CN118204275B - Phosphogypsum impurity removal method and phosphogypsum impurity removal system based on visual detection technology - Google Patents
Phosphogypsum impurity removal method and phosphogypsum impurity removal system based on visual detection technology Download PDFInfo
- Publication number
- CN118204275B CN118204275B CN202410627938.XA CN202410627938A CN118204275B CN 118204275 B CN118204275 B CN 118204275B CN 202410627938 A CN202410627938 A CN 202410627938A CN 118204275 B CN118204275 B CN 118204275B
- Authority
- CN
- China
- Prior art keywords
- phosphogypsum
- impurity
- grabbing
- target
- module
- 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.)
- Active
Links
- 239000012535 impurity Substances 0.000 title claims abstract description 271
- PASHVRUKOFIRIK-UHFFFAOYSA-L calcium sulfate dihydrate Chemical compound O.O.[Ca+2].[O-]S([O-])(=O)=O PASHVRUKOFIRIK-UHFFFAOYSA-L 0.000 title claims abstract description 198
- 238000001514 detection method Methods 0.000 title claims abstract description 49
- 238000000034 method Methods 0.000 title claims abstract description 43
- 230000000007 visual effect Effects 0.000 title claims abstract description 37
- 238000005516 engineering process Methods 0.000 title claims abstract description 23
- 238000012216 screening Methods 0.000 claims abstract description 28
- 239000000463 material Substances 0.000 claims abstract description 19
- 238000012545 processing Methods 0.000 claims abstract description 15
- 238000012549 training Methods 0.000 claims description 48
- 230000005855 radiation Effects 0.000 claims description 34
- 238000004891 communication Methods 0.000 claims description 33
- 230000008569 process Effects 0.000 claims description 23
- 210000005036 nerve Anatomy 0.000 claims description 21
- 238000005070 sampling Methods 0.000 claims description 21
- 230000005540 biological transmission Effects 0.000 claims description 19
- 230000006870 function Effects 0.000 claims description 13
- 238000003062 neural network model Methods 0.000 claims description 13
- 238000003860 storage Methods 0.000 claims description 9
- 239000006185 dispersion Substances 0.000 claims description 8
- 238000011179 visual inspection Methods 0.000 claims description 7
- 238000007781 pre-processing Methods 0.000 claims description 6
- 230000001537 neural effect Effects 0.000 claims description 5
- 238000004458 analytical method Methods 0.000 claims description 3
- 238000009877 rendering Methods 0.000 claims description 3
- 239000002245 particle Substances 0.000 description 12
- OAICVXFJPJFONN-UHFFFAOYSA-N Phosphorus Chemical compound [P] OAICVXFJPJFONN-UHFFFAOYSA-N 0.000 description 11
- 239000011574 phosphorus Substances 0.000 description 11
- 229910052698 phosphorus Inorganic materials 0.000 description 11
- 230000005496 eutectics Effects 0.000 description 9
- 230000003287 optical effect Effects 0.000 description 7
- NBIIXXVUZAFLBC-UHFFFAOYSA-N Phosphoric acid Chemical compound OP(O)(O)=O NBIIXXVUZAFLBC-UHFFFAOYSA-N 0.000 description 6
- 239000000203 mixture Substances 0.000 description 6
- 238000013528 artificial neural network Methods 0.000 description 5
- 238000012360 testing method Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 4
- 229910000147 aluminium phosphate Inorganic materials 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 229910052602 gypsum Inorganic materials 0.000 description 3
- 239000010440 gypsum Substances 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 239000002699 waste material Substances 0.000 description 3
- XLLIQLLCWZCATF-UHFFFAOYSA-N 2-methoxyethyl acetate Chemical compound COCCOC(C)=O XLLIQLLCWZCATF-UHFFFAOYSA-N 0.000 description 2
- LGDSHSYDSCRFAB-UHFFFAOYSA-N Methyl isothiocyanate Chemical compound CN=C=S LGDSHSYDSCRFAB-UHFFFAOYSA-N 0.000 description 2
- QAOWNCQODCNURD-UHFFFAOYSA-N Sulfuric acid Chemical compound OS(O)(=O)=O QAOWNCQODCNURD-UHFFFAOYSA-N 0.000 description 2
- 239000004568 cement Substances 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000007599 discharging Methods 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 239000002440 industrial waste Substances 0.000 description 2
- 229910052500 inorganic mineral Inorganic materials 0.000 description 2
- 239000011707 mineral Substances 0.000 description 2
- 210000002569 neuron Anatomy 0.000 description 2
- 239000000243 solution Substances 0.000 description 2
- 238000005406 washing Methods 0.000 description 2
- WNXJIVFYUVYPPR-UHFFFAOYSA-N 1,3-dioxolane Chemical compound C1COCO1 WNXJIVFYUVYPPR-UHFFFAOYSA-N 0.000 description 1
- DBUJFULDVAZULB-UHFFFAOYSA-N 1-methoxypentane Chemical compound CCCCCOC DBUJFULDVAZULB-UHFFFAOYSA-N 0.000 description 1
- PXGOKWXKJXAPGV-UHFFFAOYSA-N Fluorine Chemical compound FF PXGOKWXKJXAPGV-UHFFFAOYSA-N 0.000 description 1
- 239000000654 additive Substances 0.000 description 1
- 239000007864 aqueous solution Substances 0.000 description 1
- 230000004888 barrier function Effects 0.000 description 1
- 239000003795 chemical substances by application Substances 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000003912 environmental pollution Methods 0.000 description 1
