Advanced processing software and hardware are used to correlate specific points found in photographs taken at different angles, generating a point cloud which may then be interpolated into a mesh structure.
We use still and videographic imagery obtained through multiple capture techniques including diver, ROV and drone
Over the years we have greatly enhanced our delivery of underwater models through the employment of dedicated operators and the development of a networked cluster for photogrammetry processing.
The photogrammetry cluster uses several dozen computing cores distributed across four workstation-class nodes to form a parallel processing supercomputer. This network architecture allows simultaneous or sequential alignment of tens of thousands of images and the creation of terabyte-scale projects.
All of our models are ortho-rectified and scaled to permit accurate measurements to be taken from them
The models produce permanent records that can be measured in three-dimensions to much higher levels of accuracy than are possible using traditional methods. The models can be examined for a variety of physical characteristics and successive surveys can be used to quantify change over time.
Our delivery can be generated using still or videographic imagery obtained through multiple capture techniques including diver, ROV or aerial drone.