CN113591583A - Intelligent boron ore beneficiation system and method - Google Patents
Intelligent boron ore beneficiation system and method Download PDFInfo
- Publication number
- CN113591583A CN113591583A CN202110744017.8A CN202110744017A CN113591583A CN 113591583 A CN113591583 A CN 113591583A CN 202110744017 A CN202110744017 A CN 202110744017A CN 113591583 A CN113591583 A CN 113591583A
- Authority
- CN
- China
- Prior art keywords
- image data
- image
- boron ore
- boron
- ore
- 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
- ZOXJGFHDIHLPTG-UHFFFAOYSA-N Boron Chemical compound [B] ZOXJGFHDIHLPTG-UHFFFAOYSA-N 0.000 title claims abstract description 328
- 229910052796 boron Inorganic materials 0.000 title claims abstract description 328
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000005456 ore beneficiation Methods 0.000 title claims abstract description 29
- 238000013145 classification model Methods 0.000 claims abstract description 89
- 238000012545 processing Methods 0.000 claims abstract description 55
- 238000007781 pre-processing Methods 0.000 claims abstract description 34
- 238000001514 detection method Methods 0.000 claims abstract description 24
- 238000012549 training Methods 0.000 claims description 23
- 238000012360 testing method Methods 0.000 claims description 21
- 238000013527 convolutional neural network Methods 0.000 claims description 20
- 238000000605 extraction Methods 0.000 claims description 12
- 230000003287 optical effect Effects 0.000 claims description 12
- 238000005516 engineering process Methods 0.000 claims description 11
- 229910021538 borax Inorganic materials 0.000 claims description 9
- 239000004328 sodium tetraborate Substances 0.000 claims description 9
- 235000010339 sodium tetraborate Nutrition 0.000 claims description 9
- 229910021539 ulexite Inorganic materials 0.000 claims description 8
- 238000001914 filtration Methods 0.000 claims description 6
- 239000004575 stone Substances 0.000 claims description 6
- KGBXLFKZBHKPEV-UHFFFAOYSA-N boric acid Chemical compound OB(O)O KGBXLFKZBHKPEV-UHFFFAOYSA-N 0.000 claims description 5
- 239000004327 boric acid Substances 0.000 claims description 5
- 238000010606 normalization Methods 0.000 claims description 5
- 229910021540 colemanite Inorganic materials 0.000 claims description 4
- -1 whitish boroalite Inorganic materials 0.000 claims description 3
- 239000002932 luster Substances 0.000 claims description 2
- 230000008569 process Effects 0.000 abstract description 19
- 239000003086 colorant Substances 0.000 description 10
- 230000006870 function Effects 0.000 description 8
- 230000004913 activation Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 239000011159 matrix material Substances 0.000 description 5
- 238000013528 artificial neural network Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 4
- QYHKLBKLFBZGAI-UHFFFAOYSA-N boron magnesium Chemical compound [B].[Mg] QYHKLBKLFBZGAI-UHFFFAOYSA-N 0.000 description 4
- 210000002569 neuron Anatomy 0.000 description 4
- 238000013135 deep learning Methods 0.000 description 3
- 239000011435 rock Substances 0.000 description 3
- BCWKKHAIFMOLTP-UHFFFAOYSA-N [Fe].[Mg].[B] Chemical compound [Fe].[Mg].[B] BCWKKHAIFMOLTP-UHFFFAOYSA-N 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 239000013078 crystal Substances 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000011056 performance test Methods 0.000 description 2
- 230000000737 periodic effect Effects 0.000 description 2
- 238000011176 pooling Methods 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 241000219312 Chenopodium Species 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 239000004566 building material Substances 0.000 description 1
- 239000013064 chemical raw material Substances 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 150000001875 compounds Chemical class 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 229910052500 inorganic mineral Inorganic materials 0.000 description 1
- 239000011707 mineral Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- CVPJXKJISAFJDU-UHFFFAOYSA-A nonacalcium;magnesium;hydrogen phosphate;iron(2+);hexaphosphate Chemical compound [Mg+2].[Ca+2].[Ca+2].[Ca+2].[Ca+2].[Ca+2].[Ca+2].[Ca+2].[Ca+2].[Ca+2].[Fe+2].OP([O-])([O-])=O.OP([O-])([O-])=O.[O-]P([O-])([O-])=O.[O-]P([O-])([O-])=O.[O-]P([O-])([O-])=O.[O-]P([O-])([O-])=O.[O-]P([O-])([O-])=O.[O-]P([O-])([O-])=O CVPJXKJISAFJDU-UHFFFAOYSA-A 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 229910052591 whitlockite Inorganic materials 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computational Linguistics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention provides an intelligent boron ore beneficiation system and method, and the system comprises: the image acquisition module is used for acquiring image data of at least one boron ore and storing the image data into an image database; the image processing module is used for carrying out image preprocessing on the image data in the image database to select effective image data and extracting color information and image characteristic elements corresponding to the effective image data; and the intelligent classification module is used for inputting the color information and the image characteristic elements corresponding to any effective image data into a pre-trained target boron ore classification model, and performing boron ore classification on the boron ore corresponding to the effective image data by using the target boron ore classification model to obtain the boron ore type of the boron ore. In the boron ore beneficiation process, the method realizes the streamlined, modularized, automatic and efficient boron ore type identification, improves the accuracy of boron ore type detection, and saves the labor and time cost.
Description
Technical Field
The invention relates to the technical field of boron ore beneficiation, in particular to an intelligent boron ore beneficiation system and method.
Background
Boron ore is an important chemical raw material in the world, is widely distributed in nature, and more than 100 boron-containing minerals are known at present. Boron ore is mainly used for producing borax, boric acid and various boron-containing compounds, and the borides are widely applied to various departments of national economy, such as building material industry, metallurgical industry, mechanical industry, electronic industry and the like. At present, the traditional manual discrimination mode is mainly adopted for boron ore dressing. The color, shape, texture, glossiness and other aspects of the boron ore need to be analyzed and identified manually to distinguish different ore types. The main problems existing in the traditional manual identification mode are as follows: strong subjectivity, low working efficiency, limited accuracy and high working strength. Today, industrial intelligence is pursued, and the method cannot meet the requirement of industrial development increasingly.
