Determination of cutting positions of honeycomb blocks using computer vision
This addresses a domain-specific automation challenge in manufacturing or materials processing, but it is incremental as it builds on existing computer vision techniques for a niche application.
The paper tackled the problem of automating the cutting of honeycomb blocks by developing a computer vision method to determine cutting positions and angles, achieving satisfactory results with an allowable cut plane location of 0.4 times the cell face length.
The article discusses a method for automating the process of cutting a honeycomb block, and specifically obtaining points and cutting angles for the required faces. The following requirements are taken into account in the calculations: the allowable location of the cut plane is 0.4 of the length of the cell face, the cut plane must be perpendicular to the cell wall. The algorithm itself consists of two main stages: determining the honeycomb structure and searching for cut points. In the absence of significant defects in honeycomb blocks (deformation of the cell profile and a dent on the edges of the cells), the structure determination algorithm works without significant inaccuracies. The results of the cut point search algorithm can be considered satisfactory.