Process Surface by Filter
Description
This Step is used to preprocess the surface with a specific filter to obtain a better surface.
Workflow
The process of configuring this Step is shown below.
-
Configure the input. Connect the ports manually or select the input(s) under Input in the parameter configuration panel.
-
Determine whether to use the feature region, select the Filter Type, and set the filter parameters.
-
Select the output item Surface Data (selected by default).
Parameter Description
| Parameter | Description |
|---|---|
Show Advanced Filters |
Once this parameter is selected, advanced filters will be listed in the drop-down menu of Filter Type. |
Use Feature Region |
Use feature region(s) to define the region where surface preprocessing will be performed. If it is not selected, the Step will process the entire surface data. If it is selected, it is necessary to set the following parameters:
|
Filter Type |
Select a filter to process the surface data. Options: The available filters are listed below. See Filter Type for detailed information.
|
Output Description
The output of this Step is processed surface data that can be used as input to other Steps.
Appendix
Filter Type
-
Gap filling:
With the maximum or minimum Z values of the nearest neighbors or linear interpolation between neighboring values, this filter fills in missing data in the specified window.
Parameter Description Gap Filling Type
This parameter is used to select the method to fill the gaps.
-
Min Z point filling: Use the minimum Z value of the gap endpoints to fill the gap.
-
Max Z point filling: Use the maximum Z value of the gap endpoints to fill the gap.
-
Linear interpolation filling: Use Z values of the gap endpoints to perform linear interpolation and thus fill the gap.
Image for Gap Filling
This parameter is used to select the type of image for gap filling.
Value list: Depth map, intensity image, depth map and intensity image
Filter in X Direction
This parameter is used to determine whether to filter the surface data in the X-direction.
X-Direction Window Size
This parameter specifies the window size for X-direction filtering. The default value is 3.
Filter in Y Direction
This parameter is used to determine whether to filter the surface data in the Y-direction.
Y-Direction Window Size
This parameter specifies the window size for Y-direction filtering. The default value is 3.
-
-
Mean:
The mean filter can smooth the image by calculating the neighbors of pixels in the image and replacing the pixel values in the original image with the calculated average.
Parameter Description Filter in X Direction
This parameter is used to determine whether to filter the surface data in the X-direction.
X-Direction Window Size
This parameter specifies the window size for X-direction filtering. The default value is 3.
Filter in Y Direction
This parameter is used to determine whether to filter the surface data in the Y-direction.
Y-Direction Window Size
This parameter specifies the window size for Y-direction filtering. The default value is 3.
-
Median:
The median filter reduces noise in images, especially salt and pepper noise, by replacing each pixel in the image with the median of its neighboring pixels.
Parameter Description Filter in X Direction
This parameter is used to determine whether to filter the surface data in the X-direction.
X-Direction Window Size
This parameter specifies the window size for X-direction filtering. The default value is 3.
Filter in Y Direction
This parameter is used to determine whether to filter the surface data in the Y-direction.
Y-Direction Window Size
This parameter specifies the window size for Y-direction filtering. The default value is 3.
-
Decimation:
The decimation filter can reduce the image resolution. During decimation, some pixels in the raw image will be discarded.
Parameter Description Filter in X Direction
This parameter is used to determine whether to filter the surface data in the X-direction.
X-Direction Window Size
This parameter specifies the window size for X-direction filtering. The default value is 3.
Filter in Y Direction
This parameter is used to determine whether to filter the surface data in the Y-direction.
Y-Direction Window Size
This parameter specifies the window size for Y-direction filtering. The default value is 3.
-
Dilation:
This type of filter can be used to remove small holes and smooth object edges to make the image more complete.
Note that the dilation operation may magnify the noise in the image. When performing dilation on images, you should set a proper kernel size and symmetry to ensure that the dilation operation can achieve the desired result.
Parameter Description Use Intensity Image
When selected, the Step processes the intensity image; when not selected, it processes the depth map.
Keep Original Pixel Value
When selected, this parameter ensures that the original pixel values in non-edge areas of the image are not modified or lost during the dilation operation. It is not checked by default.
Kernel Size
This parameter is used to set the kernel size during dilation.
Symmetry
This parameter is used to set the applied method for dilation.
-
Horizontal symmetry: Perform dilation on the image horizontally by the set dilation kernel.
-
Vertical symmetry: Perform dilation on the image vertically by the set dilation kernel.
-
Central symmetry: Perform dilation on the image horizontally and vertically by a square dilation kernel of set size.
-
-
Erosion:
This type of filter can be used to remove small objects or noises in the image to make the image clearer.
Note that the erosion operation may lead to loss of image details. When eroding images, set appropriate kernel size and symmetry.
