RU2540489C1 - Method of mineral material dressability assessment by optic method and device for method implementation - Google Patents

Abstract

FIELD: measurement technology.

SUBSTANCE: utility and distinction areas are defined for each mineral object from a sample batch. Colour images are made for at least two sides of each mineral object, their total area and total area of all sides of mineral objects recognized as utile are determined. Source RGB arrays are converted into HLS and Yuv colour spaces, with source RGB arrays preserved. Colour correction is performed for each of nine source arrays, resulting in an aggregate of corrected RGB, HLS, Yuv arrays. Process parameters are determined and dressability is assessed for each mineral object.

EFFECT: increased speed, reliability and accuracy of measurements.

13 cl, 5 dwg

Description

The group of inventions relates to test equipment, namely, technologies and tools used for preliminary assessment of ore dressability of solid minerals and determination of their selection parameters, and can be used in geological research and testing of mineral raw materials of known, discovered deposits, as well as for analysis of the quality of the sorting of industrial or household waste, to verify the effectiveness of quality control of products or industrial products, etc.

The optical method for enrichment of solid mineral ores is successfully used in the mining industry for the extraction of gold, diamonds, nickel, calcite, feldspars, etc. The enrichment of mineral raw materials by this method is based on the selection of minerals according to their optical characteristics, such as gloss, color, reflectivity, transparency. One of the key problems that impede the development and expansion of the applicability of the optical method is the lack of a methodology for evaluating the ore dressability without direct testing on existing various optical separators, as well as equipment that implements it. In addition, it should be noted the absence of criteria in favor of the choice of an optical separator. In various optical separators, various color coordinate systems and various processing algorithms can be used to process the resulting images of mineral objects [Tako P.R. de jong. The Economic Potential of Automatic Rock Sorting / The Netherlands, Delft University of Technology, Department of Geotechnology, 2005. Manouchehri H.R. Sorting: Possibilities, Limitations and Future / Electronic resource. Access mode to the resource: http://pure.ltu.se/portal/files/299975/article - free]. Moreover, all optical separators implement the principle of color image analysis. Therefore, it is urgent to develop a method and device designed for a preliminary assessment of the degree of enrichment of mineral raw materials based on the results of processing color images of several mineral samples (both containing useful components and empty), allowing to develop a decision on the possibility (or impossibility) of separation of minerals optical method. In this case, a decision can be made for a set of possible conditions for the analysis, for example, for a number of light sources (white LEDs, fluorescent lamps, halogen lamps, etc.) and color coordinate systems used to digitize the image (RGB, HLS, Yuv, Lab, etc.).

The RGB space of color coordinates is “native” to most photodetectors, since it is a logical continuation of the method of digitizing an image obtained from a multi-element color optical radiation receiver. The color coordinates of each image element are obtained from the additive synthesis (mixing) of the values of three components: red R (red), green G (green) and blue B (blue). All color coordinates of this color space are interdependent. The data on the lightness (brightness) and color of the image have an implicit expression.

Yuv color coordinate space has one luminance (Y) and two color difference (u and v) components. Thus, the data on the "brightness" and "color" of the image are differentiated. In this case, the luminance component is independent of two interdependent color signals.

The color coordinate space HLS describes the hue from the position of the color tone H (hue), lightness L (lightness) and saturation S (saturation). Thus, the separation of information on “color”, “brightness” and “pallor” of the color shade is realized. Moreover, the color components of the HLS space are interdependent.

There is a method of evaluating the technological properties and enrichment of mineral raw materials (Patent for invention RU 2165632, G01V 9/00, published April 20, 2001), which includes determining the chemical and mineral composition of the raw material, determining the particle size distribution of the ore of the initial size and particle size distribution of the product, crushed down to the maximum degree of disclosure of valuable components, the assessment of the structural and phase characteristics of mineral raw materials, the determination of the physical properties of minerals and forecast technological parameters. Additionally, quantitative values of structural and technological parameters are determined, which include the structural element of disclosure, the boundary coarseness of ore classification in the grinding cycle, the degree of disclosure of ore minerals, the distribution of intergrowths of mineral phases in quality, while quantitatively assessing the structural phase characteristics of mineral raw materials and determining quantitative values of structural and technological parameters are carried out by automatic image analysis by optical-geometric methods odes.

The disadvantages of this method are:

- a long stage of sample preparation, due to the need for preliminary sampling and grinding of samples;

- limited information about the investigated object, because due to the lack of analysis of the "color" parameter, part of the information about the object is lost;

- a knowledge-based process for determining predictive technological indicators, since it requires the use of a significant "knowledge base" that takes into account various methods of sample preparation.

