Changelog • SuperpixelImageSegmentation

SuperpixelImageSegmentation 1.0.6

CRAN release: 2025-09-15

I updated the Makevars and Makevars.win files by adding -DARMA_USE_CURRENT (see issue: https://github.com/RcppCore/RcppArmadillo/issues/476)
I removed the -mthreads compilation option from the “Makevars.win” file
I removed the “CXX_STD = CXX11” from the “Makevars” files, and the “[[Rcpp::plugins(cpp11)]]” from the “superPx_AP_Kmeans_Segment.cpp” file due to the following NOTE from CRAN, “NOTE Specified C++11: please drop specification unless essential” (see also: https://www.tidyverse.org/blog/2023/03/cran-checks-compiled-code/#note-regarding-systemrequirements-c11)

CRAN release: 2022-02-06

CRAN release: 2021-05-04

CRAN release: 2021-03-05

I’ve included the Affinity Propagation Clustering parameters in the spixel_segmentation method of the Image_Segmentation R6 class and especially the ap_maxits (maximum number of iterations), ap_convits (convits iterations), ap_dampfact (update equation damping level) and ap_nonoise (small amount of noise to prevent degenerate cases). Although the default parameter values work for the majority of Image Segmentation tasks, adjustments might be necessary for specific use cases.

CRAN release: 2020-06-13

I’ve added the return_labels_2_dimensionsional parameter to the spixel_segmentation R6 Method, so that if TRUE then the 2-dimensional (matrix) superpixel labels will be returned
I’ve added the spixel_clusters_show method to visualize the superpixel clusters in case the return_labels_2_dimensionsional parameter is set to TRUE

CRAN release: 2019-05-29

I’ve added an error case if the kmeans_initializer parameter is not one of ‘kmeans++’, ‘random’, ‘optimal_init’ or ‘quantile_init’

CRAN release: 2018-12-30

The spixel_segmentation method allows also the user to return the kmeans clusters too (in case that the kmeans_method is set to “kmeans” or “mini_batch_kmeans”)