Interpreting the data
What are you trying to see?
Joining the dots…
Stereo Global Matching (SGM) Software
Our SGM algorithms were developed by our partners for use on the Mars Rovers missions and is currently on the Perseverance Rover. Consequently, our software has world-leading stereo matching capabilities delivering a dense point cloud for 3D analysis for stereo vision systems.
Dense stereo matching is finding corresponding points for each pixel from two or more images obtained at slightly different viewpoints, and it is a fundamental research problem in 3D scene reconstruction.
Stereo matching or disparity estimation is the process of finding the pixels in the multiscopic views that correspond to the same 3D point in the scene. The rectified epipolar geometry simplifies this process of finding correspondences on the same epipolar line
We provide demonstration software for image acquisition, stereo camera calibration, stereo matching and 3D reconstruction (point cloud output).
Our stereo vision technologies can be integrated with our machine learning algorithms and adapted to suit any industry or environment.
Our proprietary software uses an adaptive leased squares correlation technique for data matching and is a more efficient matching technique than cross-matching. It offers high matching accuracy and delivers a combination of area-based and edge-based analysis. It can be applied to multi-spectral and multi-temporal images to feature extraction, change detection, and line following, making it highly flexible for terrain mapping.
Phase is our all-in-one software for stereo vision and 3D processing. Phase makes it easy to integrate our stereo matching camera systems into other software and equipment such as for Unity for augmented reality and virtual reality mapping, robotic integrations, along with other sensors such as thermal imaging, spectrometers, and radiation detectors.
We offer this for C++, C#, and Python languages.
We have a collection of data sets for a variety of applications which will soon be integrated into the Phase Toolkit. These data sets revolve around object detection:
- Cracks & defects in glass, wire, and steel plate
- Weld defects
- Edge recognition
We continue to research and develop new software.