CN115532649A - Intelligent coal gangue sorting system - Google Patents
Intelligent coal gangue sorting system Download PDFInfo
- Publication number
- CN115532649A CN115532649A CN202211266735.XA CN202211266735A CN115532649A CN 115532649 A CN115532649 A CN 115532649A CN 202211266735 A CN202211266735 A CN 202211266735A CN 115532649 A CN115532649 A CN 115532649A
- Authority
- CN
- China
- Prior art keywords
- gangue
- coal
- target
- injection
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/34—Sorting according to other particular properties
- B07C5/342—Sorting according to other particular properties according to optical properties, e.g. colour
- B07C5/3425—Sorting according to other particular properties according to optical properties, e.g. colour of granular material, e.g. ore particles, grain
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/36—Sorting apparatus characterised by the means used for distribution
- B07C5/363—Sorting apparatus characterised by the means used for distribution by means of air
- B07C5/367—Sorting apparatus characterised by the means used for distribution by means of air using a plurality of separation means
Landscapes
- Analysing Materials By The Use Of Radiation (AREA)
Abstract
The application provides an intelligent coal gangue sorting system which comprises a conveying part, a data acquisition part, an injection part and an upper computer, wherein the conveying part comprises a motor and a belt conveyor, and the belt conveyor conveys coal gangue materials under the driving of the motor; the data acquisition part is used for emitting X rays to the coal gangue material and acquiring transmission intensity data and transmission images; the injection part is arranged at the tail end of the belt conveyor along the conveying direction and comprises a plurality of pneumatic valves for injecting coal gangue materials to realize the separation of coal and gangue; and the upper computer classifies the coal gangue materials and determines the real-time injection frequency of the injection part based on the transmission intensity data and the transmission image. The application provides a coal gangue intelligence sorting system can show and improve the proportion of selecting out coal from the gangue material, has realized the high-efficient utilization to the gangue.
Description
Technical Field
The application belongs to the technical field of coal sorting, and particularly relates to an intelligent coal gangue sorting system.
Background
Coal will be one of the most important non-renewable resources in our country and even in the world in the visible future. As the world's largest countries for coal production and consumption, coal accounts for over 70% of the energy structures in China, and such energy structures will be difficult to change for a considerable period of time in the future.
Coal dressing is the basis of clean coal technology and is also the premise of deep processing (coal water slurry preparation, coking, vaporization and liquefaction) and clean and efficient utilization of coal. The coal quality can be improved through coal dressing, most of mineral impurities in the raw coal can be removed through washing and dressing the raw coal, 65% of ash content can be reduced, 65% of pyrite can be removed, the pollution of the fire coal to the atmosphere is reduced, and the environmental benefit is very obvious. However, due to the limitation of resources, energy and technical conditions, the raw coal washed by the power plants each year only accounts for 1/4 of the raw coal consumed by all the power plants, and compared with developed countries, the method has great improvement space; in addition, the traditional coal washing method, such as a manual method, a heavy medium method, a jigging method and the like, has huge equipment and is easy to cause serious water pollution.
In addition to the problems existing in the raw coal washing process, the realization of efficient separation of coal gangue is also an urgent need for optimizing the utilization rate of coal resources. The coal gangue is a solid waste with low carbon content and high ash content formed in the coal mining and processing process, and is one of the industrial solid wastes with the largest discharge amount in China. So far, china has accumulated a large amount of coal gangue, if the coal gangue can not be effectively utilized, not only the coal resource and other effective mineral components contained in the coal gangue are wasted, so that the overall utilization rate of raw coal can not be improved, but also the utilization rate of land resources is seriously influenced by the large amount of coal gangue stacking, therefore, the coal gangue resource is required to be efficiently utilized through coal gangue sorting, so that good economic benefit and environmental benefit are realized. .
However, since the coal gangue is a product of raw coal after washing, the mechanization degree of coal mining is improved, the gangue content of the coal gangue is continuously increased, the coal gangue is sorted by using the traditional coal sorting method such as a manual method, a heavy medium method, a jigging method and the like, the difficulty of classification and identification of the coal gangue is high, the sorting cost is also sharply increased along with the increase of the gangue content, and even the sorting cost exceeds the profit of the coal gangue.
Therefore, there is a need for a gangue sorting system that can accurately identify and sort gangue, is pollution-free, and has a high profit-to-cost ratio.
Disclosure of Invention
In order to solve the problems existing in the prior art, the application aims to provide the intelligent coal and gangue sorting system which can realize accurate, efficient and intelligent coal and gangue sorting by utilizing different transmission characteristics of coal and gangue under X-rays.
The embodiment of the application can be realized by the following technical scheme:
an intelligent coal gangue sorting system comprises a conveying part, a data acquisition part, an injection part and an upper computer, wherein the conveying part comprises a motor and a belt conveyor, and the belt conveyor conveys coal gangue materials under the driving of the motor; the data acquisition part is used for emitting X rays to the coal gangue material and acquiring transmission intensity data and transmission images; the injection part is arranged at the tail end of the belt conveyor along the conveying direction and comprises a plurality of pneumatic valves for injecting the coal gangue materials to realize the separation of coal and gangue; and the upper computer classifies the coal gangue materials based on the transmission intensity data and the transmission image and determines the real-time injection frequency of the injection part.
Furthermore, the data acquisition part comprises an X-ray source and an X-ray detection device which are oppositely arranged on two sides of the conveying surface of the belt conveyor; the X-ray detection device is used for acquiring the transmission intensity data and the transmission image, wherein the transmission intensity data comprises high-energy transmission intensity data and low-energy transmission intensity data.
Further, the host computer includes: the system comprises a motor control unit, a data acquisition control unit, an image processing unit, a classification unit and a blowing control unit; the motor control unit is used for controlling the rotating speed of the motor; the data acquisition control unit is used for controlling the X-ray source to emit X-rays to the coal gangue material and receiving the transmission intensity data and the transmission image; the image processing unit extracts a coal gangue target from the transmission image; the classification unit calculates a detection vector of the coal and gangue target based on the transmission intensity data, and classifies the coal and gangue target according to the detection vector; the injection control unit controls the real-time injection frequency of the injection part based on the classification result of the coal and gangue targets.
Preferably, the transmission intensity data further comprises equivalent transmission intensity data; the transmission image is generated based on the equivalent transmission intensity data.
