An Iterative Raster Scan Algorithm for Superpixel Segmentation

機関リポジトリ (HANDLE) オープンアクセス

説明

We propose an iterative raster scan algorithm for superpixel segmentation, which is based on the K-means clustering algorithm. The proposed algorithm updates the class label of each pixel only at the boundaries of superpixels in a raster scan order, and refers to only two neighboring pixels per pixel for updating the variables. Therefore, the proposed algorithm is computationally efficient compared with existing methods. Experimental results show that the proposed algorithm generates compact and adherent superpixels in a finite number of iterations of the raster scan process.

収録刊行物

関連プロジェクト

もっと見る

キーワード

詳細情報 詳細情報について

  • CRID
    1050299981545239296
  • ISSN
    21878811
    21881758
  • HANDLE
    2324/7173592
  • 本文言語コード
    en
  • 資料種別
    journal article
  • データソース種別
    • IRDB

問題の指摘

ページトップへ