The prior art device for sorting objects according to visual characteristics (Patent for invention RU 2424859, published July 27, 2011), containing a hopper, a device for piece-feeding objects, a horizontal transport disk with a drive that performs transportation of objects to the photoelectronic unit and receiving bins , a node for dumping sortable objects into receiving bins, the number of which corresponds to the number of sorting categories, as well as a sensor that determines the exact position of the object on the transporting disk, photoelectronic The lock is made in the form of a video camera connected to the analysis and control unit, and the transporting disk has a removable background ring of a given shade installed along its edge.

The disadvantages of the known device are:

- low accuracy of analysis due to the impossibility of distinguishing between subtle color tones, since only the RGB color space is used, which is not optimal for analyzing the color of an object due to the uneven representation of color in this color space;

- the presence of an error in determining the color of the object due to the lack of a process of color correction of the matrix of the camera used;

- the ability to analyze only objects of small size: from 1.5 to 5 mm.

The prior art also knows a method for measuring the color of objects (patent RU 2356016, IPC G01J 3/46, published March 20, 2009), which consists in illuminating the surface of an object with two full-color (white) light sources, collecting color photo , a video camera of data on brightness, color hue corresponding to light beams reflected from a given set of points of the illuminated surface of the object, and processing data on brightness, color hue and color saturation of the object. Lighting is carried out alternately with the help of at least two light sources illuminating the surface of the object at different angles.

The disadvantages of this method are:

- low measurement accuracy, since the object is analyzed on one side and over the entire surface, therefore, it is necessary to average, in addition, color correction is not performed;

- limited application, due to the fact that the radiation sources are located asymmetrically relative to the analyzed object (on the one hand), this leads to the fact that only objects with a smooth surface structure can be analyzed, since in the case of rough minerals in the image, part of the object due to the possible the surface topography will be in the shade, so the color information may be unreliable due to the uneven illumination of the object;

- using only RGB and XYZ models when processing, which are not optimal for describing the color of objects, due to the uneven representation of color.

There is a method of color classification of an object (patent for invention RU 2468345, G01J 3/51, G01N 21/85, publication date 11/27/2012), which, by the combination of essential features, is closest to the claimed method and can be taken as a prototype. The method consists in the fact that at least one controlled object is placed in an analysis zone optically coupled to a radiation source and a photodetector module. A color image of each of the control objects is formed in the image plane of the analysis zone, the total color image is converted into an electrical signal, the resulting electrical signal is converted from an analog form to a digital one to obtain three two-dimensional arrays of integers, each of which contains information about the spatial distribution in the image of one of three colors - red, blue or green, then convert the resulting arrays into the HLS color space. After that, the value of the color coordinate h (x _i , y _j ), l (x _i , y _j ) or s (x _i , y _j ) of each element of the corresponding array is compared with the a priori known values of the corresponding color coordinates of the image of the analysis zone. For elements that do not coincide in the color coordinate, the color coordinates are aligned using the leveling mask. And the classification of objects is carried out by comparing the values of color-aligned elements of the arrays [H (x _i , y _j )] [L (x _i , y _j )] and [S (x _i , y _j )], at least one color coordinate with a priori known value of the corresponding color coordinate of reference objects.

The known method has several significant disadvantages:

- low accuracy of the analysis of the controlled object, since the color thresholds of discrimination and separation are not determined;

- with its help it is difficult to make a preliminary assessment of the enrichment of mineral raw materials, since the applied color analysis is limited to the classification of objects by color;

- low reliability of the analysis due to the fact that the classification features are set by the operator, and are not calculated based on the determined color parameters of the images of the studied objects.

The task to which the claimed group of inventions is directed is to achieve the following technical result, namely: to increase the efficiency, reliability and accuracy of determining technological parameters and preliminary assessment of the degree of mineral raw material enrichment while reducing the weight and overall dimensions of the device.