Further, the image processing unit extracts a gangue target from the transmission image by:
s100, denoising the transmission image through spatial smoothing processing;
s200, generating a binary edge image based on an edge detection algorithm of multi-operator fusion;
and S300, determining a coal and gangue target from the binaryzation edge image based on a clustering algorithm.
Preferably, the spatial smoothing process is to filter the transmission image using a median filter template.
Further, the multi-operator fused edge detection algorithm performs edge detection on the transmission image by using the following fractional differential operator:
where f (x, y) is the edge value at the image (x, y) coordinates, f 0 Is a fractional order differential operator of R-L, w 0 Is f 0 The weight of (a) is determined,f l ,l∈[1,…,L]is L different integer order differential operators, w l Is f l And satisfy
Further, step S300 includes the steps of:
s310, importing the binaryzation edge image;
s320, randomly selecting a plurality of edge points from the binaryzation edge image as a starting centroid;
s330, when the cluster distribution of any edge point changes, the step S340 and the step S350 are executed circularly, otherwise, the loop is skipped, wherein,
s340, sequentially executing the following steps on each edge point in the edge image:
s341, respectively calculating connectivity among centroids, distances between the centroids and cluster centers and distances between the centroids and edge points for each centroid;
s342, deleting the centroids communicated with other centroids and centroids surrounded by other clusters based on the calculation result;
s343, distributing the edge points to the clusters nearest to the edge points;
s350, for each cluster, solving a cluster mean value and updating the cluster mean value into a mass center;
and S360, determining a coal gangue target based on the clustering result.
Preferably, after determining the coal gangue target, the image processing unit further performs sampling point optimization on the coal gangue target, where the sampling point optimization further includes the following steps:
s410, dividing each coal gangue target into a plurality of sampling rings at equal intervals from a clustering center to a clustering edge;
s420, deleting the sampling ring on the outermost side of the gangue target;
s430, selecting a sampling ring for coal and gangue classification based on the area of the residual part of the coal and gangue target;
s440, traversing each sampling ring for coal gangue classification;
and S450, determining a coal and gangue target for coal and gangue classification based on the traversal result.
Further, the classification unit classifies the coal gangue target by the following steps:
A100. determining an energy detection vector for a coal gangue target based on the transmission intensity data
Wherein m is the number of sampling points in the coal gangue target,
low energy transmitted intensity, R, for the ith sample point i Is the classification characteristic of the ith sample point determined by:
wherein the content of the first and second substances,
initial values for high energy transmission intensity and low energy transmission intensity,
high energy transmission intensity for the ith sample point;
A200. calculating a classification correlation coefficient r of the coal gangue target based on the energy detection vector j ,
Wherein c is coal, g is gangue,
is composed of
Mean value of (1) i,j To be R i Fitting curve of coal/gangue
The value of the estimated value obtained is,
is I i,j The mean value of (a);
A300. and classifying the coal and gangue targets through a preset target identification strategy based on the classification correlation coefficient.
Further, the target identification policy is one of the following policies:
simple classification strategy if r c ＞r g If the coal and gangue target is coal, otherwise, the coal and gangue target is gangue;
reducing the gangue content if r c If the target is greater than the first threshold value, the coal gangue target is coal, otherwise, the coal gangue target is gangue;
reducing the coal content if r g If the second threshold value is greater than the first threshold value, the coal and gangue target is the gangue, otherwise, the coal and gangue target is the coal.
Further, the blowing control unit determines a real-time blowing frequency of the blowing section by:
B100. establishing a gangue injection queuing model (X, Y, Z, A, B and C), wherein X is gangue arrival time interval distribution, Y is injection time distribution, Z is the number of pneumatic valves, A is the maximum queuing number allowed by an optimized queuing system, B is the total quantity of gangue, and C is a queuing rule;
B200. determining injection queuing index (L) based on gangue injection queuing model s ,L q ,W s ,W q ,P n λ, μ, ρ) of L s To average number of gangue arrivals, L q To average queue length, W s Is the average latency time, W q Average waiting time for queue, P n The probability of stable injection of n pieces of waste rocks is shown, lambda is the average number of the waste rocks arriving in unit time, mu is the number of the waste rocks injected by a single pneumatic valve each time, and rho = lambda/mu is the real-time blowing frequency of a blowing part;
B300. substituting the real-time classification result of the coal gangue materials reaching the tail end of the conveyor belt into the injection queuing index, and determining the optimal value of rho based on the following optimization targets:
wherein z is the total revenue function of blowing, C s The cost coefficient of jetting is carried out for the pneumatic valve, and G is the income of a piece waste rock of jetting.
The embodiment of the application provides a gangue intelligence sorting system has following beneficial effect at least:
(1) The intelligent coal and gangue sorting system provided by the application can quickly realize the intelligent identification and sorting of targets based on different X-ray transmission characteristics of coal and gangue, and effectively solves the problems of complex process flow, poor identification rate, low sorting efficiency, easy water pollution and the like existing in various coal sorting methods in the prior art;
(2) According to the intelligent coal and gangue sorting system, the edge detection algorithm of the image processing unit with multi-operator fusion respectively carries out edge detection on coal and gangue targets, target positioning and clustering precision can be effectively improved, meanwhile, sampling points are optimized according to the size range of the coal and gangue targets, the sampling points influencing classification accuracy can be further eliminated, and classification accuracy is improved;
(3) According to the intelligent coal and gangue sorting system, the preset classification strategy is adopted to classify and recognize coal and gangue, different coal and gangue sorting requirements can be met by setting different judging conditions, and the application range of the system is greatly enlarged;
(4) According to the coal and gangue intelligent sorting system, the jetting queuing indexes are established by using the queuing model, the gangue abandoning cost and the gangue jetting benefit are introduced to construct the jetting cost function, the real-time jetting frequency of the pneumatic valve is determined based on the real-time classification result of the coal and gangue target, and the jetting scheme with the optimal comprehensive performance can be obtained on the basis of comprehensively considering various factors influencing the jetting effect through the method.
