Function (biology)

Algorithm

Process (computing)

Engineering

Biology

Construct (python library)

Statistics

Energy (signal processing)

Operating system

Computer science

Evolutionary biology

Mechanical engineering

Photogrammetry

Point cloud

Mathematics

Machine vision

Evaluation function

Artificial intelligence

Geometry

Point (geometry)

Programming language

Engineering drawing

Machining

Computer vision

Assembly Automation

Purpose This paper aims to solve the problem of free-form tubes’ machining errors which are caused by their complex geometries and material properties. Design/methodology/approach In this paper, the authors propose a multi-view vision-based method for measuring free-form tubes. The authors apply photogrammetry theory to construct the initial model and then optimize the model using an energy function. The energy function is based on the features of the image of the tube. Solving the energy function allows to use the gray features of the images to reconstruct centerline point clouds and thus obtain the pertinent geometric parameters. Findings According to the experiments, the measurement process takes less than 2 min and the precision of the proposed system is 0.2 mm. The authors used simple operations to carry out the measurements, and the process is fully automatic. Originality/value This paper proposes a method for measuring free-form tubes based on multi-view vision, which has not been attempted to the best of authors’ knowledge. This method differs from traditional multi-view vision measurement methods, because it does not rely on the data of the design model of the tube. The application of the energy function also avoids the problem of matching corresponding points and thus simplifying the calculation and improving its stability.

A measurement method of free-form tube based on multi-view vision for industrial assembly