The problem in terms of the method is solved due to the fact that in the method for assessing the degree of mineral dressability by the optical method, which consists in the fact that mineral objects from a batch of samples are placed in the analysis zone optically coupled to radiation sources and photodetector modules, color images are formed, at least two sides of each mineral object in the image plane of the analysis zone, which are then converted into electrical signals of analog form, after which they are analyzed logo-digital conversion with obtaining for each color image three two-dimensional arrays of integers in the RGB color space, each of which contains information about the spatial distribution in the image of one of the three colors - red, green or blue, then convert the RGB arrays into the HLS color space and color correction is performed each of the three arrays of transformed HLS color coordinate by comparing the values of _{h (x i, y j)} , l (x i, y j) and s (x _i, y _j) of each corresponding array element a priori izves the values of the corresponding color coordinates of the image of the analysis zone, and for elements that do not coincide in the color coordinate, align the color coordinates, thus obtaining a set of corrected HLS arrays, it is new that visually on each side of each mineral object from the batch of samples in the analysis plane determine its usefulness and zones of distinction according to preselected selective features; when obtaining color images of the sides of mineral objects, their total area and the total area of images of all sides of mineral objects recognized as useful, the conversion of the source RGB arrays to the HLS color space is carried out simultaneously with the conversion of the source RGB arrays to the Yuv color space while preserving the source RGB arrays, while the color correction of the source RGB arrays and converted Yuv arrays are similar to HLS arrays, while receiving a set of corrected Yuv arrays and a set of corrected RGB arrays, then in each of the three the resulting sets of corrected RGB, HLS, Yuv arrays for each mineral object from the batch of samples determine the thresholds for distinguishing color shades by which utility zones are found on the images on each side of the mineral objects and their total areas are determined for all mineral objects and for mineral objects recognized as useful, accordingly, according to the obtained discrimination thresholds for each set of corrected RGB, HLS, Yuv arrays, the threshold for separation of mineral objects from the sample batch is found in, then for each set of RGB, HLS, Yuv arrays, the degree of enrichment is estimated by the formula

, $$E=\frac{S_g-S_{cg}}{S_0-S_{gg}}\cdot \mathrm{one\; hundred}\%$$

,

where S ₀ - the total area of all images of the sides of the mineral objects from the batch of samples; S _g - the total area of all images of the sides of the mineral objects recognized as useful; S _cg is the total area of all utility zones of images of the sides of mineral objects recognized as useful; S _gg is the total area of utility zones of images of the sides of mineral objects from a batch of samples.

There is also an option for the development of the main technical solution, which consists in the fact that the obtained data on the studied mineral raw materials are recorded in a database.

Thus, the inventive method for assessing the degree of enrichment of mineral raw materials by the optical method with the totality of its essential features allows to achieve high accuracy and reliability of the analysis of the controlled object of mineral raw materials through the use of three color spaces and to preliminarily assess the degree of enrichment of mineral raw materials due to a comprehensive analysis of the color parameters of each mineral object such as color zones, shape and number of small inclusions, total length and width of veins uneven structure (spotting) of the surface, glare; as well as automatic determination of thresholds for distinguishing between the obtained color zones and thresholds for separating objects according to their degree of usefulness.

From the patent for invention RU 2468345, G01J 3/51, G01N 21/85, publication date 11/27/2012, an optical-electronic device for color classification of an object is also known, which, by the combination of essential features, is closest to the claimed device and can be taken as a prototype . This device includes a white light emission source and a photodetector module that are optically coupled to the analysis zone and contains an optical system and an array multi-element optical radiation detector (MPOI) mounted in the image forming plane, on each spatial element of which a red, green or blue color filter is applied. The device contains at least one photodetector module, the output of each of which is connected to the input of the corresponding conversion unit, and the input to the output of the processing unit. The output of each of the transformation units is connected to the processing unit. The device contains at least one radiation source, each of which is connected to the corresponding outputs of the source control unit, the input of which is connected to the second output of the processing unit. The main axis of the source emission indicatrix is located at an angle of 135 ° to the line of sight of the photodetector module for monitoring transparent objects, and at an angle of 45 ° for opaque objects. And the size of the analysis zone is limited by the angular field of the optical system and the linear size of the color-sensitive multi-element receiver of optical radiation, and the size of the analysis zone exceeds the total size of the objects under control.

The known device has several significant disadvantages:

- limited use due to the stationary placement of the device;

- the device needs an additional device for the supply and removal of the analyzed mineral objects;

- limited functionality due to the inability to work without an external power source.

The problem is solved in part of the device due to the fact that in the device for assessing the degree of mineral dressability, including at least two radiation sources, and at least two photodetector modules, optically coupled to the analysis zone, at least two conversion units, each of which is connected to a corresponding photodetector module, a processing unit connected to each conversion unit, a video monitoring device connected to the processing unit, at least one source control unit radiation radiation coupled to each radiation source, wherein each photodetector module comprises an optical system and a multi-element optical radiation detector mounted in the image forming plane, the main axis of the source radiation indicatrix being at an angle to the sighting axis of the photodetector module, with each multi-element optical radiation detector made in the form of a matrix photodetector with red, green and blue color filters applied to its elements, the size being the analysis exceeds the size of the mineral object, the new one is that all radiation sources are equally combined into at least two source blocks, each of which corresponds to one photodetector module and one source control unit, and each radiation source control unit is connected to the corresponding unit transformations, wherein each transform block is configured to convert the obtained images of the analysis zone into at least three different color p of the country, there are at least two attachment devices, on each of which a corresponding photodetector module and source block are fixed, the attachment device made with the possibility of their positioning, each radiation source is multi-element with the ability to change its color, while all the elements of the device are placed into the housing, in which a through hole is made corresponding to the location of the analysis zone.