Disclosure of Invention
In view of the above problems, the invention provides an intelligent boron ore beneficiation system and method, which acquire image data of boron ore, perform image preprocessing on the image data to obtain color information and image characteristic elements of effective image data, perform boron ore classification on the boron ore by using a target boron ore classification model to obtain the boron ore type of the boron ore, realize automatic intelligent detection on the boron ore type, solve the problems of low efficiency and large error in the boron ore beneficiation process, improve the accuracy and working efficiency of the boron ore beneficiation, save manpower and time cost, reduce the dependence on manual detection, and realize the flow and modularization of the boron ore beneficiation process.
According to a first aspect of the present invention, there is provided an intelligent boron ore beneficiation system, comprising:
the image acquisition module is used for acquiring image data of at least one boron ore and storing the image data into an image database;
the image processing module is used for carrying out image preprocessing on the image data in the image database to select effective image data and extracting color information and image characteristic elements corresponding to the effective image data;
and the intelligent classification module is used for inputting color information and image characteristic elements corresponding to any effective image data into a pre-trained target boron ore classification model, and performing boron ore classification on the boron ore corresponding to the effective image data by using the target boron ore classification model to obtain the boron ore type of the boron ore.
Optionally, the image processing module comprises an image preprocessing unit; the image preprocessing unit is used for:
carrying out image preprocessing on the image data one by one, and judging whether the image data is effective image data;
when the image data is effective image data, saving the effective image data;
when the image data is invalid image data, ignoring the invalid image data;
the preprocessing comprises at least one of image graying, image filtering processing and image binarization processing.
Optionally, the image preprocessing unit is further configured to:
carrying out graying processing on the image data to obtain a grayscale image corresponding to the image data;
carrying out binarization processing on the gray-scale image by utilizing an im2bw function, extracting bright spot features in the gray-scale image, and judging whether the area of the bright spots in the gray-scale image is larger than a preset pixel point or not;
when the area of the bright spot in the gray image is less than or equal to a preset pixel point, the image data is effective image data; and when the area of the bright spot in the gray-scale image is larger than the preset pixel point, the image data is invalid image data.
Optionally, the image processing module comprises:
the color information extraction unit is used for acquiring RGB color values corresponding to the effective image data, and storing the RGB color values as color information of the effective image data;
the image characteristic element extraction unit is used for extracting and storing image characteristic elements corresponding to the effective image data by utilizing a characteristic extraction algorithm; wherein the image characteristic elements comprise at least one of a gray mean, a variance, smoothness, skewness, an energy value and an entropy value.
Optionally, the intelligent classification module is further configured to:
generating a feature element set according to the color information and the image feature elements corresponding to the effective image data, and performing data normalization processing on the feature elements in the feature element set to obtain target detection data;
and inputting the target detection data into a pre-trained target boron ore classification model, and intelligently classifying the boron ore to be classified by using the target boron ore classification model according to the target detection data to obtain the boron ore type of the boron ore.
Optionally, the image processing module is further configured to: periodically judging whether newly added image data exists in the image database; when newly added image data exist in the image database, performing image preprocessing on the newly added image data to select newly added effective image data, and extracting color information and image characteristic elements corresponding to the newly added effective image data;
the intelligent classification module is further configured to: and periodically detecting newly added effective image data and corresponding color information and image characteristic elements thereof, and performing boron ore classification on the boron ore corresponding to the newly added effective image data by using the target boron ore classification model to obtain the boron ore type of the boron ore.
Optionally, the system further comprises a model building module;
the model building module is configured to:
building a boron ore classification model based on a convolutional neural network technology, and training, testing and verifying the boron ore classification model by using preset sample characteristic elements and sample class labels of a boron ore image data sample library to obtain a target boron ore classification model;
wherein the sample characteristic element comprises at least one of color, shape, texture and luster; the sample class label is a boron ore class name, including but not limited to, boromagnesite, ludwigite, natural boric acid, borax dehydrate, ulexite, colemanite, whitish boroalite, ulexite, manboracite, natural borax, tunnel stone, and pindolite.
Optionally, the intelligent classification module is further configured to: storing image data and boron ore types corresponding to the boron ores;
the model building module is further configured to: periodically acquiring image data and a boron ore type corresponding to the newly added boron ore, and updating sample characteristic elements and sample category labels in a boron ore image data sample library according to the image data and the boron ore type corresponding to the newly added boron ore; and self-learning the target boron mine classification model by using the updated sample characteristic elements and the updated sample class labels, and updating the target boron mine classification model.
Optionally, the image acquisition module comprises:
the image acquisition equipment is used for acquiring optical signals of the boron ore;
the image sensing equipment is electrically connected with the image acquisition equipment and is used for converting optical signals corresponding to the boron ore into electric signals;
and the image signal processor is electrically connected with the image sensing equipment, is used for carrying out signal processing based on the electric signal to obtain image data of the boron ore, and stores the image data into the image database.
According to a second aspect of the invention, an intelligent boron ore beneficiation method is provided, which comprises the following steps:
acquiring image data of at least one boron ore and storing the image data into an image database;
image preprocessing is carried out on image data in the image database to select effective image data, and color information and image characteristic elements corresponding to the effective image data are extracted;
and inputting color information and image characteristic elements corresponding to any effective image data into a pre-trained target boron ore classification model, and performing boron ore classification on the boron ore corresponding to the effective image data by using the target boron ore classification model to obtain the boron ore type of the boron ore.
The invention provides an intelligent boron ore beneficiation system and method based on an image processing technology, in the recorded scheme, the image data of boron ore is preprocessed, effective image data is selected, color information and image characteristic elements of the effective image data are extracted, a target boron ore classification model based on a convolutional neural network technology is constructed, the boron ore type of the boron ore is analyzed and detected by using the target boron ore classification model, efficient, flow and modularized automatic detection of the boron ore type in the boron ore beneficiation process is realized, the boron ore image data characteristic elements and boron ore type labels are stored by establishing a boron ore image data sample library, the boron ore type detection capability is improved by self-learning of the target boron ore classification model, the labor cost is greatly saved, and the error rate of the boron ore type detection is reduced.