Drawings
FIG. 1 is a schematic equipment assembly diagram of an intelligent gangue sorting system according to an embodiment of the application;
FIG. 2 is a schematic diagram of a transmission image according to an embodiment of the present application;
FIG. 3 is a schematic diagram of a system framework and data transmission of an upper computer according to an embodiment of the present application;
FIG. 4 is a graph illustrating the noise reduction of the transmission image of FIG. 2 by a value filter template according to an embodiment of the present application;
FIG. 5 is a transmission image and a plurality of mine refuse objects extracted therefrom according to an embodiment of the present application;
FIG. 6 is a schematic diagram of a gangue target being divided into a plurality of sampling rings according to an embodiment of the present application;
FIG. 7 shows the measured energy vector distribution of coal and gangue.
Detailed Description
Hereinafter, the present application will be further described based on preferred embodiments with reference to the accompanying drawings.
In addition, various components on the drawings are enlarged or reduced for convenience of understanding, but this is not intended to limit the scope of the present application.
Singular references also include plural references and vice versa.
In the description of the embodiments of the present application, it should be noted that if the terms "upper", "lower", "inner", "outer", etc. are used for indicating the orientation or positional relationship based on the orientation or positional relationship shown in the drawings or the orientation or positional relationship which is usually arranged when the products of the embodiments of the present application are used, the description is only for convenience and simplicity, but the indication or suggestion that the referred device or element must have a specific orientation, be constructed in a specific orientation and be operated, and thus, the application cannot be construed as being limited. Moreover, the terms first, second, etc. may be used in the description to distinguish between different elements, but these should not be limited by the order of manufacture or by importance to be understood as indicating or implying any particular importance, and their names may differ from their names in the detailed description of the application and the claims.
The terminology used in the description is for the purpose of describing the embodiments of the application and is not intended to be limiting of the application. It is also to be understood that, unless otherwise expressly stated or limited, the terms "disposed," "connected," and "connected" are intended to be open-ended, i.e., may be fixedly connected, detachably connected, or integrally connected; they may be mechanically coupled, directly coupled, indirectly coupled through intervening media, or interconnected between two elements. The specific meaning of the above terms in the present application will be specifically understood by those skilled in the art.
Fig. 1 is an assembly schematic diagram of an intelligent coal gangue sorting system according to some embodiments of the present application, and as shown in fig. 1, the intelligent coal gangue sorting system provided by the present application includes a conveying portion, a data collecting portion, an injection portion and an upper computer, where the conveying portion includes a motor and a belt conveyor, and the belt conveyor is driven by the motor to convey coal gangue materials; the data acquisition part is used for emitting X rays to the coal gangue material and acquiring transmission intensity data and transmission images; the injection part is arranged at the tail end of the belt conveyor along the conveying direction and comprises a plurality of pneumatic valves for injecting coal gangue materials to realize the separation of coal and gangue; and the upper computer classifies the coal gangue materials based on the transmission intensity data and the transmission image and determines the real-time injection frequency of the injection part.
In some embodiments, the data acquisition part further comprises an X-ray source and an X-ray detection device which are oppositely arranged at two sides of the conveying surface of the belt conveyor, wherein the X-ray source is connected with the upper computer through a data line (such as an RSC232 data line), and emits X-rays to the conveying surface of the belt conveyor under the control of the upper computer; the X-ray detection device further comprises an X-ray detector and a data acquisition card connected with the X-ray detector, the X-ray detector converts penetrating X-rays into electric signals through a photoelectric effect, transmission intensity data are generated after the electric signals are acquired through the data acquisition card, and the transmission intensity data are sent to an upper computer through a data line (such as a network cable based on a TCP communication protocol).
In some embodiments, the upper computer adjusts the rotation speed of the motor and the on/off of the pneumatic valve respectively through a Programmable Logic Controller (PLC) to control the conveying speed of the conveying section and the real-time blowing frequency of the blowing section.
The following table 1 shows the equipment parameters of the intelligent gangue sorting system in a specific embodiment:
TABLE 1 Intelligent coal and gangue sorting system equipment parameters
The X-ray detector comprises 20 detection cards, each detection card can collect transmission intensity data of 64 sampling points, and the width of each sampling point is 1.5mm. Each sample of 20 probe cards can form 1280 × 1 transmission intensity data, which preferably includes high energy transmission intensity data and low energy transmission intensity data. In the above embodiment, the transmission intensity data may be mapped from an analog signal into an 8-bit digital signal (0 to 255) in accordance with the intensity range thereof.
Further, by continuously sampling at certain time intervals and then stitching the results of the multiple samplings, a transmission image can be formed. For example, in some embodiments, every 320 samples, a transmission image of 1280 × 320 pixels is formed, wherein the gray value of each pixel reflects the transmission intensity of the corresponding sampling point. The above techniques for analog-to-digital conversion based on signal strength data and for generating a grayscale image are known to those skilled in the art.
In the actual X-ray transmission and detection process, because the components of the material are complex and the distribution range of the mass, the density and the thickness of the material is large, a projection image constructed by simply adopting high-energy transmission intensity data or low-energy transmission intensity data can not accurately reflect the attenuation characteristic in the X-ray transmission process. For example, when the thickness of some coal gangue materials is larger, the high-energy part of the X-ray can not penetrate through the coal and gangue parts at the same time; when the thickness of some coal gangue materials is small, the low-energy part of the X-ray penetrates through the coal and gangue parts, and the problem that the strength attenuation difference is not large and the strength data discrimination is small may exist.
To this end, in some preferred embodiments of the present application, the transmission intensity data further comprises equivalent transmission intensity data; the transmission image is generated based on the equivalent transmission intensity data. Specifically, the above-described X-ray detection apparatus is based on the high energy transmission intensity I at each sampling point H And low energy transmission intensity I L Processing to obtain equivalent transmission intensity I corresponding to the sampling point e Through the equivalent transmission intensity I e And then, establishing a uniform transmission attenuation model of different coal gangue materials:
wherein, I e0 Is the initial value of the equivalent transmission intensity, T is the thickness of the coal gangue material, u e I.e. its corresponding equivalent attenuation coefficient. By utilizing the equivalent transmission model, the comprehensive attenuation characteristics of the coal gangue materials with different qualities, densities and thicknesses on the transmission of the high-energy part and the low-energy part of the X-ray can be effectively represented, so that the discrimination of the generated transmission image on different materials is greatly increased, and the edge detection and clustering processing in the subsequent target extraction process are facilitated. FIG. 2 illustrates a schematic view of transmission images of coal refuse material at multiple locations generated in accordance with a preferred embodiment of the present application.