There is an option for the development of the main technical solution, which consists in the fact that the processing unit is configured to form a database from the received information about mineral objects.

There is an option for the development of the main technical solution, namely, that the fastening devices of the photodetector modules are equipped with a device for changing the distance from the photodetector module to the analysis zone.

There is an option for the development of the main technical solution, which consists in the fact that the video monitoring device is made in the form of a touch screen.

There is an option for the development of the main technical solution, which consists in the fact that the analysis zone is formed on a rotary table located inside the case.

There is a development option for the main technical solution, which consists in the fact that the housing is equipped with a device for carrying it.

There is an option for the development of the main technical solution, which consists in the fact that the optical system of the photodetector module is made in the form of a lens with a variable focal length.

There is an option for the development of the main technical solution, which consists in the fact that there is an autonomous power supply unit that ensures the operability of all elements of the device.

There is an option for the development of the main technical solution, which consists in the fact that there is a synchronization unit for monitoring moving mineral objects.

There is an option for the development of the main technical solution, which consists in the fact that the angle between the main axis of the source radiation indicatrix and the target axis of the photodetector module is 135 ° to control transparent mineral objects.

There is an option for the development of the main technical solution, which consists in the fact that the angle between the main axis of the source radiation indicatrix and the target axis of the photodetector module is 45 ° to control opaque mineral objects.

Thus, the claimed device in its entirety of essential features allows to achieve the claimed technical result due to the fact that:

- it is possible to change the distance from the camera to the object of analysis, i.e. adjustments to the size of the object of analysis;

- the device can operate both from a 220 V network and autonomously;

- the device has small overall dimensions due to the compact arrangement of all functional units in a single housing and low weight, i.e. it can be transferred (transported) from one deposit to another;

- due to the presence of the housing, the device can be operated in the field;

- VKU is made in the form of a touch screen, allows the operator to enter data into the program or add an input device;

- the processing unit has an additional output for connecting a data output device, which allows you to copy the database and the results obtained by evaluating the degree of enrichment, thresholds for distinguishing color zones and thresholds for separating objects for all analyzed color spaces.

The noted disadvantages of the known methods and devices allow us to conclude that the claimed inventions meet the eligibility criterion - inventive step.

Since both of the claimed technical solutions are united by a single inventive concept, in the future they will be considered together.

The invention is illustrated by drawings, in which Fig. 1 shows a generalized block diagram of a device for assessing the degree of enrichment of mineral raw materials, Fig. 2 is a diagram explaining the principle of operation of the device in static mode, Fig. 3 is a diagram explaining the principle of operation of a device in dynamic mode, figure 4 is a generalized diagram of a method for assessing the degree of enrichment of mineral raw materials by the optical method, figure 5 is an illustration of an example implementation.

The proposed method is implemented using a device for assessing the degree of enrichment of mineral raw materials (Fig. 1, 2, 3), which includes at least two radiation sources 1, and at least two photodetector modules 2, optically coupled to zone 3 analysis. The radiation sources 1 are equally divided into at least two source blocks 4, each of which corresponds to one photodetector module 2 and at least one source control unit 5 connected to each radiation source 1. Moreover, each radiation source 1 is made multi-element with the ability to change its color.

There are at least two transform units 6, each of which is connected to a respective photodetector module 2 and to a corresponding source control unit 5. In this case, there is a processing unit 7 connected to each transform unit 6. The processing unit 7 is also connected to the video monitoring device 8.

Each photodetector module 2 contains an optical system 9 and a multi-element optical radiation detector 10 installed in the image forming plane (not shown in the drawing), the main axis of the radiation indicatrix of the source 1 being located at an angle to the sighting axis of the photo-receiving module 2. Each multi-element optical receiver 2 radiation is made in the form of a matrix photodetector with light filters (not shown) on red, green and blue colors applied to its elements. The size of zone 3 of the analysis exceeds the size of the mineral object (not shown in the drawing).

In addition, each transform unit 6 is configured to convert the obtained images of the analysis zone 3 into at least three different color spaces.

There are at least two attachment devices (not shown in the drawing), each of which has a corresponding photodetector module 2 and a source unit 4. Moreover, the mounting device (not shown) is configured to position the photodetector modules 2 and source blocks 4, for example, by supplying it with linear and angular movements.

All of the above elements of the device are placed in the housing (shown conditionally in the drawing), in which a through hole is made (conditionally shown in the drawing), corresponding to the location of analysis zone 3.

Processing unit 7 may be configured to generate a database from the obtained information about the analyzed mineral objects.

Devices (not shown in the drawing) for fastening the photodetector modules 2 can be equipped with a device (not shown in the drawing) for changing the distance from the photodetector module 2 to the analysis zone 3.