- 125000001495 ethyl group Chemical group [H]C([H])([H])C([H])([H])* 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 239000008394 flocculating agent Substances 0.000 description 1
- 239000011737 fluorine Substances 0.000 description 1
- 229910052731 fluorine Inorganic materials 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 230000036571 hydration Effects 0.000 description 1
- 238000006703 hydration reaction Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 239000002367 phosphate rock Substances 0.000 description 1
- OJMIONKXNSYLSR-UHFFFAOYSA-N phosphorous acid Chemical compound OP(O)O OJMIONKXNSYLSR-UHFFFAOYSA-N 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 230000036544 posture Effects 0.000 description 1
- 239000000843 powder Substances 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- -1 road filling Substances 0.000 description 1
- 238000004904 shortening Methods 0.000 description 1
- 239000002910 solid waste Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- 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
-
- 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
- B07C5/361—Processing or control devices therefor, e.g. escort memory
- B07C5/362—Separating or distributor mechanisms
-
- 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
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- 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
- G06N3/082—Learning methods modifying the architecture, e.g. adding, deleting or silencing nodes or connections
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/22—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- 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
- B07C2501/00—Sorting according to a characteristic or feature of the articles or material to be sorted
- B07C2501/0054—Sorting of waste or refuse
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Multimedia (AREA)
- Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Quality & Reliability (AREA)
- Image Processing (AREA)
- Analysing Materials By The Use Of Radiation (AREA)
Abstract
The invention discloses a phosphogypsum impurity removal method and a phosphogypsum impurity removal system based on a visual detection technology, wherein the phosphogypsum impurity removal system comprises: phosphogypsum treatment structure, impurity grabbing parameter generation structure and impurity removal structure; the phosphogypsum treatment structure is used for screening and dispersing the input target phosphogypsum material; the impurity grabbing parameter generating structure is used for acquiring the dispersed image information of the phosphogypsum to be processed in real time, analyzing and processing the image information to obtain target impurity position information, generating grabbing parameters according to the target impurity position information, and transmitting the grabbing parameters to the impurity removing structure; and the impurity removing structure is used for grabbing target impurities from the phosphogypsum to be treated according to the grabbing parameters to obtain phosphogypsum from which the target impurities are removed. According to the invention, the phosphogypsum image is acquired by using the impurity grabbing parameter generating structure and is processed, grabbing parameters are generated, and the sorting robot is controlled to quickly and effectively grab harmful impurities in phosphogypsum according to the grabbing parameters.
Description
Technical Field
The invention relates to the field of industrial solid waste material treatment, in particular to a phosphogypsum impurity removal method and system based on a visual detection technology.
Background
Phosphogypsum is industrial waste residue generated in the process of decomposing phosphorite and extracting phosphoric acid by using sulfuric acid, wherein the content of calcium sulfate dihydrate exceeds that of common natural gypsum. In recent years, the amount of industrial waste residues such as phosphogypsum is increasing, and phosphogypsum is gradually utilized under the promotion of ecological environment protection concept, and the application range includes but is not limited to cement additives, road filling, gypsum board manufacturing, building gypsum powder manufacturing, building blocks and the like. However, phosphogypsum is mostly used for constructing road pavement base layers due to the characteristics of high treatment difficulty, environmental pollution, lack of high-efficiency high-added-value utilization ways and the like, and the current situation that mineral resources are increasingly scarce. The phosphogypsum material is applied to the pavement base layer, and then the curing agent, cement and other materials are added, so that mineral resources can be saved, the comprehensive performance of the phosphogypsum mixture can be improved to a certain extent, and the functional requirements and performance requirements of the pavement base layer can be better met.