The foregoing description is only an overview of the technical solutions of the present invention, and the embodiments of the present invention are described below in order to make the technical means of the present invention more clearly understood and to make the above and other objects, features, and advantages of the present invention more clearly understandable.
The above and other objects, advantages and features of the present invention will become more apparent to those skilled in the art from the following detailed description of specific embodiments thereof, taken in conjunction with the accompanying drawings.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:
fig. 1 is a schematic structural diagram of an intelligent boron ore beneficiation system according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of an intelligent boron ore beneficiation system according to another embodiment of the present invention;
FIG. 3 is a schematic structural diagram of an image acquisition module according to another embodiment of the present invention;
FIG. 4 shows a schematic structural diagram of a boron mine classification model provided by an embodiment of the invention;
fig. 5 shows a schematic flow chart of an intelligent boron ore beneficiation method according to an embodiment of the present invention.
Detailed Description
In order to make the technical solutions better understood by those skilled in the art, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only partial embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application. It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict.
It should be noted that the terms "first," "second," and the like in the description and claims of this application and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the application described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
An embodiment of the present invention provides an intelligent boron rock beneficiation system, as shown in fig. 1, the system may include: an image acquisition module 110, an image processing module 120, and an intelligent classification module 130.
The image acquisition module 110 is configured to acquire image data of at least one boron ore and store the image data in an image database.
The image processing module 120 is configured to perform image preprocessing on the image data in the image database to select effective image data, and extract color information and image feature elements corresponding to the effective image data.
The image preprocessing may include graying, image filtering, image binarization, etc., restoring image distortion caused by transfer of image data between different media, and denoising and interference removing the image data, so as to facilitate selection of effective image data.
The effective image data refers to image data which can reflect that the image characteristic value is in accordance with the actual situation. The color information refers to RGB values of the image data colors. The image characteristic elements refer to crystal shape, texture, gray level mean, variance, smoothness, skewness, energy value, entropy value and the like of image data.
And the intelligent classification module 130 is configured to input color information and image feature elements corresponding to any effective image data into a pre-trained target boron ore classification model, and perform boron ore classification on boron ores corresponding to the effective image data by using the target boron ore classification model to obtain boron ore types of the boron ores.
The boron mine type in the embodiment of the invention has the characteristics of multiple types, difficulty in identification by naked eyes and the like, and in order to improve the accuracy of boron mine type identification, a Convolutional Neural Network (Convolutional Neural Network) technology is used for constructing a boron mine classification model. CNN is a feedforward neural network, the artificial neurons of the network can locally respond to the surrounding neurons, the nodes of the neurons in the previous layer are output to the neurons in the next layer after passing through a nonlinear activation function, and no information is transmitted among the nodes in the same layer. Each convolution layer in the convolution neural network consists of a plurality of convolution units, and the parameter of each convolution unit is obtained through optimization of a back propagation algorithm. The low-level network structure obtains the input local features through convolution operation, and the high-level network convolutes the local features into the full local features through convolution operation. Therefore, when performing convolution operations, deeper convolution layers are typically designed to extract deep image features. And (3) constructing a target boron ore classification model based on the convolutional neural network technology by adopting the theoretical basis of the CNN convolutional neural network technology in the field of deep learning, and performing boron ore classification on the boron ore through the model to obtain the boron ore type of the boron ore.
According to the embodiment of the invention, the problems of low detection efficiency and large error in the boron ore beneficiation process are solved through the streamlined image data acquisition, processing and automatic type identification, the labor and time cost are saved, and the detection efficiency and accuracy of the boron ore type are greatly improved.
In an alternative embodiment of the present invention, as shown in fig. 2, the image capturing module 110 may include: an image acquisition device 111, an image sensing device 112, and an image signal processor 113. Each component of the image capturing module 110 may be individually configured or may be collectively configured, that is, the image capturing module 110 may be individually configured with an image capturing device such as a camera, a video camera, and the like, an image sensing device, and an image signal processor. As shown in fig. 3, the device as a whole may include an external housing, and one end of the housing is provided with an image capturing device, such as a lens in fig. three; an image sensor, an image processor and an infrared lamp can be arranged in the shell, the infrared lamp is used for providing light source compensation when the light source of the lens is weak, so that the camera can clearly image, and a USB-to-RJ 45 converter can be arranged in the shell; the other end of the casing can be provided with an Ethernet electrical interface and a twisted-pair interface, and the image processor is connected with the external Ethernet interface through a USB-to-RJ 45 converter and the twisted-pair interface so as to transmit the image data to the PC end through the Ethernet.
The image acquisition device 111 may be used to acquire optical signals of the boron ore. The image sensing device 112 may be electrically connected to the image capturing device, and is configured to convert the optical signal corresponding to the boron ore into an electrical signal. The image signal processor 113 may be electrically connected to the image sensing device, and configured to process the electrical signal into image data, and upload and store the image data into an image database.
That is to say, the image acquisition device 111 can photograph the boron ore to be classified through the camera lens to obtain the optical signal of the boron ore, the image sensing device 112 receives the optical signal of the boron ore, converts the optical signal into an electrical signal, the electrical signal is processed into image data by the image signal processor 113, the image data is uploaded to an external PC terminal through the ethernet, and an image database is established for storage, so as to facilitate subsequent processing.
In an alternative embodiment of the present invention, as shown in fig. 2, the image processing module 120 may include: an image preprocessing unit 121, a color information extraction unit 122, and an image feature element extraction unit 123.
An image preprocessing unit 131 configured to perform image preprocessing on the image data one by one and determine whether the image data is valid image data; when the image data is effective image data, saving the effective image data; when the image data is invalid image data, ignoring the invalid image data; the preprocessing comprises at least one of image graying, image filtering processing and image binarization processing.
Judging whether the image data is valid image data or not, and obtaining a gray-scale image corresponding to the image data by carrying out gray-scale processing on the image data; carrying out binarization processing on the gray-scale image by utilizing an im2bw function, extracting bright spot features in the gray-scale image, and judging whether the area of the bright spots in the gray-scale image is larger than a preset pixel point or not; when the area of the bright spot in the gray image is less than or equal to the preset pixel point, the image data is effective image data; and when the area of the bright spot in the gray-scale image is larger than the preset pixel point, the image data is invalid image data.