Video monitoring device 8 may be made in the form of a touch screen.

In this case, for controlling transparent mineral objects, the angle between the main axis of the radiation indicatrix of the radiation source 1 and the target axis of the photodetector module 2 is 135 °, and for control of opaque mineral objects, the angle between the main axis of the radiation indicatrix of the radiation source 1 and the sighting axis of photodetector module 2 is 45 ° .

In the case of measurements in the static mode (FIG. 2), the analysis zone 3 is formed on a turntable (conventionally shown in the drawing) located inside the housing (conventionally shown in the drawing).

The housing (shown conditionally in the drawing) may be equipped with a device for carrying it.

The optical system 9 of the photodetector module 2 can be made in the form of a lens with a variable focal length.

As a power source, the inventive device can be equipped with an autonomous power supply (not shown), which ensures the operability of all elements of the device. There is also a block 11 converting voltage supply.

In the case of measurements in the dynamic mode (Figure 3) for monitoring moving mineral objects, there is a synchronization unit 12 connected to the photodetector modules 2, the conversion unit 6, the control unit 5 of the radiation sources 1, and the processing unit 7. The synchronization unit 12 can be made in the form of a device with a radiation source and an optical radiation receiver coupled to it. As soon as the mineral object flies past the synchronization unit 12, the control unit 5 controls the radiation sources 1 receives the command to turn on the radiation sources 1, and the photodetector modules 2 - the command to capture a frame with the image of zone 3 analysis.

The software of the inventive device is configured to allow the operator to select any of the selective features, or a combination thereof.

In addition, the processing unit 7 may have an additional output for connecting a device (not shown) to output data.

Evaluation of the degree of enrichment of minerals by the optical method is as follows (Figure 4).

Initially, the utility and zones of differentiation of each mineral object from the sample batch are determined by pre-selected selective features, for example, by the presence of zones of a certain color on the surface of the object, by the shape of the image of the object, by the presence and length of veins on the object, etc., and record the results to the database of the processing unit.

Then, mineral objects from the batch of samples are alternately placed in the analysis zone 3, which is optically coupled to radiation sources 1 and photodetector modules 2. The analysis zone 3 with the mineral object located within it is illuminated using blocks of 4 sources. Using the photodetector modules 2, color images of at least two sides of each mineral object are formed in the image plane of the analysis zone 3, which are then converted into electrical signals of analog form.

After that, in block 6 transformations carry out their analog-to-digital conversion to obtain for each color image three two-dimensional arrays of integers in the RGB color space, each of which contains information about the spatial distribution in the image of one of the three colors - red, blue or green. Upon receipt of color images of the sides of mineral objects, their total area and the total area of images of all sides of the mineral objects recognized as useful are determined.

Then, the source RGB arrays are simultaneously converted into the HLS color space and into the Yuv color space, while the original RGB arrays are preserved. Next, color correction of each of the three initial HLS arrays is carried out by comparing the values of the color coordinates h (x _i , y _j ), l (x _i , y _j ) or s (x _i , y _j ) of each element of the corresponding array with a priori known values of the corresponding color coordinates of the image of zone 3 of the analysis, and for elements that do not coincide in the color coordinate, the color coordinates are aligned, thereby obtaining a set of corrected HLS arrays. At the same time, the color correction of the source RGB arrays and converted Yuv arrays is performed similarly to HLS arrays, while obtaining a set of corrected Yuv arrays and a set of corrected RGB arrays.

Then, in each of the three obtained sets of corrected RGB, HLS, Yuv arrays for each mineral object from the batch of samples in the processing unit, the thresholds for distinguishing color shades are determined by which utility zones are found on the images of each side of the mineral objects and their total areas for all mineral objects are determined and for mineral objects deemed useful, respectively. Then, using the obtained discrimination thresholds for each set of corrected RGB, HLS, Yuv arrays, the threshold for separation of mineral objects from the batch of samples is found.

Then, for each set of RGB, HLS, Yuv arrays, the degree of enrichment is estimated by the formula

, $$E\frac{S_g-S_{cg}}{S_0-S_{gg}}\cdot \mathrm{one\; hundred}\%$$

,

where S ₀ - the total area of all images of the sides of the mineral objects from the batch of samples;

S _g - the total area of all images of the sides of the mineral objects recognized as useful;

S _cg is the total area of all utility zones of images of the sides of mineral objects recognized as useful;

S _gg is the total area of utility zones of images of the sides of mineral objects from a batch of samples.

All obtained data on the studied mineral raw materials are recorded in the database of the processing unit, analyzing which, the specialist concludes that it is possible to enrich this type of mineral raw materials with the optical method with the choice of a separator (the previously defined discrimination and separation thresholds can be used to configure the selected separator).

An example of the implementation of the method of evaluating the enrichment.

