Automatic ore drawing experiment method and system based on image recognition

Abstract

The invention discloses an automatic ore drawing experiment method and system based on image recognition. The method comprises the steps that the placement direction of mark particles is marked through colors and labels; the ore drawing equipment is filled with ore bulk solids and mark particles layer by layer through a mechanical arm; ore drawing equipment is started, and images of ore bulk solids and mark particles on each layer are shot; the images are processed to form an image training set so as to train a convolutional neural network; the discrete body image is recognized through the trained convolutional neural network to obtain the color and number of each mark particle and the corresponding released body volume when the mark particles appear; and the data is processed by using a numbering direct reduction method and a pore-forming quantity method, and a release volume form is drawn. According to the method, manual weight measurement, mark particle selection and identificationand data recording are not needed, a camera and a neural network are used for identification and analysis to complete the release body reduction process, continuous operation is adopted, the pore reaching quantity value of each mark particle can be measured more accurately, and the obtained release body form is more accurate.

Description

technical field [0001] The invention relates to the field of caving mining methods, and is particularly suitable for measuring the flow properties of bulk materials, in particular to an automatic ore drawing experiment method and system based on image recognition. Background technique [0002] In caving mining, ore drawing is a very important process, which directly affects the ore recovery rate and waste rock mixing rate, and is related to the economic benefits of the entire enterprise. If the design of the structural parameters of the stope is unreasonable or the ore drawing process is not managed properly, it is easy to increase the loss rate and dilution rate, resulting in the waste of resources and the decline of mine profits. According to statistics, the loss rate of ferrous metal mines mined by the caving method in my country is 20-30%, and the dilution rate is 25-35%, which is about 10-15% higher than that of foreign advanced mines. However, the properties of bulk o...

AI Technical Summary

Problems solved by technology

Due to the limitation of on-site ore-drawing experiments, the current main method for studying the law of bulk flow and measuring the flow parameters of bulk materials is to use similar materials in the laboratory for physical ore-drawing experiments, but the traditional ore-drawing experiments are discontinuous between each process. It is necessary to manually weigh, pick out the marker particles and record the data after each release of a certain amount of material, which leads to time-consuming and labor-intensive experiments, which is not conducive to the research work of the experiment

Images