For example, the preset pixel points are 1000 pixel points, when the area of the bright spot in the gray image is identified to be larger than 1000 pixel points, the gray image is considered to have a huge defect, so that the characteristic value of the image deviates from the actual situation, the gray image is taken as invalid image data, the system automatically ignores the invalid image data, and the next image data is analyzed. When the area of the bright spots in the gray image is recognized to be less than or equal to 1000 pixel points, the gray image is considered to accord with the actual condition of the characteristic value and is used as effective image data to enter the subsequent processing.
And the color information extracting unit 122 is configured to acquire an RGB color value corresponding to the effective image data, and store the RGB color value as color information of the effective image data.
After the effective image data is selected, color information of the effective image data needs to be acquired, and different boron ore types have different colors, for example, the boron-magnesium ore colors are white, gray and light yellow, the boron-magnesium iron ore colors include black and black green, the boron-magnesium ore colors are colorless and white, and the like. The color information of the boron ore is obtained by converting the effective image data color into the RGB value, so that the system can more accurately identify the type of the boron ore.
And an image feature element extraction unit 123, configured to extract and store an image feature element corresponding to the valid image data by using a feature extraction algorithm.
The feature extraction algorithm is a process for automatically extracting features of the convolutional layer by utilizing a CNN convolutional neural network technology in the field of deep learning. The principle is that after a convolution operation of a convolution kernel, one image matrix obtains another matrix, which is called feature mapping. Each convolution kernel may extract a particular feature, with different convolution kernels extracting different features. For example, an image of a human face is input, a certain convolution kernel is used to extract the features of the eyes, another convolution kernel is used to extract the features of the mouth, and so on. The feature mapping is a feature value matrix obtained by convolution operation of a certain image. CNN can be understood as comparing images in blocks, and this "small block" to which it is compared is called a feature (features). And (3) finding some rough features at approximately the same positions in the two images for matching, wherein the CNN can better see the similarity of the two images, and compared with the traditional method for comparing the whole images one by one, each feature is just like a small valued two-dimensional array, and different features match different features in the images. CNN does not know exactly which parts of the original are to be matched by these features, so it tries at every possible position in the original, i.e. it uses the convolution kernel to perform a sliding operation on the image, and performs a convolution operation once per sliding operation to obtain a feature value. Finally, after the whole image is convolved, a characteristic matrix with characteristic values in the range of-1 to-1 is obtained, wherein the closer the value in the matrix is to 1, the closer the corresponding position is to the characteristic represented by the convolution kernel, and the closer the value is to-1, the closer the corresponding position is to the inverse characteristic represented by the convolution kernel.
The image characteristic elements extracted by the convolutional neural network based on deep learning of the embodiment of the invention can comprise gray level mean, variance, smoothness, skewness, energy value, entropy value and the like.
Further, as shown in fig. 2, the intelligent boron rock beneficiation system provided by the embodiment of the present invention may further include: a model building module 140.
The model construction module 140 may construct a boron mine classification model based on a convolutional neural network technology, and train, test, and verify the boron mine classification model by using a preset sample feature element and a sample class label of a boron mine image data sample library to obtain a target boron mine classification model.
The sample characteristic elements may include color, shape, texture, gloss, and the like. The sample class label is a boron ore class name and may include, boromagnesite, ferroboron, maurite, natural boric acid, borax dehydrate, ulexite, colemanite, whitish boroalite, ulexite, manboracite, natural borax, tunnel stone, stigmatite, and the like.
Firstly, determining a learning procedure of boron ore classification training, namely a learning process of model training, and compiling the learning process of model training to obtain a preliminarily constructed boron ore classification model. The TensorFlow is an open source code software library which uses a data flow graph to carry out numerical calculation, is good at training a deep neural network, and can use Pycharm software to construct a CNN boron mine classification model by using the TensorFlow.
The method comprises the steps of firstly screening, processing and classifying various boron ore pictures according to different characteristics of the boron ore pictures, using sample characteristic elements such as color, shape and texture of the boron ore as input values of a model, and uniformly converting the picture data into the same size so as to train and test a boron ore classification model.
Specifically, the training set and the test set may respectively select a plurality of pieces of typical boron ore image data for each boron ore type as samples of the training set and the test set, where each sample in the training set samples may include image data and a corresponding type label, and each sample in the test set samples may include image data without a type label corresponding to the image data. The system can automatically read the image data and the type labels from the folders corresponding to the training set samples and the testing set samples into the array, and automatically process the quantity of the boron mine types and the quantity of the image data corresponding to each boron mine type. As shown in fig. 4, which is a schematic structural diagram of a boron mine classification model, a training set sample array is subjected to multiple iterations through 3 convolutional layers and pooling layers containing activation functions, 1 flat layer, 2 full-link layers containing activation functions, and 1 output layer until training is finished. And testing the trained boron ore classification model through the test set sample, and if the performance test of the trained boron ore classification model on the test set sample is good, using the boron ore classification model as a target boron ore classification model and storing the boron ore classification model for calling when identifying the boron ore type in the future.
In an alternative embodiment of the present invention, as shown in fig. 2, the intelligent classification module 130 may further be configured to: generating a feature element set according to the color information and the image feature elements corresponding to the effective image data, and performing data normalization processing on the feature elements in the feature element set to obtain target detection data; and inputting the target detection data into a pre-trained target boron ore classification model, and intelligently classifying the boron ore to be classified according to the target detection data by using the target boron ore classification model to obtain the boron ore type of the boron ore.
The system combines color information of image data corresponding to the boron ore and image characteristic elements to generate a characteristic element set, normalizes the characteristic elements in the characteristic element set, inputs the normalized characteristic elements into a target boron ore classification model, intelligently matches boron ore types corresponding to the boron ore in the image data according to the characteristic elements in the characteristic element set, and outputs corresponding results through a PC (personal computer) end.
Further, the image processing module 120 may be further configured to: periodically judging whether newly added image data exists in an image database; and when the newly added image data exists in the image database, performing image preprocessing on the newly added image data to select newly added effective image data, and extracting color information and image characteristic elements corresponding to the newly added effective image data.