Using the proposed method for assessing the degree of enrichment by the optical method, a representative sample of gold ore from the Konevinskoye deposit was analyzed. The sample contained 117 mineral samples.

The task of the analysis was to determine the possibility of ore beneficiation of the indicated deposit by the optical method, in particular:

- search for thresholds of discrimination in the images of samples of useful areas belonging to minerals: quartz (white and light yellow shades - the most useful mineral), oxidized berezite (brown shades - a useful mineral), berezite (swamp green shades - the least useful mineral) and granodiarite (gray shades - empty carrier rock);

- determination of the optimal threshold for the separation of mineral samples into useful and empty;

- calculation of the assessment of the degree of enrichment for several separation thresholds close to optimal.

Figure 5 shows the most typical samples of a representative sample of gold-bearing ore of the Konevinskoye deposit (the given images of the samples were saved during their analysis): a - sample No. 753 (a useful sample of beresite, characterized by the presence of oxides), b - sample No. 794 (the most useful sample containing quartz), c - sample No. 783 (empty sample of bearing rock).

During the analysis of this representative sample, thresholds were obtained for distinguishing utility zones for white and light yellow, brown, and marsh green shades (table) in the RGB, Yuv, and HLS color spaces. Color shades that do not belong to those considered were automatically assigned to gray.

Table 1 shows the obtained thresholds for distinguishing useful color shades for RGB, Yuv, and HLS color spaces.

Table 1 Light shades Swamp green shades Brown shades RGB R / g G B / g R / g G B / g R / g G B / g (2.12; 100) (0; 255) (0.6; 100) (0.71; 2.9) (120; 255) (0.7; 0.9) (0.45; 0.7) (0; 255) (0.65; 0.9) Yuv Y u V Y u v Y u v [80; 100] [-50; 50] [-50; 50] [-50; 40] [-20; -3) [-50; -3] (10; 40) [-50; -1) (10.5; 50] [20; 100] [-50; -3) (3; 10.5] [0; 10) [-50; -3] (7; 50] Hls H L S H L S H L S [0; 360] [90; 100] [0; 100] [105; 170] [25; 53] (0; 80] [340; 360] [1; 15] [0; 100] [0; 70] [18; 90) [0; 100] [0; 50] [1; 15] [0; 100]

For low-contrast objects, floating thresholds were used (Table 1), i.e. for channels R and B, thresholds normalized along channel G were set.

In addition, to determine the necessary color shades in the color spaces Yuv and HLS, discrimination thresholds were obtained for the five working areas.

It was determined that for the separation of useful mineral samples from samples of empty bearing rock, the relative percentage of “gray” shades in the image of the studied mineral sample of a representative sample should be used.

The optimum separation threshold value for the RGB and Yuv color spaces was determined to be 75% of “gray” shades (if the percentage of “gray” shades in the image of the sample is greater than the specified threshold, then it is assigned to samples of empty carrier rock, if less, then it is useful). For the HLS color space, the optimum separation threshold is 70% gray.

When using the separation threshold, the value of which is less than optimal, samples of useful rock will be determined as empty. In the case of assigning a separation threshold, the value of which is more than optimal, samples of an empty bearing rock will be determined as useful.

Preliminary estimates of the degree of enrichment of a representative sample of gold ore from the Konevinskoye deposit were also determined for a number of separation thresholds close to its optimal value in all color spaces under consideration.

Table 2 shows the values of preliminary estimates of the degree of enrichment of a representative sample of gold ore from the Konevinskoye deposit at various separation thresholds.

table 2 Color space Separation threshold values,% 60 70 75 80 90 Assessment of sample enrichment,% RGB 1,093541 8,775127 17,41088 39,00831 95,05763 Ynv 2,878684 9,737966 18,8907 36.9642 76,32731 Hls 28,00367 52,55025 67.75355 77.8498 96,40542

According to the results presented in the second table, it is most effective to analyze the representative sample considered using the HLS color space, and the least effective way is to use the RGB color space.

An example of a specific implementation of the device.

In the presented embodiment, the inventive device for implementing the method for evaluating the optical dressability of mineral raw materials is intended for the analysis of mineral objects of a size range from 10 mm to 100 mm.

