Image Filtering¶
Function¶
This Step can perform filtering process on the input images with different kinds of filters.
Usage Scenario¶
This Step is widely used for image processing, and there is no specific usage scenario.
Parameters¶
The following types of filters are available:
GaussianFilter¶
Utilize Gaussian function to process images.
- Size X
- Instruction: This parameter is used to set the width of the Gaussian kernel size. When the value is increased, the image is blurred.Default Value: 3Suggested Value: Please set according to the actual scenario.
- Size Y
- Instruction: This parameter is used to set the height of the Gaussian kernel size. When the value is increased, the image is blurred.Default Value: 3Suggested Value: Please set according to the actual scenario.
- Sigma X
- Instruction: This parameter is used to set the standard deviation of the Gaussian kernel in X direction. When the value is increased, the image is blurred.Default Value: 3Suggested Value: Please set according to the actual scenario.
- Sigma Y
- Instruction: This parameter is used to set the standard deviation of the Gaussian kernel in Y direction. When the value is increased, the image is blurred.Default Value: 3Suggested Value: Please set according to the actual scenario.
BilateralFilter¶
- Diameter
- Instruction: This parameter determines the range of filtering area. When the value is increased, more original image information will be retained, so that the image is clearer.Default Value: 3Suggested Value: Please set according to the actual scenario.
Hint
Standard Deviation of Color Space Filter and Standard Deviation of Coordinate Space Filter affect the filter result jointly.
- Standard Deviation of Color Space Filter
- Instruction: This parameter is used to set the standard deviation of pixel values.Default Value: 3Suggested Value: Please set according to the actual scenario.
- Standard Deviation of Coordinate Space Filter
- Instruction: This parameter is used to set the standard deviation of the spatial distance between pixels.Default Value: 3
Boxfilter¶
- Kernel Size
- Instruction: This parameter is used to set the kernel size during the filtering process and its unit is pixel. When the value is increased, the image obtained is blurred.Default Value: 3Suggested Value: Please set according to the actual scenario.
ConditionalMeanFilter¶
- Kernel Size
- Instruction: This parameter is used to set the kernel size.Default Value: 3Suggested Value: Please set according to the actual scenario.
- Threshold
- Instruction: This parameter is used to set the threshold of the distance between pixels and its unit is pixel (px). When the distance is greater than this threshold, the pixel will be retained. When it is less than this threshold, the pixel will not be included in filtering.Default Value: 100 pxSuggested Value: Please set according to the actual scenario.
DftFilter¶
Utilize discrete Fourier transform to process images and filter out the signals with high frequencies, noise and edges for example. This filter is capable of noise reduction and edge enhancement.
GuidedFilter¶
- Kernel Size
- Instruction: This parameter is used to set the kernel size. When the value is increased, the image smoothing effect is more obvious.Default Value: 3Suggested Value: Please set according to the actual scenario.
- Eps
- Instruction:This parameter is used to set the regularization parameter of the guided filter.Default Value: 0.01Suggested Value: Please set according to the actual scenario.
HaarFilter¶
Smooth the image and remove some noise.
MedianFilter¶
- Kernel Size
- Instruction: This parameter is used to set the kernel size. When the value is increased, the image smoothing effect is more obvious.Default Value: 3Suggested Value: Please set according to the actual scenario.