Parameter Description Use Intensity Image
When selected, the Step processes the intensity image; when not selected, it processes the depth map.
Keep Original Pixel Value
When selected, this parameter ensures that the original pixel values in non-edge areas of the image are not modified or lost during the dilation operation. It is not checked by default.
Kernel Size
This parameter is used to set the kernel size during erosion.
Symmetry
This parameter is used to set the applied method for erosion.
-
Horizontal Symmetry: Perform erosion on the image horizontally by the set erosion kernel.
-
Vertical Symmetry: Perform erosion on the image vertically by the set erosion kernel.
-
Central symmetry: Perform erosion on the image horizontally and vertically by a square erosion kernel of set size.
-
-
Opening:
Opening is essentially erosion followed by dilation. This type of filter can be used to remove the noise in the details of the image while preserving the main features of the image.
Parameter Description Use Intensity Image
When selected, the Step processes the intensity image; when not selected, it processes the depth map.
Keep Original Pixel Value
When selected, this parameter ensures that the original pixel values in non-edge areas of the image are not modified or lost during the dilation operation. It is not checked by default.
Kernel Size
This parameter is used to set the kernel size for opening.
-
Closing:
Closing is essentially dilation followed by erosion. This type of filter can be used to fill small holes in objects and smooth edges without significantly altering their areas. Secondly, the closing operation can effectively improve the issue of uneven object edges caused by noise.
Parameter Description Use Intensity Image
When selected, the Step processes the intensity image; when not selected, it processes the depth map.
Keep Original Pixel Value
When selected, this parameter ensures that the original pixel values in non-edge areas of the image are not modified or lost during the dilation operation. It is not checked by default.
Kernel Size
This parameter is used to set the kernel size for closing.
-
Morph gradient:
Morphological gradient is an operation in image morphological processing. It highlights the edge information in an image with the difference between dilation and erosion. This type of filter can enhance the object edges in the image and make the edges more prominent.
Parameter Description Use Intensity Image
When selected, the Step processes the intensity image; when not selected, it processes the depth map.
Keep Original Pixel Value
When selected, this parameter ensures that the original pixel values in non-edge areas of the image are not modified or lost during the dilation operation. It is not checked by default.
Kernel Size
This parameter is used to set the kernel size for gradient operation. The default value is 3.
-
Gaussian:
Gaussian filter can remove a certain amount of noise in the image. This type of filter can be used to effectively smooth the object edges while preserving the edges and details of the image. Therefore, it is usually used for image smoothing, removing high-frequency noise, and preprocessing before edge detection.
Parameter Description Use Intensity Image
When selected, the Step processes the intensity image; when not selected, it processes the depth map.
Kernel Size
This parameter is used to set the kernel size for Gaussian filtering. The default value is 3.
-
Sobel:
Sobel is capable of performing operations within a certain range around pixels and is often used for image edge detection.
Parameter Description Use Intensity Image
When selected, the Step processes the intensity image; when not selected, it processes the depth map.
Kernel Size
This parameter is used to set the kernel size during edge detection. The default value is 3.
Symmetry
This parameter is used to set the applied method for edge detection.
-
Horizontal Symmetry: Apply the set kernel on the image horizontally to detect edges.
-
Vertical Symmetry: Apply the set kernel on the image vertically to detect edges.
-
Central symmetry: Perform edge detection on the image horizontally and vertically by a square kernel of set size.
Min Threshold
After filtering, points with Z values greater than this threshold will be considered edge points.
Max Threshold
Limit the Z values of the filtered points to this threshold or below.
-
-
Laplacian:
Laplacian can effectively detect edge and regions where grayscale changes abruptly in images.
Parameter Description Use Intensity Image
When selected, the Step processes the intensity image; when not selected, it processes the depth map.
Kernel Size
This parameter is used to set the kernel size during edge detection. The default value is 3.
Min Threshold
After filtering, points with Z values greater than this threshold will be considered edge points.
Max Threshold
This parameter is used to limit the Z values of the filtered points to this threshold or below.
-
Negative:
During image processing, each pixel of the image can undergo color inversion or depth value inversion to achieve a visual effect similar to that of a negative film.
Parameter Description Use Intensity Image
When selected, the Step processes the intensity image; when not selected, it processes the depth map.
Include Null Points
This parameter is used to determine whether to include null points in the processed image. Null points are points without depth or intensity information, which are often the result of data loss or inappropriate settings of the feature region(s).
Once this parameter is selected, set Null Filling Value of Depth.
Null Filling Value of Depth
Set the depth value for filling null points. It is visible only when Null Filling Value of Depth is selected.
-
Normalization:
Normalization is a method for enhancing image contrast, which can improve the local contrast of an image without affecting the overall contrast.
