Earticle

초록

영어

In this paper, a dynamic algorithm is proposed that is aimed at removing raindrops within fixed background video images according to the volume of raindrops. In surveillance videos with fixed backgrounds, when the field of vision is obstructed because of falling rain or snow, obtaining picture information on the objects becomes difficult. Therefore, obstacles such as raindrops that obstruct the field of vision should be removed from pictures to obtain more specific picture information. The proposed technique obtains images on a periodical basis from a fixed background picture and assesses whether there are raindrops in the present screen among the obtained images. The proposed technique predicts the volume of raindrops by comparing images obtained periodically and removes raindrops by applying a different dynamic algorithm based on the predicted volume of raindrops. This study embodied and experimented with the proposed technique and verified that the result was excellent.

목차

Abstract
 1. Introduction
 2. Related Work
  2.1. Picture Processing Equipment and Methods to Remove Rain from Pictures
  2.2. Rain Removal in Videos by Using Kalman Filter [9]
  2.3. Detection and Removal of Rain from Videos [3]
  2.4. Rain Removal in Videos by Combining Temporal and Chromatic Properties [10]
 3. Proposed Algorithm
  3.1. Picture Photographing Interval
  3.2. Extraction of Candidate Raindrops
  3.3. Algorithm to Remove Raindrops
 4. Experiments
  4.1. Experimental Setups
  4.2. Experimental Results
 5. Conclusion
 Acknowledgments
 References

키워드

저자

• Nam-Bong Ha [ Department of Computer Science, Kyonggi University ]
• Namgi Kim [ Department of Computer Science, Kyonggi University ] Corresponding author

참고문헌

발행기관

간행물

표지보기 관심저널 등록

• 간행물명

  International Journal of Multimedia and Ubiquitous Engineering

• 간기

  월간

• pISSN

  1975-0080

• 수록기간

  2008~2016

• 등재여부

  SCOPUS

• 십진분류

  KDC 505 DDC 605