3D Filters and Statistics¶

3D Filters¶

The EFilters ^.NET ^Python class contains static methods used to apply filters to ZMaps or depth maps.

The EFilters::RemoveNoise ^.NET ^Python method removes outliers from a depth map or a ZMap.

It takes a depth map or a ZMap as input and generates a depth map or a ZMap respectively.

The undefined points are not taken into account.

It is based on a square moving kernel. The size of the kernel is (2 × halfKernelSize + 1) where halfKernelSize is a parameter of the method.
The threshold parameter is scaled with regard to the Z resolution of the filtered depth map or ZMap.
There are 3 variations of this filter, depending on the method parameter:
- ENoiseRemovalMethod_AbsoluteDifferenceFromMean ^.NET ^Python removes a point when it deviates from the average in the neighborhood, including itself. The threshold is an absolute difference.
- ENoiseRemovalMethod_RelativeDifferenceFromMean ^.NET ^Python removes a point when it deviates from the average in the neighborhood, including itself. The threshold is a multiple of the standard deviation.
- ENoiseRemovalMethod_HighStandardDeviation ^.NET ^Python removes a point when the standard deviation in the neighborhood, including itself, is higher than a defined threshold.

The EFilters::Median ^.NET ^Python method removes the outliers from a depth map or a ZMap.

It takes a depth map or a ZMap as input and generates a depth map or a ZMap respectively.
- The undefined points are not taken into account.
There are options for this method:
- You can adjust the width and the height of the median filter.
- The function can ignore the undefined pixels or take them into account (as the lowest possible value) when computing the medians.
  - Taking the undefined pixels into account removes spickles in the middle of undefined pixels and smoothens the edges between defined and undefined pixels.
  - Ignoring them does not. It can also be much slower.

The EStatistics ^.NET ^Python class contains static methods used to compute statistics on ZMaps or depth maps.

The EStatistics::ComputeAverageMap ^.NET ^Python method computes the local average map.

You can use this method as a low-pass filter.
Undefined points are not taken into account.
This method is based on a square moving kernel. The size of the kernel is (2 × halfKernelSize + 1) where halfKernelSize is a parameter of the method.

The EStatistics::ComputeStandardDeviationMap ^.NET ^Python method computes a map of the local standard deviation.

You can use this method to visually determine the threshold value to use with the EFilters::RemoveNoise ^.NET ^Python method when using the ENoiseRemovalMethod_HighStandardDeviation ^.NET ^Python setting.

Note

Be aware, however, that in the generated map, a pixel with the value 0 can either be undefined or have a standard deviation equal to zero.

The EStatistics::ComputePixelStatistics ^.NET ^Python method returns basic statistical information about pixel values:

Use an ERegion ^.NET ^Python object to specify the region of the ZMap or depth map used to compute the statistics.

The EStatistics::ComputeStatistics ^.NET ^Python method returns the same information as the EStatistics::ComputePixelStatistics ^.NET ^Python method, but scaled with respect to the Z resolution.

Use an ERegion ^.NET ^Python object to specify the region of the ZMap or depth map used to compute the statistics.