Work

Development of a Machine Vision System for Monitoring the Frothing Process

Year: 2026

Type: article

Abstract: An approach to the construction of a contactless automatic control system for frothing during flotation is considered. The structure of the machine vision system is proposed. The choice of the measuring unit installation location is justified. Two approaches for constructing a contactless automatic froth control system in the flotation process based on machine vision are considered. The first is based on the traditional Watershed preprocessing and segmentation algorithm, while the second uses a convolutional neural network. The U-Net deep neural network image segmentation algorithm achieves high segmentation accuracy (IoU = 0.6947). It is resistant to noise, uneven lighting and dynamic changes in the froth.

Source: Eng—Advances in Engineering

DOI: 10.3390/eng7070319

Authors: Natalia Vasilyeva, Pavel Ivanov, Aleksandra Sniatkova

Institutions: Saint Petersburg Mining University, Stockholm School of Economics in Russia

Language: English

Topic: Industrial Vision Systems and Defect Detection

Subfield: Industrial and Manufacturing Engineering

Field: Engineering

Domain: Physical Sciences

SDG: Industry, innovation and infrastructure

Open Access status: gold

Estimated APC paid: $1,448