, однако линейный размер матрицы, равный 4,86 мм × 3,62 мм, вносит дополнительное ограничение на размер зоны анализа. Полезное поле зрения объектива составляет по горизонтали 2ω_g=33,8° и по вертикали 2ω_v=25,5°. Контролируемые объекты расположены на расстоянии a=250 мм от объектива. Таким образом, действительный размер зоны анализа по горизонтали составляет x_g=2a·tg(ω_g)=2·250·tg(16,9°)=151,875 (мм), а по вертикали x_v=2a·tg(ω_v)=2·250·tg(12,75°)=113,125 (мм), т.е. 152 мм × 113 мм. Размер выравнивающей маски b×b равен 3×3 пикселей. Таким образом, минимальный размер изображения объекта $$r_{p\min }=b^2\cdot r_{el}\cdot \sqrt{2}=9\cdot 3,75\cdot {10}^{-3}\cdot \sqrt{2}=5\cdot {10}^{-2}\text{ (}мм)$$

. Минимальный размер определяемого вкрапления объекта, соответственно $$r_{ob\min }=(x_g/1294)\cdot b^2\cdot \sqrt{2}=(151,875/1294)\cdot 9\cdot \sqrt{2}=1,5\text{(}мм)$$

.This sample device contains two photodetector modules. Each photodetector module is made in the form of a digital camera, the optical system of which is made in the form of a photographic lens with an angular field of 2ω = 48.6 ° and focal length f '= 8 mm, and a multi-element optical image receiver is in the form of a CMOS matrix with 1294 × 964 spatial elements (pixels), each of which is 3.75 μm × 3.75 μm in size (r _el = 3.75 μm). The size of the analysis zone is limited by the angular field of the optical system and the linear size of the color-sensitive multi-element optical radiation detector. Since the focal length of the lens is f '= 8 mm, and the angular field is 2ω = 48.6 °, it follows that the image size of the analysis zone for each coordinate cannot exceed $${x^{\prime }}_{\max }^{}=2\cdot tg(\omega )\cdot f\text{'}\text{'}=2\cdot tg(24,3)\cdot 8=7,2\text{mm}$$

however, a linear matrix size of 4.86 mm × 3.62 mm introduces an additional restriction on the size of the analysis zone. The useful field of view of the lens is 2ω _g = 33.8 ° horizontally and 2ω _v = 25.5 ° vertically. Controlled objects are located at a distance of a = 250 mm from the lens. Thus, the actual size of the analysis area horizontally is x _g = 2 a · tg (ω _g ) = 2 · 250 · tg (16.9 °) = 151.875 (mm), and vertically x _v = 2 a · tg ( ω _v ) = 2 · 250 · tg (12.75 °) = 113.125 (mm), i.e. 152 mm × 113 mm. The size of the b × b alignment mask is 3 × 3 pixels. Thus, the minimum image size of an object $$r_{p\min }={\mathrm{<figure-callout\; id="2"\; label="b"\; filenames="00000008.png,00000009.png"\; state="\{\{state\}\}">b</figure-callout>}}^2\cdot r_{el}\cdot \sqrt{2}=9\cdot 3,75\cdot {10}^{-3}\cdot \sqrt{2}=5\cdot {10}^{-2}\text{(}mm)$$

. The minimum size of the determined interspersed object, respectively $$r_{ob\min }=(x_g/1294)\cdot {\mathrm{<figure-callout\; id="2"\; label="b"\; filenames="00000008.png,00000009.png"\; state="\{\{state\}\}">b</figure-callout>}}^2\cdot \sqrt{2}=(151,875/1294)\cdot 9\cdot \sqrt{2}=\mathrm{one},5\text{(}mm)$$

.

Each photodetector module corresponds to two blocks of radiation sources. Each block of radiation sources is made in the form of a matrix of 66 semiconductor emitting luminescent white SMD diodes (3 rows of 22 LEDs). Blocks of radiation sources are arranged in pairs in the horizontal and vertical planes. For the analysis of opaque mineral objects, the main axis of the radiation indicatrix of the blocks of radiation sources are located at an angle of ± 45 ° to the sight axis of the photodetector module. To control transparent objects, the same number of blocks of radiation sources, but the main axis of their radiation indicatrix are located at an angle of ± 135 ° to the sight axis of the photodetector module.

The source control unit is made in the form of an electronic device containing an ARDUINO UNO chip, and controls the operation of four blocks of radiation sources.

The conversion unit consists of an analog-to-digital converter operating under the control of a microcontroller. The conversion unit is made as a single unit with a photodetector module.

The processing unit is made in the form of an electronic computer.

The video control device is made in the form of a monitor, mouse and keyboard.

Thus, the inventive method for assessing the degree of enrichment of mineral raw materials by the optical method and a device for its implementation on the basis of the above set of features provide high efficiency, reliability and accuracy of determining technological parameters and preliminary assessment of the degree of enrichment of mineral raw materials due to a comprehensive analysis of the color parameters of each mineral object with a decrease weight and overall dimensions of the device.

