A rapid and accurate computer vision system for measuring the volume of axi-symmetric natural products based on cubic spline interpolation
Joko Siswantoro, Endah Asmawati, Muhammad Zain Fawwaz Nuruddin Siswantoro · Journal of Food Engineering · 2022
The volume contains prominent information used to evaluate the visual quality of natural products and can be measured using a computer vision system. The computer vision system provides an alternative method for measuring the volume of natural products, especially for axi-symmetric natural products. The volume of axi-symmetric natural products can be obtained using integral for the volume of solid of revolution by rotating the boundary curve of the upper half object cross-section. This study proposes a computer vision system for measuring the volume of axi-symmetric natural products based on cubic spline interpolation. A piecewise cubic polynomial was constructed using cubic spline interpolation to approximate the boundary curve of the upper half object cross-section based on the boundary points of the object extracted from the image of the object. The polynomial was integrated to obtain the volume of the object. The experimental result reveals that the proposed system achieved an accurate volume measurement result with a mean absolute relative error of 1.03% compared to the water displacement method result.