Abstract
Particle sorting has proven to be extremely profitable in mine processes, upgrading the grade of ore and the overall economics of a mine drastically. This is accomplished by the separation of valuable ore from barren material, reducing overall comminution and operational expenditures. Modern usage of particle sorting in precious metals mining predominantly features XRT, Laser, and XRF sensors, but optical sorting has yet to achieve the success it has in the field of food and plastic processing due to the physical characteristics of the ore. Color sorting has been implemented in limestone, feldspars, and talcum. Color sorting lacks the fine-tuning as seen with XRT, Laser, and XRF as those sensors have physical and chemical characteristics to exploit but provide advantages because of lower associated costs and the ability to handle larger throughputs. This leaves color sorting with immense potential for improvement as a more suitable algorithm is found.
Precious metals mining in the industry has progressively resulted in the mining of lower-grade ore as valuable deposits are exhausted. This paper aims to determine the efficacy of color-sorting AI algorithms on gold sulfide deposits.