Accordingly, the intelligent classification module 130 is further configured to: and periodically detecting the newly added effective image data and the corresponding color information and image characteristic elements thereof, and performing boron ore classification on the boron ore corresponding to the newly added effective image data by using a target boron ore classification model to obtain the boron ore type of the boron ore.
That is, the operator may set regular automatic image processing through the PC, for example, automatically detect whether the image database of the system contains new image data every 1 minute, perform image preprocessing on the new image data one by one, select new effective images in the new image, and automatically perform boron mine type identification on the new effective image data. By setting the regular detection, the image data can be processed more flexibly, the manual operation time is saved, and the detection efficiency of the boron ore type is greatly improved.
Further, the intelligent classification module 130 may also be configured to: storing the image data corresponding to the boron ore and the boron ore type.
Accordingly, the model building module 140 may also be configured to: periodically acquiring image data and a boron ore type corresponding to the newly added boron ore, and updating sample characteristic elements and sample category labels in a boron ore image data sample library according to the image data and the boron ore type corresponding to the newly added boron ore; and self-learning the target boron ore classification model by using the updated sample characteristic elements and the sample class labels, and updating the target boron ore classification model.
That is, after the boron ore type of the image data is identified by using the target boron ore classification model, the image data and the boron ore type corresponding to the boron ore are used as new sample data, the system updates the sample characteristic elements and the sample class labels in the boron ore image data sample library at intervals according to the newly input sample data, and retrains the target boron ore classification model again, so that the self-learning of the target boron ore classification model is realized, and the identification precision and the classification accuracy of the target boron ore classification model are improved.
The intelligent boron ore beneficiation system provided by the invention can complete automatic identification of boron ore types in the boron ore beneficiation process, realize high-efficiency, flow and modularized automatic detection, and carry out self-learning update on the target boron ore classification model by establishing the boron ore image data sample library, realize improvement of identification precision and classification accuracy of the target boron ore classification model, greatly save labor cost, reduce error rate of boron ore type identification, and improve flexibility of system work and convenience of use of operators by setting up periodic processing.
Based on the same inventive concept, the embodiment of the invention also provides an intelligent boron ore beneficiation method, as shown in fig. 5, the method at least comprises the following steps S501 to S503.
Step S501, image data of at least one boron ore is obtained and stored in an image database.
Through the image acquisition equipment, the lighting system, the image sensing equipment and the image signal processor which are arranged in the system, the boron ore to be classified is photographed and transmitted or stored to the computer equipment for subsequent processing.
Step S502, image preprocessing is carried out on the image data in the image database to select effective image data, and color information and image characteristic elements corresponding to the effective image data are extracted.
The image preprocessing can include graying, image filtering, image binarization and the like, effective image data with image characteristic values meeting actual conditions are selected, and color information and image characteristic elements corresponding to the effective image data are extracted according to a preset algorithm.
The color information refers to RGB values of the colors of the image data, and the image feature elements refer to crystal shapes, textures, gray level means, variances, smoothness, skewness, energy values, entropy values, and the like of the image data.
Step S503, inputting color information and image characteristic elements corresponding to any effective image data into a pre-trained target boron ore classification model, and performing boron ore classification on boron ores corresponding to the effective image data by using the target boron ore classification model to obtain boron ore types of the boron ores.
And (3) constructing a boron ore classification model by using a convolutional neural network technology to identify the boron ore type of the boron ore to be classified. Among the types of boron ore may include ludwigite, malbolite, natural boric acid, borax barren, ulexite, colemanite, whitlockite, ulexite, manboracite, natural borax, chenopodium stone, stigmatite, and the like.
Specifically, a feature element set can be generated according to the color information and the image feature elements corresponding to the effective image data, and data normalization processing is performed on the feature elements in the feature element set to obtain target detection data; and intelligently classifying the boron ore to be classified according to the target detection data by using the target boron ore classification model to obtain the boron ore type of the boron ore.
That is to say, the system combines the color information of the image data corresponding to the boron ore and the image characteristic elements to generate a characteristic element set, normalizes the characteristic elements in the characteristic element set, inputs the normalized characteristic elements into the target boron ore classification model, intelligently matches the boron ore type corresponding to the boron ore in the image data according to the characteristic elements in the characteristic element set, and outputs a corresponding result through the PC terminal.
According to the embodiment of the invention, the image data of the boron ore is collected, the image data is subjected to image preprocessing to obtain the color information and the image characteristic elements of the effective image data, and the boron ore is classified by using the target boron ore classification model to obtain the boron ore type of the boron ore, so that the automatic intelligent detection of the boron ore type is realized.
Further, in the intelligent boron ore beneficiation method provided by the embodiment of the present invention, the step S301 may further include the following substeps S501-1 to S502-3:
step S501-1: and collecting optical signals of the boron ore by using image collection equipment.
Step S501-2: and the image sensing equipment is electrically connected with the image acquisition equipment, so that optical signals corresponding to the boron ore are converted into electric signals.
Step S501-3: and the image data acquisition device is electrically connected with the image sensing equipment, performs signal processing based on the electric signals to obtain image data of the boron ore, and stores the image data in an image database.
The image processing device can be provided with an external casing, one end of the casing is provided with a camera lens and an infrared lamp, a USB-to-RJ 45 converter can be arranged inside the casing, and the other end of the casing can be provided with an Ethernet electrical interface and a twisted pair interface. Image acquisition equipment can shoot through camera lens to the boron ore of treating categorised, obtains boron ore stone light signal, and image sensing equipment receives boron ore stone light signal, turns into the signal of telecommunication with light signal conversion, and the signal of telecommunication is image data by image signal processor handles, passes to outside PC end through the ethernet to establish the image database and keep, so that follow-up processing.
Further, in the intelligent boron ore beneficiation method provided by the embodiment of the present invention, the step S302 may further include the following substeps S502-1 to S502-3:
step S502-1: carrying out image preprocessing on the image data one by one, and judging whether the image data is effective image data; when the image data is effective image data, saving the effective image data; when the image data is invalid image data, ignoring the invalid image data; the preprocessing comprises at least one of image graying, image filtering processing and image binarization processing.