Fig. 3 further shows a schematic diagram of the system frame of the upper computer in fig. 1 and data transmission between the system frame and the transmission part, the data acquisition part and the blowing part, as shown in fig. 3, in an embodiment of the present application, the upper computer includes a motor control unit, a data acquisition control unit, an image processing unit, a classification unit and a blowing control unit.
Specifically, the motor control unit is used for controlling the rotating speed of the motor; the data acquisition control unit is used for controlling the X-ray source to emit X-rays to the coal gangue material and receiving transmission intensity data and transmission images; the image processing unit extracts a coal gangue target from the transmission image; the classification unit calculates a detection vector of the coal and gangue target based on the transmission intensity data, and classifies the coal and gangue target according to the detection vector; the injection control unit controls the real-time injection frequency of the injection part based on the classification result of the coal and gangue targets. The connection between each unit and the transmission unit, the data acquisition unit and the blowing unit has been described above, and the following describes in detail the embodiments of the image processing unit, the classification unit and the blowing control unit in the upper computer with reference to the drawings and preferred embodiments.
In some embodiments of the present application, the image processing unit extracts a gangue target from the transmission image by:
s100, denoising the transmission image through spatial smoothing processing;
s200, generating a binary edge image based on an edge detection algorithm of multi-operator fusion;
s300, determining a coal and gangue target from the binarization edge image based on a clustering algorithm.
Specifically, the image processing unit first performs noise reduction processing on the transmission image through step S100, and as shown in fig. 2, the original transmission image acquired by the X-ray detection device contains much noise. The noise is generated due to various reasons, such as dust, moisture and debris covering the radiation tube, gaussian noise generated by the cross flow of the electronic circuit, sharp noise generated by the mode-to-electricity conversion of the probe card, etc., which are present in actual working conditions. The existence of the noise causes serious interference to subsequent target extraction and classification, so that the transmission image needs to be denoised firstly.
The analysis of the noise existing in the transmission image shows that the noise is frequency-independent noise, so that the noise can be performed in a spatial smoothing mode. The essence of filtering based on spatial smoothing is blurring, which blurs image details while reducing noise. Excessive smoothing has some effect on subsequent target positioning. Compared with a common spatial smoothing filtering algorithm through experiments, the median filtering is found to be capable of better overcoming image detail blurring caused by a linear filter (mean filtering), and has an obvious effect on filtering pulse interference and image scanning noise. In a preferred embodiment of the present application, therefore, median filtering is selected to reduce the noise of the transmission image. Fig. 4 shows the result of denoising the transmission image of fig. 2 using a 5 x 5 median filter template in some preferred embodiments.
After the transmission image is denoised, the image processing unit performs edge detection on the transmission image through the step S200; and then extracting the coal gangue target through the clustering algorithm of the step S300.
The target recognition and extraction is to divide the image into sub-images taking the material as the center according to the distribution of the material from the angle of the change of the image attribute, and support the subsequent target recognition. The obvious change of the image attribute is represented as image attribute discontinuity, such as gray level discontinuity, texture trend discontinuity, material attribute discontinuity and the like, and reflects the distribution change of the material in the image. Because the image processing unit in the embodiment of this application needs to carry out the discernment and the extraction of coal gangue target rapidly and accurately from the coal gangue material of continuous conveying, its requirement to the real-time has abandoned most image target positioning algorithm. In addition, in order to ensure the accuracy of coal and gangue target identification, the strength of image filtering and noise reduction operation is weak, more noise remains, and certain influence is brought to subsequent edge processing operation. Therefore, most irrelevant attribute information is removed through the edge detection algorithm with multi-operator fusion, the structural attribute information of important materials is reserved on the premise of guaranteeing the real-time performance of the algorithm, and the follow-up clustering and target extraction are guaranteed.
Specifically, in a preferred embodiment of the present application, the multi-operator fused edge detection algorithm performs edge detection on the transmission image using the following fractional differential operator:
where f (x, y) is the edge value at the image (x, y) coordinates, f 0 Is a fractional order differential operator of R-L, w 0 Is f 0 Weight of (f) l ,l∈[1,…,L]Is L different integer order differential operators, w l Is f l And satisfy
In particular, the R-L fractional order differential operator f 0 Can be obtained by Riemann-Liouville fractional order differential dispersion:
wherein the content of the first and second substances,
is an R-L fractional discrete operator, and f (t) is a continuous differentiable function near t; alpha and t are integral upper and lower limits of the differential operator, n is the minimum integer larger than alpha and is the differential order of the R-L fractional order differential operator, and alpha is larger than 0 and n-1 is larger than or equal to alpha and smaller than or equal to n.
In particular, the integer order differential operator f l May be one or more of Prewitt, laplacian, canny.
The image edge detection using common first-order differential operators, such as Robert operator, prewitt operator, sobel operator, canny operator, etc., is well known to those skilled in the art, however, the edge detection of the gangue transmission image using the above operators alone has certain drawbacks, such as: the Robert operator is accurate in positioning, but does not include smoothing processing, so that the Robert operator is sensitive to noise; the Prewitt operator and the Sobel operator both have differential operators with mean filtering property, the former is average filtering, the latter is weighted average filtering, and the detected image edge is possibly more than 2 pixels; canny's operator is highly resistant to noise at edges, but is prone to misinterpret isolated and clustered noise as boundaries. Therefore, the edge of the coal gangue target cannot be well extracted by using the first order differential operator alone.
Based on the reasons, the Robert operator with accurate positioning, the Prewitt operator with high noise resistance and the Sobel operator are selected from the integral order differential operators and are further fused with the R-L fractional order differential operator, and accurate material edge detection under complex noise is achieved.
Further, after acquiring the binary edge image, the image processing unit further extracts a coal and gangue target through the following steps:
s310, importing the binaryzation edge image;
s320, randomly selecting a plurality of edge points from the binaryzation edge image as a starting centroid;
s330, when the cluster allocation of any edge point changes, executing step S340 and step S350 in a loop, otherwise, jumping out of the loop, wherein,
s340, sequentially executing the following steps on each edge point in the edge image:
s341, for each centroid, respectively calculating connectivity among the centroids, the distance between the centroid and the cluster center and the distance between the centroid and the edge point;
s342, deleting the centroids communicated with other centroids and centroids surrounded by other clusters based on the calculation result;
s343, distributing the edge points to the clusters closest to the edge points;
s350, for each cluster, solving a cluster mean value and updating the cluster mean value into a mass center;
and S360, determining a coal gangue target based on the clustering result.
