Author(s): Ichiro FUJITA, Tatshushi Shibano, Kojiro Tani
Linked Author(s): Ichiro FUJITA, Kojiro Tani
Keywords: Space-time image velocimetry; Discharge measurement; Wave-frequency spectrum; KU-STIV;
Abstract: This paper aims to develop a method for improving the measurement accuracy of river surface velocity in using the space-time image velocimetry (STIV), which is one of the image-based techniques using river surface video images. STIV has been recognized as an efficient technique to measure river discharge remotely by using existing river monitoring cameras because it can measure streamwise velocity components more efficiently and accurately than the other imaging technique such as the large-scale particle image velocimetry (LSPIV). However, there are cases where adequate measurement accuracy is difficult to maintain due to worse image shooting conditions of river surface. In this research, textures that appears in a space-time image (STI) generated by STIV were classified into ten types and developed image filters mainly based wavenumber-frequency spectrum to improve the texture quality of STI. It was made clear that quality of STI can be remarkably improved with the filters and contribute to the improvement of STIV.