Judging whether the image data is valid image data or not, and obtaining a gray-scale image corresponding to the image data by carrying out gray-scale processing on the image data; carrying out binarization processing on the gray-scale image by utilizing an im2bw function, extracting bright spot features in the gray-scale image, and judging whether the area of the bright spots in the gray-scale image is larger than a preset pixel point or not; when the area of the bright spot in the gray image is less than or equal to the preset pixel point, the image data is effective image data; and when the area of the bright spot in the gray-scale image is larger than the preset pixel point, the image data is invalid image data.
For example, the preset pixel points are 1000 pixel points, when the area of the bright spot in the gray image is identified to be larger than 1000 pixel points, the gray image is considered to have a huge defect, so that the characteristic value of the image deviates from the actual situation, the gray image is taken as invalid image data, the system automatically ignores the invalid image data, and the next image data is analyzed. When the area of the bright spots in the gray image is recognized to be less than or equal to 1000 pixel points, the gray image is considered to accord with the actual condition of the characteristic value and is used as effective image data to enter the subsequent processing.
Step S502-2: and acquiring the RGB color value corresponding to the effective image data, and storing the RGB color value as the color information of the effective image data.
After the effective image data is selected, color information of the effective image data needs to be acquired, and different boron ore types have different colors, for example, the boron-magnesium ore colors are white, gray and light yellow, the boron-magnesium iron ore colors include black and black green, the boron-magnesium ore colors are colorless and white, and the like. The color information of the boron ore is obtained by converting the effective image data color into the RGB value, so that the system can more accurately identify the type of the boron ore.
Step S502-3: and extracting and storing image characteristic elements corresponding to the effective image data by using a characteristic extraction algorithm.
The image feature element may include a gray level mean, a variance, a smoothness, a skewness, an energy value, an entropy value, and the like.
Optionally, the intelligent boron rock beneficiation method provided by the embodiment of the present invention may further include step S504:
step S504: and constructing a boron ore classification model based on a convolutional neural network technology, and training, testing and verifying the boron ore classification model by using preset sample characteristic elements and sample class labels of a boron ore image data sample library to obtain a target boron ore classification model.
Firstly, determining a learning procedure of boron ore classification training, namely a learning process of model training, and compiling the learning process of model training to obtain a preliminarily constructed boron ore classification model. The TensorFlow is an open source code software library which uses a data flow graph to carry out numerical calculation, is good at training a deep neural network, and can use Pycharm software to construct a CNN boron mine classification model by using the TensorFlow.
The method comprises the steps of firstly screening, processing and classifying various boron ore pictures according to different characteristics of the boron ore pictures, using sample characteristic elements such as color, shape and texture of the boron ore as input values of a model, and uniformly converting the picture data into the same size so as to train and test a boron ore classification model.
Specifically, the training set and the test set may respectively select a plurality of pieces of typical boron ore image data for each boron ore type as samples of the training set and the test set, where each sample in the training set samples may include image data and a corresponding type label, and each sample in the test set samples may include image data without a type label corresponding to the image data. The system can automatically read image data and type labels from folders corresponding to the training set samples and the test set samples into an array, automatically process the quantity of the boron mine types and the quantity of the image data corresponding to each boron mine type, and perform multiple iterations and finish training through 3 convolutional layers and pooling layers containing activation functions, 1 flat layer, 2 full-connection layers containing activation functions and 1 output layer. And testing the trained boron ore classification model through the test set sample, and if the performance test of the trained boron ore classification model on the test set sample is good, using the boron ore classification model as a target boron ore classification model and storing the boron ore classification model for calling when identifying the boron ore type in the future.
Optionally, the intelligent boron ore beneficiation method provided by the embodiment of the invention can also periodically judge whether new image data exists in the image database; when newly added image data exist in the image database, performing image preprocessing on the newly added image data to select newly added effective image data, and extracting color information and image characteristic elements corresponding to the newly added effective image data; and periodically detecting the newly added effective image data and the corresponding color information and image characteristic elements thereof, and performing boron ore classification on the boron ore corresponding to the newly added effective image data by using a target boron ore classification model to obtain the boron ore type of the boron ore.
That is, an operator can control the start or the end of the operation of the boron mine identification program through the PC terminal, and can also set regular automatic image processing, set the time interval of the automatic image processing, for example, automatically detect whether the image database of the system contains new image data every 1 minute, set the buffer area of the image processing, and continue to wait if no new image data exists in the buffer area; if the new image data exists in the buffer area, the new image data is read one by one and subjected to image preprocessing, new effective images in the new images are selected, corresponding color information and image characteristic elements are automatically extracted from the new effective image data, data normalization is carried out on the color information and the image characteristic elements, a target boron ore classification model is input, boron ore type identification is carried out, and the identification result is output to a PC end page for display. And after the boron ore type identification is finished, emptying the image data of the image processing buffer area, and continuously and circularly detecting whether the system has newly added image data or not so as to facilitate subsequent processing.
Optionally, the intelligent boron ore beneficiation method provided by the embodiment of the present invention may further include the following steps:
storing image data and boron ore types corresponding to the boron ores; periodically acquiring image data and a boron ore type corresponding to the newly added boron ore, and updating sample characteristic elements and sample category labels in a boron ore image data sample library according to the image data and the boron ore type corresponding to the newly added boron ore; and self-learning the target boron ore classification model by using the updated sample characteristic elements and the sample class labels, and updating the target boron ore classification model.
That is, after the boron ore type of the image data is identified by using the target boron ore classification model, the image data corresponding to the boron ore and the boron ore type are used as new sample data, the system updates the sample characteristic elements and the sample class labels in the boron ore image data sample library at intervals according to the newly input sample data, and retrains the target boron ore classification model again, so that the self-learning of the target boron ore classification model is realized, and the identification precision and the classification accuracy of the target boron ore classification model are improved.
The intelligent boron ore beneficiation system and method provided by the embodiment of the invention extract color information and image characteristic elements corresponding to boron ore image data by preprocessing the image data of the boron ore to be classified, training and testing a pre-constructed boron ore classification model, selecting the boron ore classification model with higher accuracy and smaller error as a target boron ore classification model, completing automatic identification of boron ore types in the boron ore beneficiation process, realizing high-efficiency, flow-based and modular automatic detection, and self-learning update is carried out on the target boron ore classification model by establishing a boron ore image data sample library, so that the identification precision and the classification accuracy of the target boron ore classification model are improved, the labor cost is greatly saved, the error rate of boron ore type identification is reduced, and by setting the periodic processing, the flexibility of system work and the convenience of use of operators are improved.
