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

Main page Implementations Industrial machine vision System for the quality assessment of the sugar beets inside of a truck bed

System for the quality assessment of the sugar beets inside of a truck bed

When receiving the shipment at Znamensky sugar plant in the Tambov region (RUSAGRO-CENTER LLC), the facility evaluates the quality of delivered sugar beets that are located inside of a truck bed. The quality of delivered sugar beet is assessed based on the presence of dirt, sugar beet tops and other foreign objects, chipping and frostbites on the sugar beet roots. The engineers at Mallenom Systems have been tasked to develop automated system for quality evaluation of the received sugar beets, which is now classified as VISCONT.SugarBeet.

Principle of operation

At the point of delivery, an IP video camera is installed under the roof of a shed. The computer receives the video data stream. The computer itself is located at a remote location or inside of a protective cabinet. When VISCONT.SugarBeet receives the request from the plant control system, the obtained frame from a camera is analysed using the specially designed algorithms for this task. With the use of neural networks, the contents inside of a truck bed are located and the search for presence of dirt, beet tops, chipping and snow is performed. Then, based on the obtained data, the system classifies received shipment into quality categories. The final results with the quality evaluation are sent into the plant control system database.

 

For each quality parameter the system applies the weight coefficient and calculates the final quality score based on the all collected data. Then, this result is compared with floor quality value which is dynamically set and depends on the utilisation capacity of the plant. Finally, the system assigns the designation for the received shipment which can be either “storage” or “processing”.

System implementation goals:

  1. Improvement of the accuracy and minimisation of the “human error” during the assignment of designation for the received shipment
  2. Development of the sugar beet planning supply system for the plant (during an integration with the corporate information system)

System implementation stages:

  1. Instalment and calibration of equipment.
  2. Instalment and setting of system libraries and interfaces in accordance with intended task.
  3. Integration with corporate information system.
  4. Collection of samples for further improvement/learning of the models.