Through the above steps S310 to S360, different gangue targets may be extracted from the gangue material, and fig. 5 shows a transmission image and a plurality of extracted gangue targets in a specific embodiment, where each gangue target is surrounded by an edge line to form a corresponding region, and sampling points in the region may be used for further gangue classification.
However, due to the large size distribution span of different gangue objects in the gangue material, especially when X-rays irradiate a material with a large size, the problem of transmission intensity data distortion may occur. Through statistics on the shapes of coal gangue materials of various sizes, the materials are found to be wedge-shaped from the center to the edges, namely the center is thick and the edges are thin, and the center thickness increases with the increase of the sizes. The characteristics enable the transmission intensity data of the materials with different sizes and thicknesses to present different usability, for example, for a large-size material with a higher central thickness, an area with an overlarge thickness can not penetrate through the material, so that the transmission intensity data of the area can not truly reflect the attenuation characteristic of the X-ray; the X-ray penetration rate of the edge area is higher, but the transmission intensity is correspondingly influenced by noise to a higher degree; in addition, sampling points contained in the image of the larger material are too many, so that the classification calculation time is too long, the instantaneity of the sorting system is seriously influenced, and further the subsequent injection is influenced. Therefore, it is necessary to perform further sample point optimization on the gangue targets extracted in step S300.
Specifically, in some preferred embodiments of the present application, the optimization of the gangue target sampling point is performed by the following steps:
s410, dividing each coal gangue target into a plurality of sampling rings at equal intervals from a clustering center to a clustering edge;
s420, deleting the sampling ring on the outermost side of the coal gangue target;
s430, selecting a sampling ring for coal and gangue classification based on the area of the rest part of the coal and gangue target;
s440, traversing each sampling ring for coal gangue classification;
and S450, determining a coal and gangue target for coal and gangue classification based on the traversal result.
The number of the sampling rings in the above steps can be set according to the size distribution range of the coal and gangue target, the resolution of the coal and gangue target image and the like (for example, under the condition that the size of most materials is distributed in the range of [10mm,300mm ], the number of the divided sampling rings can be set to be 10-20), after the coal and gangue target is divided into a plurality of sampling rings, the edge part is removed firstly, then the sampling ring used for subsequent coal and gangue classification is selected according to the area of the rest part, then the sampling points which are not penetrated by X-rays (namely the sampling points with the transmission intensity of 0) are removed, and finally the optimized coal and gangue target used for subsequent coal and gangue classification is obtained.
FIG. 6 shows a schematic diagram of the sample ring division for a specific gangue target, which is divided into 10 sample rings from the cluster center to the cluster edge at equal intervals.
Table 2 below shows a sampling loop selection strategy under a partitioning scheme with a variable number of sampling loops.
Table 2 sampling loop selection strategy (division scheme with variable number of sampling loops)
After the image processing unit extracts the coal and gangue targets through the steps, the classification unit can classify the coal and gangue of each coal and gangue target. In an embodiment of the application, the classification of the coal and gangue targets is based on the coal and gangue exhibiting different attenuation characteristics for the transmission of X-rays.
Statistics show that the true density of coal is about 1.3-1.8, and the true density of gangue is about 1.8-2.3. The density of the coal and the gangue is greatly different, and the difference enables the penetration attenuation of the coal and the gangue to the X-ray to show different characteristics, so that the coal and the gangue can be identified and sorted by the X-ray. In an embodiment of the application, the classification unit classifies the gangue targets by the following steps:
A100. determining an energy detection vector for a coal gangue target based on the transmission intensity data
Wherein m is the number of sampling points in the coal gangue target,
low energy transmitted intensity, R, for the ith sample point i For the ith sample point determined byClassification characteristics:
wherein the content of the first and second substances,
initial values for high energy transmission intensity and low energy transmission intensity,
high energy transmission intensity for the ith sample point;
A200. calculating a classification correlation coefficient r of the coal and gangue target based on the energy detection vector j ,
Wherein c is coal, g is gangue,
is composed of
Mean value of (1) i,j To be R i Fitting curve of coal/gangue
The estimated value of the current value obtained is,
is shown as I i,j The mean value of (a);
A300. and classifying the coal and gangue targets through a preset target identification strategy based on the classification correlation coefficient.
Specifically, step a100 is used to construct energy detection vectors of each sampling point of each gangue target, wherein,
the high-energy X-ray and the low-energy X-ray can be detected and obtained under the condition of no coal gangue material.
For coal and gangue with different transmission attenuation characteristics, the energy detection vectors thereof show different R-I L FIG. 7 shows the distribution results of the energy detection vectors of coal and gangue obtained by a plurality of actual measurements, and it can be seen from FIG. 7 that the energy detection vectors of coal and gangue are in the ranges from R to I L Obvious clustering characteristics are shown in the characteristic space, and the separation degree of the aggregation areas is good, so that the energy detection vector constructed by the method can be effectively applied to classification and identification of the coal and the gangue.
For convenience of classification of coal and gangue, the classification can be carried out by y = a · e bx Performing curve fitting on a plurality of actually measured energy detection vectors of the coal and the gangue through the functional relation to obtain a fitting curve of the coal
Fitting curve with gangue
Then, the classification correlation coefficient r of each coal and gangue target corresponding to coal and gangue is respectively obtained through the step A200 c And r g 。
R at each coal gangue target c And r g Then, the gangue targets may be classified using a preset target identification strategy in step a300. In some preferred embodiments of the present application, the preset target recognition policy includes:
simple classification strategy if r c ＞r g If the coal and gangue target is coal, otherwise, the coal and gangue target is gangue;
reducing the gangue content if r c If the target is greater than the first threshold value, the coal gangue target is coal, otherwise, the coal gangue target is gangue;
reducing coal content strategy if g If the second threshold value is greater than the first threshold value, the coal and gangue target is the gangue, otherwise, the coal and gangue target is the coal.
