Main page › Implementations › Industrial machine vision › Color classification system for diamonds
Intelligent (self-learning) system for detection and classification of diamonds by color based on the optoelectronic stereoscopic method.
In 2015-2016, Mallenom Systems completed the project for development of mathematical algorithm and software for automatic sorting of natural diamonds by color. This project was carried out for the order placed by ALROSA, and the general contractor for this order was NPP Burevestnik JSC. Optoelectronic stereoscopic method for control, machine vision technologies and machine learning all have been at the core of this system.
Principle of operation:
The video cameras capture the diamond movement while it is free falling. The software collects the images from the cameras and analyses color properties of the diamond. Diamond classification is performed based on the models constructed with the help of machine learning methodology and algorithms for video analysis developed by Mallenom Systems.
The learning of the developed models has taken place with the use of diamond samples provided by ALROSA. This has ensured the highest classification reliability for this particular company. To improve the quality of classification further the system can undergo gradual additional learning by studying new types of diamonds, including the rare ones. Additionally, the system allows to create unique classification markers for each specific diamond lot for much more detailed record keeping.
Separate software was developed for generation of new classification models, which allows to continue learning of already existing and perform learning for new models directly by the system operator.
The developed system sorts diamonds at a rate of 20 diamonds per second. The number of classification classes is only limited by the capacity of the sorting machinery.