It can be clearly understood by those skilled in the art that the specific working processes of the system, the module and the unit described above may refer to the corresponding processes in the foregoing method embodiments, and for the sake of brevity, the detailed description is omitted here.
In addition, the functional units in the embodiments of the present invention may be physically independent of each other, two or more functional units may be integrated together, or all the functional units may be integrated in one processing unit. The integrated functional units may be implemented in the form of hardware, or in the form of software or firmware.
Those of ordinary skill in the art will understand that: the integrated functional units, if implemented in software and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computing device (e.g., a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention when the instructions are executed. And the aforementioned storage medium includes: u disk, removable hard disk, Read Only Memory (ROM), Random Access Memory (RAM), magnetic or optical disk, and other various media capable of storing program code.
Alternatively, all or part of the steps of implementing the foregoing method embodiments may be implemented by hardware (such as a computing device, e.g., a personal computer, a server, or a network device) associated with program instructions, which may be stored in a computer-readable storage medium, and when the program instructions are executed by a processor of the computing device, the computing device executes all or part of the steps of the method according to the embodiments of the present invention.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments can be modified or some or all of the technical features can be equivalently replaced within the spirit and principle of the present invention; such modifications or substitutions do not depart from the scope of the present invention.
Claims (10)
1. An intelligent boron ore beneficiation system, comprising:
the image acquisition module is used for acquiring image data of at least one boron ore and storing the image data into an image database;
the image processing module is used for carrying out image preprocessing on the image data in the image database to select effective image data and extracting color information and image characteristic elements corresponding to the effective image data;
and the intelligent classification module is used for inputting color information and image characteristic elements corresponding to any effective image data into a pre-trained target boron ore classification model, and performing boron ore classification on the boron ore corresponding to the effective image data by using the target boron ore classification model to obtain the boron ore type of the boron ore.
2. The system of claim 1, wherein the image processing module comprises an image pre-processing unit; the image preprocessing unit is used for:
carrying out image preprocessing on the image data one by one, and judging whether the image data is effective image data;
when the image data is effective image data, saving the effective image data;
when the image data is invalid image data, ignoring the invalid image data;
the preprocessing comprises at least one of image graying, image filtering processing and image binarization processing.
3. The system of claim 2, wherein the image pre-processing unit is further configured to:
carrying out graying processing on the image data to obtain a grayscale image corresponding to the image data;
carrying out binarization processing on the gray-scale image by utilizing an im2bw function, extracting bright spot features in the gray-scale image, and judging whether the area of the bright spots in the gray-scale image is larger than a preset pixel point or not;
when the area of the bright spot in the gray image is less than or equal to a preset pixel point, the image data is effective image data; and when the area of the bright spot in the gray-scale image is larger than the preset pixel point, the image data is invalid image data.
4. The system of claim 1, wherein the image processing module comprises:
the color information extraction unit is used for acquiring RGB color values corresponding to the effective image data, and storing the RGB color values as color information of the effective image data;
the image characteristic element extraction unit is used for extracting and storing image characteristic elements corresponding to the effective image data by utilizing a characteristic extraction algorithm; wherein the image characteristic elements comprise at least one of a gray mean, a variance, smoothness, skewness, an energy value and an entropy value.
5. The system of claim 1, wherein the intelligent classification module is further configured to:
generating a feature element set according to the color information and the image feature elements corresponding to the effective image data, and performing data normalization processing on the feature elements in the feature element set to obtain target detection data;
and inputting the target detection data into a pre-trained target boron ore classification model, and intelligently classifying the boron ore to be classified by using the target boron ore classification model according to the target detection data to obtain the boron ore type of the boron ore.
6. The system of claim 1,
the image processing module is further configured to: periodically judging whether newly added image data exists in the image database; when newly added image data exist in the image database, performing image preprocessing on the newly added image data to select newly added effective image data, and extracting color information and image characteristic elements corresponding to the newly added effective image data;
the intelligent classification module is further configured to: and periodically detecting newly added effective image data and corresponding color information and image characteristic elements thereof, and performing boron ore classification on the boron ore corresponding to the newly added effective image data by using the target boron ore classification model to obtain the boron ore type of the boron ore.
7. The system of claims 1-6, further comprising a model building module;
the model building module is configured to:
building a boron ore classification model based on a convolutional neural network technology, and training, testing and verifying the boron ore classification model by using preset sample characteristic elements and sample class labels of a boron ore image data sample library to obtain a target boron ore classification model;
wherein the sample characteristic element comprises at least one of color, shape, texture and luster; the sample class label is a boron ore class name, including but not limited to, boromagnesite, ludwigite, natural boric acid, borax dehydrate, ulexite, colemanite, whitish boroalite, ulexite, manboracite, natural borax, tunnel stone, and pindolite.
8. The system of claim 7,
the intelligent classification module is further configured to: storing image data and boron ore types corresponding to the boron ores;
the model building module is further configured to: periodically acquiring image data and a boron ore type corresponding to the newly added boron ore, and updating sample characteristic elements and sample category labels in a boron ore image data sample library according to the image data and the boron ore type corresponding to the newly added boron ore; and self-learning the target boron mine classification model by using the updated sample characteristic elements and the updated sample class labels, and updating the target boron mine classification model.
9. The system of claim 1, wherein the image acquisition module comprises:
the image acquisition equipment is used for acquiring optical signals of the boron ore;
the image sensing equipment is electrically connected with the image acquisition equipment and is used for converting optical signals corresponding to the boron ore into electric signals;
and the image signal processor is electrically connected with the image sensing equipment, is used for carrying out signal processing based on the electric signal to obtain image data of the boron ore, and stores the image data into the image database.