Parameter Description Use Intensity Image
When selected, the Step processes the intensity image; when not selected, it processes the depth map.
Min Threshold
The pixel values of points will be normalized to the range of Min Threshold to Max Threshold.
Max Threshold
The pixel values of points will be normalized to the range of Min Threshold to Max Threshold.
-
Binarization:
This type of filter sets the depth values of points in the image to custom values according to the height threshold.
Parameter Description Use Intensity Image
When selected, the Step processes the intensity image; when not selected, it processes the depth map.
Min Threshold
When the actual depth value of a data point is less than or equal to the set Depth Threshold, set the depth value of the point to the value of Min Threshold.
Max Threshold
When the actual depth value of a data point is above the set Depth Threshold, set the depth value of the point to the value of Max Threshold.
Depth Threshold
Depth threshold for image binarization. If the actual depth value of a data point is greater than or smaller than that of Depth Threshold, reset the depth value of the data point to that of Max Threshold or Min Threshold, respectively.
-
Percentile:
By setting Low Threshold Percentile and High Threshold Percentile, you can remove points that do not belong to this percentile range.
Parameter Description Use Intensity Image
When selected, the Step processes the intensity image; when not selected, it processes the depth map.
Low Threshold Percentile
Points with depth or intensity values within the range from the low threshold percentile to the high threshold percentile will be retained.
High Threshold Percentile
Points with depth or intensity values within the range from the low threshold percentile to the high threshold percentile will be retained.
-
Relative threshold:
Set the low and high thresholds to determine the height range of retained data points. When the Use Feature Region option is selected, the height range is relative to the mean height of data points defined by the feature region.
Parameter Description Use Intensity Image
When selected, the Step processes the intensity image; when not selected, it processes the depth map.
Low Threshold
Points with height (or intensity) or relative height (or intensity) within the range of the low threshold to the high threshold will be retained.
High Threshold
Points with height (or intensity) or relative height (or intensity) within the range of the low threshold to the high threshold will be retained.
-
Crop:
This type of filter can remove unqualified data in the feature region.
Parameter Description Output as Single Image
By default, this option is selected, indicating that the original positions of the cropped area will be retained and merged into a single output image. When selected, multiple feature regions will each output a separate image for each cropped area.
Troubleshooting
|
CV-W2001
Error: The selected filter type is invalid.
Possible cause: The selected filter type was not Negative, Normalization, or Binarization.
Solution: Make sure the selected filter type is Negative, Normalization, or Binarization.
CV-W2301
Error: The “Kernel Size” value of the selected filter should exceed 0.
Possible cause: The “Kernel Size” value is less than or equal to 0.
Solution: Make sure the parameter value is greater than 0.
CV-W2302
Error: The selected filter type is invalid.
Possible cause: The selected filter type was not Gaussian, Sobel, or Laplacian.
Solution: Make sure the selected filter type is Gaussian, Sobel, or Laplacian.
CV-W2401
Error: The X-direction window size or Y-direction window size should exceed 0.
Possible cause: The parameter value is less than or equal to 0.
Solution: Make sure the parameter value is greater than 0.
CV-W2402
Error: The selected gap filling type is invalid.
Possible cause: The gap filling type was not selected properly.
Solution: Select Gap Filling Type again in the drop-down list.
CV-W2403
Error: The set “Image for Gap Filling” is invalid.
Solution: Select a valid image type for gap filling in the drop-down list.
CV-W2501
Error: The selected filter type is invalid.
Possible cause: The selected filter type was not Dilation, Erosion, Opening, Closing, or Morph gradient.
Solution: Make sure the selected filter type is Dilation, Erosion, Opening, Closing, or Morph gradient.
CV-W2502
Error: The “Kernel Size” value of the selected filter should exceed 0.
Possible cause: The parameter value is less than or equal to 0.
Solution: Make sure the parameter value is greater than 0.
CV-W2601
Error: The X-direction window size or Y-direction window size should exceed 0.
Possible cause: The parameter value is less than or equal to 0.
Solution: Make sure the parameter value is greater than 0.
CV-W2602
Error: The selected filter type is invalid.
Possible cause: The selected filter type was not Mean, Median, or Decimation.
Solution: Make sure the selected filter type is Mean, Median, or Decimation.
CV-W2701
Error: The “Low Threshold” value should not be greater than that of “High Threshold,” or either of them is not within the range of 0 to 100%.
Possible causes:
-
The low threshold is not less than the high threshold;
-
The filter type is “Percentile,” and the threshold is not within the range of 0 to 1.
Solution:
-
Make sure the low threshold is less than or equal to the high threshold;
-
When the filter type is “Percentile”, make sure the threshold is within the range of 0 to 1.