Calculation of precipitation intensity using processing the sequence of video frames
The article describes the method for automatic precipitation detection and its intensity estimation from a sequence of video frames. The convolution operation used for filtration is optimized in order to decrease computational complexity. The accuracy of precipitation intensity estimation is improve...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
EDP Sciences
2019-01-01
|
| Series: | ITM Web of Conferences |
| Online Access: | https://www.itm-conferences.org/articles/itmconf/pdf/2019/07/itmconf_crimico2019_12018.pdf |
| Summary: | The article describes the method for automatic precipitation detection and its intensity estimation from a sequence of video frames. The convolution operation used for filtration is optimized in order to decrease computational complexity. The accuracy of precipitation intensity estimation is improved by accounting for cross-correlation between sequential video frames. The proposed method has reduced computational complexity. The results of the study are demonstrated on a few images of different size. The computational complexity of proposed algorithm is appropriate for real-time implementation. |
|---|---|
| ISSN: | 2271-2097 |