The different target recognition strategies described above correspond to different sorting requirements, wherein,simple classification strategy by simply pairing r c And r g The sizes of the coal and gangue targets are compared to classify the coal and gangue targets, and the method can be used in common coal and gangue separation occasions; the coal and gangue containing strategy is reduced by identifying the coal and gangue targets of the first threshold value which is less than or equal to the first threshold value as the gangue, so that the coal content of the coal and gangue targets identified as coal can be improved, the coal and gangue containing strategy is suitable for occasions needing to extract high-quality coal, the coal and gangue containing strategy is reduced by identifying the targets greater than the second threshold value as the gangue, the coal and gangue target content identified as the gangue can be improved, and the coal and gangue containing strategy is suitable for occasions where effective mineral components in the material with high coal gangue content are identified and separated. In a specific implementation process, the first threshold and the second threshold can be determined based on sampling detection data of different batches of coal gangue materials and combined with actual sorting requirements.
After the classification unit classifies all the coal and gangue targets continuously transmitted on the conveyor belt, the injection control unit can combine different coal and gangue targets to reach the injection part of the conveyor belt along the conveying direction according to the classification result, and the real-time injection frequency of a plurality of pneumatic valves arranged at the tail end of the conveyor belt can be passed through.
In some optional embodiments, the real-time blowing and spraying frequency of the pneumatic valves can be determined according to the proportion of coal and gangue in the classification result, for example, when a coal gangue material with the proportion (quantity proportion or area proportion) of coal smaller than a certain value reaches the end of the conveyor belt, the pneumatic valves are started to blow, so that the material at the moment is blown away, otherwise, the blowing and spraying are stopped, so that the material directly falls, and thus, the coal gangue separation is realized.
In the actual blowing control process, because a single pneumatic valve cannot realize continuous blowing on continuously coming gangue, a blowing part is generally required to be formed by a plurality of pneumatic valves, and the blowing control unit controls the pneumatic valves to blow alternately. However, there is a time delay in the mechanical operation of the pneumatic valve and the control of the pneumatic valve becomes a joint point that restricts the blowing performance of the system, namely: if the real-time blowing and spraying frequency of the plurality of pneumatic valves is set unreasonably, the gangue cannot be blown in time or the pneumatic valves are idle for a long time for waiting. In order to avoid the wrong large spraying and the small large spraying belt, the implementation blowing frequency of the pneumatic valve needs to be optimized according to the classification result of the gangue of the material reaching the tail end of the conveying belt.
The essence of the injection optimization is an optimized scheduling problem for queuing and injecting gangue targets, namely, the minimum pneumatic valve capable of meeting the injection requirement is found in queuing, so that the injection cost and index loss are minimum under the condition of meeting the injection condition, and the injection system efficiency is highest. The randomness of the arrival sequence and time of the gangue can be described through probability, and an optimization solution is found through a queuing theory.
Specifically, in some preferred embodiments of the present application, the blowing control unit determines the real-time blowing frequency of the blowing section by:
B100. establishing a gangue injection queuing model (X, Y, Z, A, B and C), wherein X is gangue arrival time interval distribution, Y is injection time distribution, Z is the number of pneumatic valves, A is the maximum queuing number allowed by an optimized queuing system, B is the total quantity of gangue, and C is a queuing rule;
B200. determining injection queuing index (L) based on gangue injection queuing model s ,L q ,W s ,W q ,P n λ, μ, ρ) of L s To average number of gangue reached, L q To average queue length, W s To average latency, W q Average wait time for queue, P n The probability of stable injection of n pieces of waste rocks is shown, lambda is the average number of the waste rocks arriving in unit time, mu is the number of the waste rocks injected by a single pneumatic valve each time, and rho = lambda/mu is the real-time blowing frequency of a blowing part;
B300. substituting the real-time classification result of the coal gangue materials reaching the tail end of the conveyor belt into the injection queuing index, and determining the optimal value of rho based on the following optimization targets:
wherein z is the total revenue function of blowing, C s The cost coefficient of jetting is carried out for the pneumatic valve, and G is the income of a piece waste rock of jetting.
Specifically, the arrival of the waste rocks along with the conveyor belt can be regarded as infinite, the waste rocks are not correlated, the arrival times of the waste rocks are independent, and the arrival rules are random; furthermore, in the delta t time, the probability of arrival of a piece of gangue is set to be lambda delta t; the probability of no gangue arriving is 1-lambda delta t; treating delta t as infinitesimal small, and simultaneously enabling the number of arrived waste rocks to be less than or equal to 1; the gangue reaches compliance lambda poise.
And establishing a queuing model (X, Y, Z, A, B and C) based on the setting, wherein X is the arrival time interval distribution of the gangue, Y is the injection time distribution, Z is the number of pneumatic valves, A is the maximum allowable queuing number of an optimized queuing system, B is the total quantity of the gangue, and C is a queuing rule. In actual sorting, different queuing rules can be set according to different sorting targets, such as: first comes and blows, big gangue takes precedence, gangue which is not adjacent to coal takes precedence, and the like.
Further, a blowing queuing index (L) is determined based on the queuing model s ,L q ,W s ,W q ,P n λ, μ, ρ) of L s To average number of gangue arrivals, L q To average queue length, W s Is the average latency time, W q Average wait time for queue, P n The probability of stable blowing of n pieces of waste rocks is shown, lambda is the average number of the waste rocks arriving in unit time, mu is the number of the waste rocks blown by a single pneumatic valve each time, and rho = lambda/mu is the real-time blowing frequency of a blowing part.
The blowing queuing indexes have the following relations: l is s ＝λW s ；L q ＝λW q ；L s ＝L q +λ/μ；W s ＝W q +1/μ. Wherein when the mine spoil falls from the conveyor, the corresponding pneumatic valve is identified. The plurality of gangue corresponds to the same pneumatic valve and is blown first. The valve body action time is relatively independent, the first blowing is carried out first, and the valve body action time is relatively independent. Obeying a negative exponential distribution of μ.
