Using Commercial-Grade Digital Camera Images in the Estimation of Hydraulic Flume Bed Changes: Case Study
A methodology for utilizing commercial-grade digital camera images, surveying measurements, and digital photogrammetry techniques to substitute traditional measurement methods for creating a digital elevation model (DEM) of hydraulic flume bed was developed and tested. A Sony DSC T9 camera, Sokkia set3 total stations, and Leica Photogrammetry Suite software were used. Two different image resolutions were tested for two cases of the hydraulic flume bed representing initial and posthydraulic test cases. The DEMs were compared with the ones created using manual measurements, using a steel ruler for the initial case. The developed methodology provided a reliable and cost-effective alternative to traditional methods for medium- to low-accuracy applications. A maximum of 12-mm elevation RMSE was obtained for the check points in the initial case. The corresponding RMSE for the posthydraulic test case was 10 mm. No significant gain in obtained accuracy was achieved when higher-resolution images were used. Absolute values of the differences between the DEMs created using six- and three-megapixel-resolution image sets for the initial and post-hydraulic test cases were less than 2 cm for 80 percent and 86 percent of the pixels, respectively. This percentage increases to above 90 percent if the difference threshold is increased to 3 cm. The differences between photogrammetry derived DEMs and manually collected data were less than 2 cm for 71 percent of the DEM pixels and less than 3 cm for 90 percent of the pixels.
No Reference information available - sign in for access.
No Citation information available - sign in for access.
No Supplementary Data.
No Article Media