smooth_imagesmooth_imageSmoothImageSmoothImage (Operator)


smooth_imagesmooth_imageSmoothImageSmoothImage — Smooth an image using various filters.


smooth_image(Image : ImageSmooth : Filter, Alpha : )

Herror smooth_image(const Hobject Image, Hobject* ImageSmooth, const char* Filter, double Alpha)

Herror T_smooth_image(const Hobject Image, Hobject* ImageSmooth, const Htuple Filter, const Htuple Alpha)

void SmoothImage(const HObject& Image, HObject* ImageSmooth, const HTuple& Filter, const HTuple& Alpha)

HImage HImage::SmoothImage(const HString& Filter, double Alpha) const

HImage HImage::SmoothImage(const char* Filter, double Alpha) const

HImage HImage::SmoothImage(const wchar_t* Filter, double Alpha) const   (Windows only)

static void HOperatorSet.SmoothImage(HObject image, out HObject imageSmooth, HTuple filter, HTuple alpha)

HImage HImage.SmoothImage(string filter, double alpha)


smooth_imagesmooth_imageSmoothImageSmoothImageSmoothImage smooths gray images using recursive filters originally developed by Deriche and Shen and using the non-recursive Gaussian filter. The following filters can be choosen via the parameter FilterFilterFilterFilterfilter: 'deriche1', 'deriche2', 'shen' and 'gauss'. The “filter width” (i.e., the range of the filter and thereby result of the filter) can be of any size. In the case that the Deriche or Shen is choosen it decreases by increasing the filter parameter AlphaAlphaAlphaAlphaalpha and increases in the case of the Gauss filter (and AlphaAlphaAlphaAlphaalpha corresponds to the standard deviation of the Gaussian function). An approximation of the appropiate size of the filterwidth AlphaAlphaAlphaAlphaalpha is performed by the operator info_smoothinfo_smoothInfoSmoothInfoSmoothInfoSmooth.

Non-recursive filters like the Gaussian filter are often implemented using filter-masks. In this case the runtime of the operator increases with increasing size of the filter mask. The runtime of the recursive filters remains constant; except the border treatment becomes a little bit more time consuming. The Gaussian filter becomes slow in comparison to the recursive ones but is in contrast to them isotropic (the filter 'deriche2' is only weakly direction sensitive). A comparable result of the smoothing is achieved by choosing the following values for the parameter: Alpha(deriche2) = Alpha(deriche1) / 2, Alpha(shen) = Alpha(deriche1) / 2, Alpha(gauss) = 1.77 / Alpha(deriche1).

For an explanation of the concept of smoothing filters see the introduction of chapter Filters / Smoothing.


Note that filter operators may return unexpected results if an image with a reduced domain is used as input. Please refer to the chapter Filters.

Execution Information


ImageImageImageImageimage (input_object)  (multichannel-)image(-array) objectHImageHImageHobject (byte / uint2 / real)

Image to be smoothed.

ImageSmoothImageSmoothImageSmoothImageSmoothimageSmooth (output_object)  (multichannel-)image(-array) objectHImageHImageHobject * (byte / uint2 / real)

Smoothed image.

FilterFilterFilterFilterfilter (input_control)  string HTupleHTupleHtuple (string) (string) (HString) (char*)


Default value: 'deriche2' "deriche2" "deriche2" "deriche2" "deriche2"

List of values: 'deriche1'"deriche1""deriche1""deriche1""deriche1", 'deriche2'"deriche2""deriche2""deriche2""deriche2", 'gauss'"gauss""gauss""gauss""gauss", 'shen'"shen""shen""shen""shen"

AlphaAlphaAlphaAlphaalpha (input_control)  real HTupleHTupleHtuple (real) (double) (double) (double)

Filterparameter: small values cause strong smoothing (vice versa by using bei 'gauss'"gauss""gauss""gauss""gauss").

Default value: 0.5

Suggested values: 0.1, 0.2, 0.3, 0.5, 1.0, 1.5, 2.0, 2.5, 3.0, 4.0, 5.0, 7.0, 10.0

Typical range of values: 0.01 ≤ Alpha Alpha Alpha Alpha alpha ≤ 50.0

Minimum increment: 0.01

Recommended increment: 0.1

Restriction: Alpha > 0

Example (HDevelop)


Example (C)


Example (HDevelop)


Example (HDevelop)


Example (HDevelop)



If the parameter values are correct the operator smooth_imagesmooth_imageSmoothImageSmoothImageSmoothImage returns the value 2 (H_MSG_TRUE). The behavior in case of empty input (no input images available) is set via the operator set_system('no_object_result',<Result>)set_system("no_object_result",<Result>)SetSystem("no_object_result",<Result>)SetSystem("no_object_result",<Result>)SetSystem("no_object_result",<Result>). If necessary an exception is raised.

Possible Predecessors


Possible Successors

thresholdthresholdThresholdThresholdThreshold, dyn_thresholddyn_thresholdDynThresholdDynThresholdDynThreshold, regiongrowingregiongrowingRegiongrowingRegiongrowingRegiongrowing


binomial_filterbinomial_filterBinomialFilterBinomialFilterBinomialFilter, gauss_filtergauss_filterGaussFilterGaussFilterGaussFilter, mean_imagemean_imageMeanImageMeanImageMeanImage, derivate_gaussderivate_gaussDerivateGaussDerivateGaussDerivateGauss, isotropic_diffusionisotropic_diffusionIsotropicDiffusionIsotropicDiffusionIsotropicDiffusion

See also

info_smoothinfo_smoothInfoSmoothInfoSmoothInfoSmooth, median_imagemedian_imageMedianImageMedianImageMedianImage, sigma_imagesigma_imageSigmaImageSigmaImageSigmaImage, anisotropic_diffusionanisotropic_diffusionAnisotropicDiffusionAnisotropicDiffusionAnisotropicDiffusion


R.Deriche: “Fast Algorithms for Low-Level Vision”; IEEE Transactions on Pattern Analysis and Machine Intelligence; PAMI-12, no. 1; S. 78-87; 1990.