TY - JOUR ID - 70126 TI - Application of Structure from Motion (SFM) Method to Determine the Bed Surface Particles Sizes in Gravel Bed Rivers JO - Iranian Journal of Soil and Water Research JA - IJSWR LA - en SN - 2008-479X AU - Zamani, Parisa AU - Mohajeri, Seyed Hossein AU - Samadi, Amir AD - Graduated M.Sc., Water Eng. Dept., Faculty of Agriculture and Natural Resources, Imam Khomeini International University (IKIU), Qazvin, Iran. AD - Assistant Professor, Civil Eng. Dept., Faculty of Eng. and Tech., Kharazmi University, Karaj, Iran. AD - Assistant Professor, Water Eng. Dept., Faculty of Agriculture and Natural Resources, Imam Khomeini International University (IKIU), Qazvin, Iran. Y1 - 2019 PY - 2019 VL - 50 IS - 1 SP - 215 EP - 230 KW - Close-Range Photogrammetry KW - Digital Elevation Model KW - Laser Scanner KW - Rough bed KW - Structure from Motion DO - 10.22059/ijswr.2018.254906.667879 N2 - Accurate and precise characterization of the natural rough bed has great importance. Without any doubt, there is not any example of natural flow or flow near hydraulic structures with no roughness on their surrounding walls Although the traditional bed roughness characterization approach is based on the grain size distribution curve, in the recent approach, roughness determination is based on the point-to-point height measurement of the bed, which cannot be easily determined. Therefore, despite of many studies and various methods and tools which have been developed for determining the digital model elevation and statistical properties of such substrates, there is a lack of simple and low-cost method with high accuracy. In the present study, the capabilities of a close-range photogrammetric method called the Structure from Motion (SFM) have been investigated for determining the bed surface particles sizes. For verification, the digital elevation of various objects with  regular geometric shapes, such as spheres and cubes, was determined using SFM method and compared with the theoretical values ​​derived from their mathematical equation. The results of the model derived by the structure from motion method for irregular geometric shapes was performed using a laser scanner and a caliper which indicated  the high precision of the simple and low-cost SFM method. The results showed that the SFM method could accurately developed a digital model of an artificial gravel and sand bed (absolute error of 0.19 to 1 mm). Furthermore, this method was applied in the real environment;  Kordan River bed and the size distribution of the point to point  bed particles  were calculated based on the cloud points of the developed digital model, indicating the capability of the method for determining the natural roughness of the river bed based on the concepts of statistical methods. UR - https://ijswr.ut.ac.ir/article_70126.html L1 - https://ijswr.ut.ac.ir/article_70126_77fd44d8b9945e86489fc4c183e19191.pdf ER -