10. An intelligent boron ore beneficiation method is characterized by comprising the following steps:
acquiring image data of at least one boron ore and storing the image data into an image database;
image preprocessing is carried out on image data in the image database to select effective image data, and color information and image characteristic elements corresponding to the effective image data are extracted;
and inputting color information and image characteristic elements corresponding to any effective image data into a pre-trained target boron ore classification model, and performing boron ore classification on the boron ore corresponding to the effective image data by using the target boron ore classification model to obtain the boron ore type of the boron ore.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110744017.8A CN113591583A (en) | 2021-06-30 | 2021-06-30 | Intelligent boron ore beneficiation system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110744017.8A CN113591583A (en) | 2021-06-30 | 2021-06-30 | Intelligent boron ore beneficiation system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113591583A true CN113591583A (en) | 2021-11-02 |
Family
ID=78245689
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110744017.8A Pending CN113591583A (en) | 2021-06-30 | 2021-06-30 | Intelligent boron ore beneficiation system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113591583A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114565820A (en) * | 2022-03-01 | 2022-05-31 | 中科海慧（北京）科技有限公司 | Mineral sample identification system based on space-time big data analysis |
| CN117705198A (en) * | 2024-02-05 | 2024-03-15 | 天津矿山工程有限公司 | Double-side feeding ore bin unloading monitoring system and method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102297822A (en) * | 2010-06-22 | 2011-12-28 | 中国矿业大学 | Method for predicting ash content of coal particles by utilizing image analysis |
| CN105068136A (en) * | 2015-07-27 | 2015-11-18 | 中国地质调查局武汉地质调查中心 | Potential positioning evaluation method for copper and gold mine of Indo-China peninsula demonstration zone based on multi-source information |
| CN107766856A (en) * | 2016-08-20 | 2018-03-06 | 湖南军芃科技股份有限公司 | A kind of ore visible images method for separating based on Adaboost machine learning |
| CN110232419A (en) * | 2019-06-20 | 2019-09-13 | 东北大学 | A kind of method of side slope rock category automatic identification |
| CN111881909A (en) * | 2020-07-27 | 2020-11-03 | 精英数智科技股份有限公司 | Coal and gangue identification method and device, electronic equipment and storage medium |
| CN112580659A (en) * | 2020-11-10 | 2021-03-30 | 湘潭大学 | Ore identification method based on machine vision |
-
2021
- 2021-06-30 CN CN202110744017.8A patent/CN113591583A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102297822A (en) * | 2010-06-22 | 2011-12-28 | 中国矿业大学 | Method for predicting ash content of coal particles by utilizing image analysis |
| CN105068136A (en) * | 2015-07-27 | 2015-11-18 | 中国地质调查局武汉地质调查中心 | Potential positioning evaluation method for copper and gold mine of Indo-China peninsula demonstration zone based on multi-source information |
| CN107766856A (en) * | 2016-08-20 | 2018-03-06 | 湖南军芃科技股份有限公司 | A kind of ore visible images method for separating based on Adaboost machine learning |
| CN110232419A (en) * | 2019-06-20 | 2019-09-13 | 东北大学 | A kind of method of side slope rock category automatic identification |
| CN111881909A (en) * | 2020-07-27 | 2020-11-03 | 精英数智科技股份有限公司 | Coal and gangue identification method and device, electronic equipment and storage medium |
| CN112580659A (en) * | 2020-11-10 | 2021-03-30 | 湘潭大学 | Ore identification method based on machine vision |
Non-Patent Citations (1)
| Title |
|---|
| 胡发焕;刘国平;胡华;董增文;: "基于模糊支持向量机和D-S证据理论的钨矿石初选方法", 光子学报, no. 07, pages 240 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114565820A (en) * | 2022-03-01 | 2022-05-31 | 中科海慧（北京）科技有限公司 | Mineral sample identification system based on space-time big data analysis |
| CN117705198A (en) * | 2024-02-05 | 2024-03-15 | 天津矿山工程有限公司 | Double-side feeding ore bin unloading monitoring system and method |
| CN117705198B (en) * | 2024-02-05 | 2024-05-24 | 天津矿山工程有限公司 | Double-side feeding ore bin unloading monitoring system and method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108960245B (en) | Tire mold character detection and recognition method, device, equipment and storage medium | |
| CN105678332B (en) | Converter steelmaking end point judgment method and system based on flame image CNN recognition modeling | |
| Amara et al. | A deep learning-based approach for banana leaf diseases classification | |
| WO2019233297A1 (en) | Data set construction method, mobile terminal and readable storage medium | |
| Zheng et al. | Automatic inspection of metallic surface defects using genetic algorithms | |
| CN111257341B (en) | Underwater building crack detection method based on multi-scale features and stacked full convolution network | |
| CN106886216B (en) | Robot automatic tracking method and system based on RGBD face detection | |
| Izadi et al. | A new intelligent method for minerals segmentation in thin sections based on a novel incremental color clustering | |
| CN113591583A (en) | Intelligent boron ore beneficiation system and method | |
| CN111401246B (en) | Smoke concentration detection method, device, equipment and storage medium | |
| CN109902662B (en) | Pedestrian re-identification method, system, device and storage medium | |
| Laga et al. | Image-based plant stornata phenotyping | |
| CN113869300A (en) | Workpiece surface defect and character recognition method and system based on multi-vision fusion | |
| CN109934221B (en) | Attention mechanism-based automatic analysis, identification and monitoring method and system for power equipment | |
| CN110728201B (en) | Image processing method and device for fingerprint identification | |
| CN110334727B (en) | Intelligent matching detection method for tunnel cracks | |
| CN112862757A (en) | Weight evaluation system based on computer vision technology and implementation method | |
| CN113963041A (en) | Image texture recognition method and system | |
| CN108460344A (en) | Dynamic area intelligent identifying system in screen and intelligent identification Method | |
| Jenifa et al. | Classification of cotton leaf disease using multi-support vector machine | |
| CN110688966B (en) | Semantic guidance pedestrian re-recognition method | |
| CN115007474A (en) | Coal dressing robot and coal dressing method based on image recognition | |
| CN110569716A (en) | Goods shelf image copying detection method | |
| CN113297408A (en) | Image matching and scene recognition system and method based on Sift algorithm | |
| CN112270404A (en) | Detection structure and method for bulge defect of fastener product based on ResNet64 network |
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 |