Further, in the actual sorting process, substituting the real-time classification result of the coal gangue materials reaching the tail end of the conveyor belt into the injection queuing index, and respectively calculating the probability of stable injection of the system: p n ＝(1-ρ)ρ n Optimizing queuing indexes: l is s ＝λ/(μ-λ)，L q ＝ρλ/(μ-λ)，W s ＝1/(μ-λ)，W q = ρ/(μ - λ) and optimization parameters: effective arrival rate λ e ＝λ(1-P n ) Effective blowing strength
Average number of gangue reached
Average queue length L q ＝L s -(1-P 0 ) Average waiting blowing duration
Average waiting time of queue
Finally, parameters Cs and G are introduced, where C s And G is the yield of a piece of gangue for blowing. Searching for the maximum total blowing yield z through the queuing indexes and the parameters, namely meeting rho of the following formula:
in some specific embodiments, the total revenue blowing function z can be expressed in an iterative fashion as:
wherein z is (k) 、z (k+1) And d, iteratively calculating rho by continuously changing rho of the total injection income respectively calculated for the kth iterative computation and the kth +1 th iterative computation, so that the total injection income gradually approaches to the maximum value in an iterative manner, stopping iterative computation when the sum of the two adjacent iterative computation results is less than a preset threshold value, and taking the rho at the moment as an optimal value.
The blowing scheme with the maximum total blowing income is the blowing optimization queuing result. The time complexity of the queuing optimization algorithm is O (nlogn). Due to the limitation of the width and the speed of the belt, the queuing optimization time is less than 100ms, and the real-time blowing requirement is met. When the width and the speed of the belt are increased, the queuing index of queuing optimization can be simplified, so that the queuing optimization time is further reduced.
While the present invention has been described in detail and with reference to specific embodiments thereof, it will be apparent to one skilled in the art that various changes and modifications can be made therein without departing from the spirit and scope thereof as defined in the appended claims.
Claims (12)
1. The utility model provides a coal gangue intelligence sorting system, includes conveying part, data acquisition portion, jetting portion and host computer, its characterized in that:
the conveying part comprises a motor and a belt conveyor, and the belt conveyor conveys the coal gangue materials under the driving of the motor;
the data acquisition part is used for emitting X rays to the coal gangue material and acquiring transmission intensity data and a transmission image;
the injection part is arranged at the tail end of the belt conveyor along the conveying direction, comprises a plurality of pneumatic valves and is used for injecting the coal gangue materials to realize the separation of coal and gangue;
and the upper computer classifies the coal gangue materials based on the transmission intensity data and the transmission image and determines the real-time injection frequency of the injection part.
2. The intelligent gangue sorting system according to claim 1, wherein:
the data acquisition part comprises an X-ray source and an X-ray detection device which are oppositely arranged on two sides of the conveying surface of the belt conveyor;
the X-ray detection device is used for acquiring the transmission intensity data and the transmission image, wherein the transmission intensity data comprises high-energy transmission intensity data and low-energy transmission intensity data.
3. The intelligent coal gangue sorting system according to claim 2, wherein the upper computer comprises:
the system comprises a motor control unit, a data acquisition control unit, an image processing unit, a classification unit and a blowing control unit;
the motor control unit is used for controlling the rotating speed of the motor;
the data acquisition control unit is used for controlling the X-ray source to emit X-rays to the coal gangue material and receiving the transmission intensity data and the transmission image acquired by the X-ray detection device;
the image processing unit extracts a coal gangue target from the transmission image;
the classification unit calculates a detection vector of the coal and gangue target based on the transmission intensity data, and classifies the coal and gangue target according to the detection vector;
the injection control unit controls the real-time injection frequency of the injection part based on the classification result of the coal and gangue targets.
4. The intelligent gangue sorting system of claim 3, wherein:
the transmission intensity data further comprises equivalent transmission intensity data;
the transmission image is generated based on the equivalent transmission intensity data.
5. The intelligent gangue sorting system of claim 3, wherein the image processing unit extracts gangue targets from the transmission images by:
s100, denoising the transmission image through spatial smoothing;
s200, generating a binary edge image based on an edge detection algorithm of multi-operator fusion;
s300, determining a coal and gangue target from the binarization edge image based on a clustering algorithm.
6. The intelligent gangue sorting system according to claim 5, wherein:
the spatial smoothing process is to filter the transmission image using a median filter template.
7. The intelligent coal and gangue sorting system according to claim 5, wherein the multi-operator fused edge detection algorithm performs edge detection on the transmission images using the following fractional differential operators:
where f (x, y) is the edge value at the image (x, y) coordinates, f 0 Is R-L fractional order differential operator, w 0 Is f 0 Weight of (a), f l ,l∈[1,…,L]Is L different integer order differential operators, w l Is f l And satisfy
8. The intelligent gangue sorting system according to claim 5, wherein the step S300 further comprises the following steps:
s310, importing the binaryzation edge image;
s320, randomly selecting a plurality of edge points from the binarization edge image as an initial centroid;
s330, when the cluster distribution of any edge point changes, the step S340 and the step S350 are executed circularly, otherwise, the loop is skipped, wherein,
s340, sequentially executing the following steps on each edge point in the edge image:
s341, respectively calculating connectivity among centroids, distances between the centroids and cluster centers and distances between the centroids and edge points for each centroid;
s342, deleting the centroids communicated with other centroids and centroids surrounded by other clusters based on the calculation result;
s343, distributing the edge points to the clusters nearest to the edge points;
s350, for each cluster, solving a cluster mean value and updating the cluster mean value into a mass center;
and S360, determining a coal gangue target based on the clustering result.
9. The intelligent gangue sorting system according to claim 5, wherein:
after the image processing unit determines the coal and gangue target, the image processing unit also optimizes sampling points of the coal and gangue target, and the optimization of the sampling points further comprises the following steps:
s410, dividing each coal and gangue target into a plurality of sampling rings at equal intervals from a clustering center to a clustering edge;
s420, deleting the sampling ring on the outermost side of the coal gangue target;
s430, selecting a sampling ring for coal and gangue classification based on the area of the rest part of the coal and gangue target;
s440, traversing each sampling ring for coal gangue classification;
and S450, determining a coal and gangue target for coal and gangue classification based on the traversal result.
10. The intelligent gangue sorting system according to claim 3, wherein the classifying unit classifies gangue targets by:
A100. determining an energy detection vector for a coal gangue target based on the transmission intensity data
Wherein m is the number of sampling points in the coal gangue target,
low energy transmitted intensity, R, for the ith sample point i Is the classification characteristic of the ith sample point determined by:
wherein the content of the first and second substances,
initial values for high energy transmission intensity and low energy transmission intensity,
high energy transmission intensity for the ith sample point;
A200. calculating a classification correlation coefficient r of the coal and gangue target based on the energy detection vector j ,
Wherein c is coal, g is gangue,
is composed of
Mean value of (1) i,j To be R i Fitting curve of coal/gangue
The value of the estimated value obtained is,
is I i,j The mean value of (a);
a300: and classifying the coal and gangue targets through a preset target identification strategy based on the classification correlation coefficient.
