Manufacturing and manufacturing industries

Vol. 2 No. 31 (2026) / 2026-06-30 / Number of views: 115

COMPUTER VISION CONTROL OF ORE MATERIAL DISTRIBUTION IN BUNKERS

Authors

K. Kulazhanov Kazakh University of Technology and Business
https://orcid.org/0000-0001-5202-3501
LLP «AG Tech»
https://orcid.org/0009-0003-3613-0130
K. Kulazhanov Kazakh University of Technology and Business
https://orcid.org/0000-0001-8069-0037
K. Kulazhanov Kazakh University of Technology and Business
https://orcid.org/0000-0001-9143-4581
K. Kulazhanov Kazakh University of Technology and Business
https://orcid.org/0009-0001-1288-4136
LLP «AG Tech»
https://orcid.org/0009-0009-2520-7781
LLP «AG Tech»
https://orcid.org/0009-0004-9133-6593
LLP «AG Tech»
https://orcid.org/0009-0004-8436-5943
LLP «AG Tech»
https://orcid.org/0009-0009-6261-7093
Toraighyrov University
https://orcid.org/0000-0003-4688-7869

Keywords

computer vision, bunker node, lump ore, segregation, coefficient of unevenness, video monitoring, congestion

Link to DOI:

https://doi.org/10.58805/kazutb.v.2.31-1413

How to quote

Akishev, K.M., et al. “COMPUTER VISION CONTROL OF ORE MATERIAL DISTRIBUTION IN BUNKERS”. Vestnik KazUTB, vol. 2, no. 31, June 2026, doi:10.58805/kazutb.v.2.31-1413.

Abstract

The need to improve methods of controlling the processes of loading and outflow of lump ore in the bunker nodes of mining and processing enterprises is an important task, since the uneven distribution of material, segregation, and congestion formation have a significant impact on equipment productivity and the effectiveness of subsequent technological transformations.

The paper develops and implements a computer vision method for monitoring the condition of ore material in a bunker based on the analysis of video data obtained from cameras installed above the bunker.

The essence of the method consists in image segmentation, extraction of geometric and textural features of the embankment surface, formation of the coefficient of uneven distribution of Kh, indicators of textural analysis based on the matrices of the common occurrence of gray levels (GLCM).

The scientific novelty of the study is to substantiate the use of visual features as virtual measuring parameters, with the possibility of integration into the contour of automated process control systems.

It is shown that an increase in the coefficient of uneven distribution of Kh correlates with a decrease in the productivity of feeders by 4-5%, a decrease in the extraction of a useful component in subsequent alterations by 2-3%.

The possibility of early detection of segregation zones and congestion has been confirmed.

The practical significance of the study is to increase the stability of the technological process through contactless monitoring of hopper loading.

The prospects for further research are related to the development of predictive models and the adaptation of the method to various operating conditions.