Learning salient seeds refer to the manifold ranking and background-prior strategy

Yung Chen Chou, Yu Wei Nien, Ying Chin Chen, Bo Li, Jung San Lee

Research output: Contribution to journalArticle

Abstract

Recently, the key technique of image processing has been widely applied to pattern recognition, content retrieval, and object segmentation. These applications have brought much higher complexity in image computation. Accordingly, the processed results may be interfered due to the interlacing reference. To overcome this problem, researchers have developed the object detection mechanism, which is a preprocessing procedure to extract significant feature to stand for the whole image. However, the error rate of detection is a crucial challenge in this research field. Based on the concept of manifold ranking, we have designed a brand-new object detection method considering both local and global features. The experimental results have demonstrated that the new method is able to lower down the detection error rate in case that the object located near the boundary.

Original languageEnglish (US)
Pages (from-to)5859-5879
Number of pages21
JournalMultimedia Tools and Applications
Volume79
Issue number9-10
DOIs
StatePublished - Mar 1 2020

Keywords

  • Edge detecting prior
  • Manifold ranking
  • Object detection
  • Salient map

ASJC Scopus subject areas

  • Software
  • Media Technology
  • Hardware and Architecture
  • Computer Networks and Communications

Fingerprint Dive into the research topics of 'Learning salient seeds refer to the manifold ranking and background-prior strategy'. Together they form a unique fingerprint.

  • Cite this