11. The intelligent gangue sorting system of claim 10, wherein the target identification policy is one of the following policies:
simple classification strategy if c ＞r g If the coal and gangue target is coal, otherwise, the coal and gangue target is gangue;
reducing the gangue content if r c If the target is greater than the first threshold value, the coal and gangue target is coal, otherwise, the coal and gangue target is gangue;
reducing coal content strategy if g If the second threshold value is larger than the first threshold value, the coal gangue target is gangue, otherwise, the coal gangue target is coal.
12. The intelligent coal gangue sorting system according to claim 3, wherein the injection control unit determines the real-time injection frequency of the injection part by the following steps:
B100. establishing a gangue injection queuing model (X, Y, Z, A, B and C), wherein X is gangue arrival time interval distribution, Y is injection time distribution, Z is the number of pneumatic valves, A is the maximum queuing number allowed by an optimized queuing system, B is the total quantity of gangue, and C is a queuing rule;
B200. determining injection queuing index (L) based on gangue injection queuing model s ,L q ,W s ,W q ,P n λ, μ, ρ) of L s To average number of gangue reached, L q To average queue length, W s To average latency, W q Average waiting time for queue, P n The probability of stable injection of n pieces of waste rocks is shown, lambda is the average number of the waste rocks arriving in unit time, mu is the number of the waste rocks injected by a single pneumatic valve each time, and rho = lambda/mu is the real-time blowing frequency of a blowing part;
B300. substituting the real-time classification result of the coal gangue materials reaching the tail end of the conveyor belt into the injection queuing index, and determining the optimal value of rho based on the following optimization targets:
wherein z is the total revenue function of blowing, C s And G is the yield of a piece of gangue for blowing.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211266735.XA CN115532649A (en) | 2022-10-17 | 2022-10-17 | Intelligent coal gangue sorting system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211266735.XA CN115532649A (en) | 2022-10-17 | 2022-10-17 | Intelligent coal gangue sorting system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115532649A true CN115532649A (en) | 2022-12-30 |
Family
ID=84735797
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211266735.XA Pending CN115532649A (en) | 2022-10-17 | 2022-10-17 | Intelligent coal gangue sorting system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115532649A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116140243A (en) * | 2023-04-18 | 2023-05-23 | 北京霍里思特科技有限公司 | Mining blowing sorting method, sorting system, equipment and storage medium |
| CN116871177A (en) * | 2023-09-05 | 2023-10-13 | 国擎(山东)信息科技有限公司 | Method and system for separating kaolin crude ore based on multispectral technology |
| CN117571543A (en) * | 2024-01-16 | 2024-02-20 | 清华大学 | Method and system for online measurement of true density of bulk material by utilizing X/gamma rays |
-
2022
- 2022-10-17 CN CN202211266735.XA patent/CN115532649A/en active Pending
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116140243A (en) * | 2023-04-18 | 2023-05-23 | 北京霍里思特科技有限公司 | Mining blowing sorting method, sorting system, equipment and storage medium |
| CN116140243B (en) * | 2023-04-18 | 2023-08-15 | 北京霍里思特科技有限公司 | Mining blowing sorting method, sorting system, equipment and storage medium |
| CN116871177A (en) * | 2023-09-05 | 2023-10-13 | 国擎(山东)信息科技有限公司 | Method and system for separating kaolin crude ore based on multispectral technology |
| CN116871177B (en) * | 2023-09-05 | 2023-11-17 | 国擎(山东)信息科技有限公司 | Method and system for separating kaolin crude ore based on multispectral technology |
| CN117571543A (en) * | 2024-01-16 | 2024-02-20 | 清华大学 | Method and system for online measurement of true density of bulk material by utilizing X/gamma rays |
| CN117571543B (en) * | 2024-01-16 | 2024-04-09 | 清华大学 | Method and system for online measurement of true density of bulk material by utilizing X/gamma rays |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115532649A (en) | Intelligent coal gangue sorting system | |
| CN205289011U (en) | Building rubbish sorting device based on machine vision | |
| CN114445387A (en) | Fiberboard quality classification method based on machine vision | |
| WO2016172827A1 (en) | Stepwise-refinement pavement crack detection method | |
| CN108971190A (en) | A kind of separating domestic garbage method based on machine vision | |
| CN206139527U (en) | Panoramic vision potato is selected separately and defect detecting device | |
| CN102855641A (en) | Fruit level classification system based on external quality | |
| CN105447514B (en) | A kind of metal recognition methods based on comentropy | |
| CN107121436B (en) | The Intelligent detecting method and identification device of a kind of silicon material quality | |
| CN102253050A (en) | Automatic detection method and device for magnetic tile surface defect based on machine vision | |
| CN111277243A (en) | Photon counting laser radar adaptive filtering algorithm for water depth extraction | |
| CN109886277B (en) | Contour analysis-based fresh tea leaf identification method | |
| CN111215342A (en) | Industrial garbage classification and sorting system | |
| CN112893159B (en) | Coal gangue sorting method based on image recognition | |
| CN110766702A (en) | Agaricus bisporus grading judgment method | |
| CN109916826A (en) | A kind of solid waste online recognition system and recognition methods based on EO-1 hyperion detection | |
| CN110243732B (en) | Whole-course ore pulp grinding granularity online detection system with granularity range of 2-500um | |
| CN113552028B (en) | Mixture visual analysis device, and method and system for detecting granularity composition of mixture | |
| CN201997485U (en) | Garbage auto-separation device | |
| CN116630813A (en) | Highway road surface construction quality intelligent detection system | |
| CN115661646A (en) | Intelligent garbage classification method based on computer vision | |
| CN104899595A (en) | Male and female silkworm chrysalis sorting and counting device based on SIFT (Scale Invariant Feature Transform) feature image | |
| CN115266728A (en) | Method for identifying inclusion defects on surface of cold-rolled strip steel | |
| CN113743523B (en) | Building rubbish fine classification method guided by visual multi-feature | |
| CN111539920A (en) | Automatic detection method for fermented grain quality in white spirit brewing process |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |