Back to Search
Start Over
Motion robust rain detection and removal from videos
- Source :
- MMSP
- Publication Year :
- 2012
- Publisher :
- IEEE, 2012.
-
Abstract
- Weather such as rain and snow cause difficulties in processing the videos captured. Since the appearance of rain drops can affect the performance of human tracking and reduce the efficiency of video compression, detection and removal of rain is a challenging problem in outdoor surveillance systems. In this paper, we propose a new algorithm for rain detection, which is based on joint spatial and wavelet domain features. This approach is robust to the videos with moving objects in the rain. Experimental results demonstrated its better performance in comparison with the existing approaches in the subjective quality.
- Subjects :
- Computer science
business.industry
Feature extraction
Wavelet transform
ComputerApplications_COMPUTERSINOTHERSYSTEMS
GeneralLiterature_MISCELLANEOUS
Object detection
Wavelet
Video tracking
Computer vision
Artificial intelligence
Joint (audio engineering)
business
Rain and snow mixed
ComputingMethodologies_COMPUTERGRAPHICS
Data compression
Subjects
Details
- Database :
- OpenAIRE
- Journal :
- 2012 IEEE 14th International Workshop on Multimedia Signal Processing (MMSP)
- Accession number :
- edsair.doi...........3dd149b6f95f910994b252f2b8e61f09