Bipolar Log-Intensity-Variance Histogram Method for Local Image Patch Intensity Change Measurements

Han Wang, Quan Shi, Zhihuo Xu, Ming Wei, Hanseok Ko

Research output: Contribution to journalArticle

Abstract

For a fixed-position camera, the intensity changes of an image pixel are often caused by object movement or illumination change. This paper focuses on such a problem: given two adjacent local image patches, how can the causes of intensity change be determined? A bipolar log-intensity-variance histogram is proposed to describe the intensity variations on the chaos phase plot subspace. This is combined with two sigmoid functions to construct a probabilistic measure function. Experimental results show that the proposed measurements are more effective and robust than conventional methods to the cause of variation in image intensity.

Original languageEnglish
Article number3651529
JournalMathematical Problems in Engineering
Volume2019
DOIs
Publication statusPublished - 2019 Jan 1

ASJC Scopus subject areas

  • Mathematics(all)
  • Engineering(all)

Fingerprint Dive into the research topics of 'Bipolar Log-Intensity-Variance Histogram Method for Local Image Patch Intensity Change Measurements'. Together they form a unique fingerprint.

  • Cite this