Claims (13)

1. A method for assessing the degree of mineral mineral enrichment by the optical method, which consists in placing mineral objects from a batch of samples in an analysis zone optically coupled to radiation sources and photodetector modules, color images are formed of at least two sides of each mineral object in image planes of the analysis zone, which are then converted into electrical signals of an analog form, after which they are converted by analog-to-digital conversion to obtain for each color images of three two-dimensional arrays of integers in the RGB color space, each of which contains information about the spatial distribution in the image of one of the three colors - red, green or blue, then convert the RGB arrays to the HLS color space and perform color correction of each of the three converted HLS arrays by comparing the color coordinate values of _{h (x i, y j)} , l (x i, y j) and s (x _i, y _j) corresponding to each element of the array with a priori known values of the respective color image coordinates analysis zones, and for elements that do not coincide in the color coordinate, color coordinates are equalized, thereby obtaining a set of corrected HLS arrays, characterized in that before installing each mineral object from a batch of samples in the analysis plane, its utility and zones of discrimination are determined visually on each side of it pre-selected selective features, when obtaining color images of the sides of mineral objects, their total area and total area of images of all sides are determined mineral objects recognized as useful, the conversion of the source RGB arrays to the HLS color space is carried out simultaneously with the conversion of the source RGB arrays to the Yuv color space while preserving the source RGB arrays, while the color correction of the source RGB arrays and the converted Yuv arrays is similar to HLS arrays, thus obtaining a combination corrected Yuv arrays and a set of corrected RGB arrays, then in each of the three resulting sets of corrected RGB, HLS, Yuv arrays for for each mineral object from the batch of samples, thresholds for distinguishing color shades are determined by which utility zones are found on the images of each side of mineral objects and their total areas for all mineral objects and for mineral objects recognized as useful are determined, respectively, according to the obtained discrimination thresholds for each set of corrected RGB, HLS, Yuv arrays find the threshold for the separation of mineral objects from a batch of samples, then for each set of RGB, HLS, Yuv arrays, ku degree of enrichment by the formula
$$E=\frac{S_g-S_{cg}}{S_0-S_{gg}}\cdot \mathrm{one\; hundred}\%$$

,
where S ₀ is the total area of all images of the sides of mineral objects from the batch of samples, S _g is the total area of all images of the sides of mineral objects recognized as useful, S _cg is the total area of all utility zones of images of the sides of mineral objects recognized as useful, S _gg is the total area utility zones of images of sides of mineral objects from a batch of samples. 2. The method according to claim 1, characterized in that the obtained data on the studied mineral raw materials are recorded in the database. 3. A device for assessing the degree of enrichment of mineral raw materials, including at least two radiation sources, and at least two photodetector modules, optically coupled to the analysis zone, at least two conversion units, each of which is connected to the corresponding photodetector module, a processing unit connected to each conversion unit, a video monitoring device connected to the processing unit, at least one radiation source control unit connected to each radiation source, when ohm, each photodetector module contains an optical system and a multi-element optical radiation detector mounted in the image forming plane, the main axis of the radiation indicatrix of the source being at an angle to the sighting axis of the photodetector module, and each multi-element optical radiation detector is made in the form of a matrix photodetector with the elements applied to it filters of red, green and blue colors, and the size of the analysis zone exceeds the size of the mineral object, different I mean that all radiation sources are combined into at least two source blocks, each of which corresponds to one photodetector module and one source control unit, and each source control unit is connected to a corresponding conversion unit, and each conversion unit is made with the ability to convert the obtained images of the analysis zone into at least three different color spaces, there are at least two attachment devices, on each and which is fixed the corresponding photoreceptor module and power source, wherein the device attachment is capable of their ranking, each radiation source configured multielement able to change its color, all elements of the device are placed in a housing, wherein a through hole corresponding to the arrangement assay zone. 4. The device according to claim 3, characterized in that the processing unit is configured to form a database from the received information about the mineral object. 5. The device according to claim 3, characterized in that the fastening devices of the photodetector modules are equipped with a device for changing the distance from the photodetector module to the analysis zone. 6. The device according to claim 3, characterized in that the video monitoring device is made in the form of a touch screen. 7. The device according to claim 3, characterized in that the analysis zone is formed on a rotary table located inside the housing. 8. The device according to claim 3, characterized in that the housing is equipped with a device for carrying it. 9. The device according to claim 3, characterized in that the optical system of the photodetector module is made in the form of a lens with variable focal length. 10. The device according to claim 3, characterized in that there is an autonomous power supply unit that ensures the operability of all elements of the device. 11. The device according to claim 3, characterized in that for the control of moving mineral objects there is a synchronization unit. 12. The device according to claim 3, characterized in that for controlling transparent mineral objects, the angle between the main axis of the indicatrix of the radiation source and the target axis of the photodetector module is 135 °. 13. The device according to claim 3, characterized in that for controlling opaque mineral objects, the angle between the main axis of the indicatrix of the radiation source and the target axis of the photodetector module is 45 °.

Images