In HALCON 13, surface-based 3D matching has been improved to be more robust when dealing with flat surfaces. The new matching algorithm now analyses edges in both the object model and the 3D scene. This means that ambiguities from flat surfaces can now be solved using edge data. For example, a flat object will have a similar 3D surface profile as the conveyor belt on which it is located. This would be difficult to accurately locate in a 3D scene using only the 3D point information. With the improved surface-based 3D matching algorithm, the edges of the object can be used along with the 3D point information for more accurate and robust results. This improvement particularly supports applications like robotic picking of boxes.
HALCON 13 also offers a new method to reconstruct 3D objects from multiple cameras with high quality. This new method uses the information of all camera views at once which leads to more robust results than those provided by common stereo reconstruction methods.
Additionally, HALCON 13 transfers the colour information from the input images to the reconstructed objects. This is useful for better visualisation or further processing (e.g., with a 3D printer).
HALCON 13 is due to be released on November 1, 2016
In the meantime, why not check out HALCON 12?