SLIC-superpixel-with-OpenCV icon indicating copy to clipboard operation
SLIC-superpixel-with-OpenCV copied to clipboard

Implementation of the SLIC superpixel algorithm to work with OpenCV2

SLIC Superpixel Modification

We modified the SLICO version of slic algorithm to work with OpenCV2, and make it easy to use.

Class explanation

  • void SLIC::GenerateSuperpixels(cv::Mat& img, UINT numSuperpixels)

    Perform SLIC algorithm on the given image with the given number of superpixels.

    img - 24-bit or 8-bit cv::Mat

    numSuperpixels - an appropriate number, no more than total number of pixels

  • cv::Mat SLIC::GetImgWithContours(cv::Scalar color)

    Get the result image with contours on the given color.

    color - If input image is 24-bit, the color may be cv::Scalar(255, 0, 0) or cv::Scalar(0, 255, 0) or cv::Scalar(0, 0, 255)

    If input image is 8-bit, the color may be cv::Scalar(0) or cv::Scalar(128) or cv::Scalar(255)

  • int* SLIC::GetLabel()

    Get label on each pixel which shows the number of superpixel it belongs to.

Example

  • original image

bird_color.jpg

  • 200 superpixels

bird_color.jpg

  • 500 superpixels

bird_color.jpg