+7 8202 20-16-35
Mon-Fri 9:00AM – 6:00PM (MSK)

Home Products Machine vision for industry VISCONT.Granules


A system for the granulometric composition assessment of bulk materials

VISCONT.Granules is intelligent system for assessing the granulometric composition of bulk materials on a conveyor belt. Developed with the use of convolutional neural networks, the system determines the granulometric composition with 90% accuracy even in the harshest industrial environment.  

Ask a question

Most often, the assessment of granulometric composition of bulk materials on a conveyor belt is performed manually by a laboratory analysis of randomly picked samples and by visual inspection performed by an operator of a conveyor belt. On average, the accuracy of analysis of the granulometric composition of bulk materials and their dimensions performed by the production line operator reaches 70-80 percent mark. However, the cost of potential mistake is exceptionally high.

In turn, the solution based on the machine vision system reaches the accuracy of 90% even in the harshest industrial environments. Furthermore, unlike the visual inspection performed by an operator of conveyor belt, the machine vision system analyses the granulometric composition for entire surface area of bulk materials.

VISCONT.Granules is the solution developed by Mallenom Systems for visual analysis and control of conveyor belt which transports the bulk materials. Based on the deep neural networks models, the system assesses contents at the conveyor belt as following:

  • First, the system detects and identifies the linear dimensions of individual elements (i.e. granules) out of entire bulk material
  • Then, it classifies the granules into different types based on the obtained linear dimensions
  • It also detects oversized and atypical granules

The reliable completion of these steps is crucial for preparation of bulk materials to progress to the fragmentation stage of the production process that may involve crushing and/or grinding of the matter.  The accurate identification and measurement are vital for the smooth and efficient operation of the production plant as well as for prevention of machinery damage.

Developed machine vision algorithms enable to identify the oversized individual elements that might be present on the conveyor belt and pose the danger to further production processes. When the system detects such oversized elements, it generates a signal that contains photo and video capture of these outliers. If the need arises, the operators can stop the conveyor belt to address these risk factors.     


System integration results:

  • Reduction in depreciation and wear of fixed assets
  • Minimisation of the expenses on granular matter laboratory analysis
  • Minimisation of the plant operational costs
  • Optimisation of production processes and quality control of mined minerals
  • Provision of necessary means for the growth in the quality of operational management
  • Elimination of the human factor influence
  • Reduction of the personal exposure to the fine dust particles and other harmful elements

Stages of system integration:

  • Installation and setting of equipment at the production facility
  • Installation and setting of system libraries and interfaces to optimise the performance of the system
  • Collection of materials for further system models learning
  • Integration with existing information system at the production facility

For the inquiries regarding purchase, please feel free to contact:

Manager for international development