCloud2
)Output normals. (point type is
pcl::Normal
)~output_with_xyz
(sensor_msgs/PointCloud2
)Output point cloud with normal field. (point type is
pcl::PointXYZRGBNormal
)
Parameters¶
max_depth_change_factor
(Double, default: 0.02)max depth change factor
normal_smoothing_size
(Double, default: 20.0)normal smoothing size parameter
estimation_method
(Int, default: 1)Estimation method.
0: AVERAGE_3D_GRADIENT
1: COVARIANCE_MATRIX
2: AVERAGE_DEPTH_CHANGE
depth_dependent_smoothing
(Boolean, default: false)Use depth dependent smoothing.
border_policy_ignore
(Boolean, default: true)Ignore border policy.
Parameters above can be changed by dynamic_reconfigure
.
Sample¶
roslaunch jsk_pcl_ros sample_normal_estimation_integral_image.launch