A fast search algorithm for MPEG video using multiple histogram descriptors

Feifei Lee, Koji Kotani, Qiu Chen, Tadahiro Ohmi

Research output: Contribution to journalArticlepeer-review

4 Citations (Scopus)


In this paper, we present a fast and robust video search algorithm using multiple histogram descriptors for large video database. There are two types of histogram descriptors used to generate more robust features. The first one is based on the adjacent pixel intensity difference quantization (APIDQ) algorithm, which has been reliably applied to human face recognition previously. An APIDQ histogram is utilized as the feature vector of the frame image. The other one is ordinal histogram descriptor which is robust to color distortion. Combined with active search, a temporal pruning algorithm, fast and robust video search can be realized. Experimental results show our proposed algorithm can detect a given 15 second video clip from a 6 hours of MPEG video in merely 100ms, and achieve Equal Error Rate (ERR) of 1.5%, which is more accurately and robust than conventional fast video search algorithm.

Original languageEnglish
Pages (from-to)164-171
Number of pages8
JournalInternational Journal of Digital Content Technology and its Applications
Issue number9
Publication statusPublished - 2011 Sep


  • Adjacent pixel intensity difference quantization (APIDQ)
  • DC image
  • Fast search
  • Histogram descriptor
  • Ordinal feature

ASJC Scopus subject areas

  • Software
  • Computer Networks and Communications


Dive into the research topics of 'A fast search algorithm for MPEG video using multiple histogram descriptors'. Together they form a unique fingerprint.

Cite this