In the practical application process, because phosphogypsum contains a plurality of harmful impurities such as phosphorus, fluorine, organic matters and the like, the harmful impurities are easy to absorb water and form a lump, so that the hydration process is blocked, the setting time is prolonged, the pH value of the aqueous solution is reduced, the strength of the mixture is reduced, and the application effect of the phosphogypsum is greatly influenced. At present, a water washing method is generally adopted to remove organic impurities of phosphogypsum firstly, but the water washing method has low removal efficiency and is difficult to thoroughly remove harmful impurities, so that the performance of phosphogypsum mixed materials is not effectively improved.
In view of this, how to provide a phosphogypsum impurity removal method and a phosphogypsum impurity removal system which can rapidly and effectively remove harmful impurities in phosphogypsum and improve the performance of phosphogypsum mixture based on a visual detection technology are technical problems which need to be solved by those skilled in the art.
Disclosure of Invention
In view of the above, the invention provides a phosphogypsum impurity removal method and a phosphogypsum impurity removal system based on a visual detection technology, which solve the problems of low removal efficiency and poor removal effect of harmful impurities in phosphogypsum at present.
In order to achieve the above purpose, the present invention adopts the following technical scheme:
scheme one: a phosphogypsum impurity removal method based on visual detection technology comprises the following steps:
constructing a grabbing parameter generation model based on a deep neural network model frame and a neural radiation field, and training the grabbing parameter generation model;
Acquiring image information of phosphogypsum to be processed in real time;
inputting the image information into a trained grabbing parameter generation model for analysis processing to obtain target impurity position information, and generating grabbing parameters according to the target impurity position information; the grabbing parameters comprise preset grabbing positions and grabbing gesture information of target impurities;
and grabbing target impurities from the phosphogypsum to be treated according to the grabbing parameters to obtain phosphogypsum with the target impurities removed.
Further, training the grabbing parameter generation model, wherein the specific process is as follows:
preprocessing the image information of phosphogypsum to be processed to obtain an impurity grabbing sample data set;
Inputting the impurity grabbing sample data set into a deep neural network model frame, outputting a target impurity image marked with the position of a target impurity, and continuously carrying out iterative updating on the deep neural network model frame according to the target impurity image and related loss until a first preset model training ending condition is met;
inputting the target impurity image into a nerve radiation field to obtain an initial nerve radiation field;
Constructing an initial variant sub-model for learning geometric features of the target impurity image on the nerve radiation field, and determining a plurality of visual angles for training the initial nerve radiation field;
And training the initial nerve radiation field according to the target impurity image and the projection area corresponding to each view angle until a second preset model training ending condition is met.
Scheme II: phosphogypsum impurity removal system based on visual detection technology, comprising: phosphogypsum treatment structure, impurity grabbing parameter generation structure and impurity removal structure;
the impurity grabbing parameter generating structure and the impurity removing structure are arranged near the phosphogypsum processing structure;
the impurity grabbing parameter generating structure is connected with the impurity removing structure;
The phosphogypsum treatment structure is used for screening an input target phosphogypsum material to obtain phosphogypsum to be treated containing target impurities, and carrying out dispersion treatment on the phosphogypsum to be treated;
The impurity grabbing parameter generating structure is used for acquiring the dispersed image information of phosphogypsum to be processed in real time, analyzing and processing the image information to obtain target impurity position information, generating grabbing parameters according to the target impurity position information, and transmitting the grabbing parameters to the impurity removing structure; the grabbing parameters comprise preset grabbing positions and grabbing gesture information of target impurities;
And the impurity removing structure is used for grabbing target impurities from the phosphogypsum to be treated according to the grabbing parameters to obtain phosphogypsum from which the target impurities are removed.
According to the invention, the image information of phosphogypsum to be processed is obtained in real time through the impurity grabbing parameter generating structure, and is analyzed and processed to obtain the target impurity position information, meanwhile, grabbing parameters are generated according to the target impurity position information, and the sorting robot is controlled to accurately grab target impurities from the phosphogypsum to be processed according to the grabbing parameters. The grabbing parameter generation process is not limited by illumination change and image acquisition information, target impurities in phosphogypsum can be detected, identified and positioned more accurately, harmful impurities in phosphogypsum can be removed rapidly and accurately, and the performance of phosphogypsum mixture is improved.
