Package org.opencv.ximgproc

Class SuperpixelSLIC

  • public class SuperpixelSLIC
extends Algorithm
    Class implementing the SLIC (Simple Linear Iterative Clustering) superpixels algorithm described in CITE: Achanta2012. SLIC (Simple Linear Iterative Clustering) clusters pixels using pixel channels and image plane space to efficiently generate compact, nearly uniform superpixels. The simplicity of approach makes it extremely easy to use a lone parameter specifies the number of superpixels and the efficiency of the algorithm makes it very practical. Several optimizations are available for SLIC class: SLICO stands for "Zero parameter SLIC" and it is an optimization of baseline SLIC described in CITE: Achanta2012. MSLIC stands for "Manifold SLIC" and it is an optimization of baseline SLIC described in CITE: Liu_2017_IEEE.

    • Constructor Detail

      • SuperpixelSLIC

        protected SuperpixelSLIC​(long addr)

    • Method Detail

      • __fromPtr__

        public static SuperpixelSLIC __fromPtr__​(long addr)

      • getNumberOfSuperpixels

        public int getNumberOfSuperpixels()
        Calculates the actual amount of superpixels on a given segmentation computed and stored in SuperpixelSLIC object.
        Returns:
        automatically generated

      • iterate

        public void iterate​(int num_iterations)
        Calculates the superpixel segmentation on a given image with the initialized parameters in the SuperpixelSLIC object. This function can be called again without the need of initializing the algorithm with createSuperpixelSLIC(). This save the computational cost of allocating memory for all the structures of the algorithm.
        Parameters:
        num_iterations - Number of iterations. Higher number improves the result. The function computes the superpixels segmentation of an image with the parameters initialized with the function createSuperpixelSLIC(). The algorithms starts from a grid of superpixels and then refines the boundaries by proposing updates of edges boundaries.

      • iterate

        public void iterate()
        Calculates the superpixel segmentation on a given image with the initialized parameters in the SuperpixelSLIC object. This function can be called again without the need of initializing the algorithm with createSuperpixelSLIC(). This save the computational cost of allocating memory for all the structures of the algorithm. The function computes the superpixels segmentation of an image with the parameters initialized with the function createSuperpixelSLIC(). The algorithms starts from a grid of superpixels and then refines the boundaries by proposing updates of edges boundaries.

      • getLabels

        public void getLabels​(Mat labels_out)
        Returns the segmentation labeling of the image. Each label represents a superpixel, and each pixel is assigned to one superpixel label.
        Parameters:
        labels_out - Return: A CV_32SC1 integer array containing the labels of the superpixel segmentation. The labels are in the range [0, getNumberOfSuperpixels()]. The function returns an image with the labels of the superpixel segmentation. The labels are in the range [0, getNumberOfSuperpixels()].

      • getLabelContourMask

        public void getLabelContourMask​(Mat image,
                                boolean thick_line)
        Returns the mask of the superpixel segmentation stored in SuperpixelSLIC object.
        Parameters:
        image - Return: CV_8U1 image mask where -1 indicates that the pixel is a superpixel border, and 0 otherwise.
        thick_line - If false, the border is only one pixel wide, otherwise all pixels at the border are masked. The function return the boundaries of the superpixel segmentation.

      • getLabelContourMask

        public void getLabelContourMask​(Mat image)
        Returns the mask of the superpixel segmentation stored in SuperpixelSLIC object.
        Parameters:
        image - Return: CV_8U1 image mask where -1 indicates that the pixel is a superpixel border, and 0 otherwise. are masked. The function return the boundaries of the superpixel segmentation.

      • enforceLabelConnectivity

        public void enforceLabelConnectivity​(int min_element_size)
        Enforce label connectivity.
        Parameters:
        min_element_size - The minimum element size in percents that should be absorbed into a bigger superpixel. Given resulted average superpixel size valid value should be in 0-100 range, 25 means that less then a quarter sized superpixel should be absorbed, this is default. The function merge component that is too small, assigning the previously found adjacent label to this component. Calling this function may change the final number of superpixels.

      • enforceLabelConnectivity

        public void enforceLabelConnectivity()
        Enforce label connectivity. superpixel. Given resulted average superpixel size valid value should be in 0-100 range, 25 means that less then a quarter sized superpixel should be absorbed, this is default. The function merge component that is too small, assigning the previously found adjacent label to this component. Calling this function may change the final number of superpixels.