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 journalArticlepeer-review


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
Publication statusPublished - 2019

Bibliographical note

Funding Information:
This work was supported by the Natural Science Foundation of China under grants 61872425, 61771265, and 61801247; the University Science Research of Jiangsu Province Foundation under grant 17KJB520029; the Natural Science Foundation of Jiangsu Province of China under grant BK20180945; Nantong University (17R31); the “333” Talents of Jiangsu (BRA2017475); the Science and Technology Project of Nan-tong (GY12017006); and the Basic Science Research Program of NRF (NRF-2017R1A2B4012720).

Publisher Copyright:
© 2019 Han Wang et al.

ASJC Scopus subject areas

  • General Mathematics
  • General Engineering


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