Further, the phosphogypsum treatment structure comprises a feeding port, a transmission module, a screening module, a flattening transmission module and a discharging port;
the feeding port, the transmission module, the screening module and the flattening transmission module are sequentially connected;
the screening module is connected with the discharge port;
the feeding port is used for inputting target phosphogypsum materials;
the transmission module is used for conveying the target phosphogypsum material to the screening module;
The screening module is used for screening the target phosphogypsum material according to a preset rule to obtain phosphogypsum to be treated and phosphogypsum which is not required to be treated, conveying the phosphogypsum to the flattening transmission module, and conveying the phosphogypsum which is not required to be treated to the discharge port;
and the flattening transmission module is used for carrying out dispersion treatment on the phosphogypsum to be treated.
Further, the impurity grabbing parameter generating structure comprises a first control module, an image collector, an image processor, a visual detection module and a first communication module;
The first control module is respectively connected with the image collector, the image processor, the visual detection module and the first communication module;
The image collector is connected with the image processor;
the image processor is connected with the visual detection module;
The first communication module is respectively connected with the visual detection module and the impurity removal structure;
The image acquisition device is used for acquiring the dispersed image information of phosphogypsum to be processed in real time and transmitting the image information to the image processor;
The image processor is used for preprocessing the received image information to obtain preprocessed image information and transmitting the preprocessed image information to the visual detection module;
The vision detection module analyzes and processes the preprocessed image information by using a trained grabbing parameter generation model to obtain target impurity position information, and generates grabbing parameters according to the target impurity position information; the grabbing parameters comprise preset grabbing positions and grabbing gesture information of target impurities;
The first communication module is used for transmitting the grabbing parameters to the impurity removing structure;
The first control module is used for controlling the normal operation of each sub-module in the impurity grabbing parameter generating structure.
Further, the impurity removing structure includes a plurality of sorting robots;
The sorting robot comprises a sorting module, a second control module and a second communication module;
The second control module is respectively connected with the sorting module and the second control module;
the second communication module is connected with the first communication module;
The second communication module is used for transmitting the received grabbing parameters to the second control module;
And the second control module is used for controlling the sorting module to grasp target impurities from the phosphogypsum to be treated according to the grasping parameters.
Further, the impurity removing structure further comprises a force sensor;
the force sensor is respectively connected with the sorting module and the second control module;
the force sensor is used for collecting force information of the sorting module and transmitting the force information to the second control module; the second control module judges and processes the force information to obtain a grabbing result, and the grabbing result is transmitted to the first communication module through the second communication module.
Further, the impurity grabbing parameter generating structure further comprises a result display module;
The result display module is respectively connected with the first control module, the visual detection module and the first communication module;
the result display module is used for displaying the grabbing result; and the vision detection module adjusts the grabbing parameters according to the grabbing results and the related loss function.
Further, the device also comprises a crushing structure;
The crushing structure is arranged between the flattening transmission module and the discharge port, and is used for crushing phosphogypsum for removing target impurities and conveying the crushed phosphogypsum to the discharge port.
Scheme III: a computer readable storage medium having stored therein at least one program code loaded and executed by a computer processor to implement the phosphogypsum impurity removal method based on visual inspection techniques of scheme one.
Compared with the prior art, the invention has the following advantages:
According to the invention, the input target phosphogypsum material is screened and dispersed through the phosphogypsum treatment structure to obtain partial phosphogypsum containing harmful impurities to be removed, the impurity grabbing parameter generation structure is utilized to carry out high-precision detection on the harmful impurities in the phosphogypsum, and parameter data for grabbing the detected harmful impurities is automatically generated, so that the impurity removal structure thoroughly removes the harmful impurities in the phosphogypsum, the grabbing parameter generation model is not limited by illumination change and image acquisition information, the harmful impurities in the phosphogypsum can be detected, identified, positioned and grabbed more accurately, the harmful impurities in the phosphogypsum can be thoroughly sorted, the phosphogypsum mixture performance is improved, and the practical requirements of users on the phosphogypsum can be fully met. In addition, the whole phosphogypsum impurity removal process can realize full-automatic treatment, the data processing efficiency of the grabbing parameter generation model is higher, the whole process does not need to take too long, and the high-efficiency and accurate sorting of the harmful impurities in the phosphogypsum can be realized.
Drawings
In order to more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the drawings that are needed in the examples will be briefly described below, it being understood that the following drawings only illustrate some examples of the present invention and therefore should not be considered as limiting the scope, and that other related drawings may be obtained from these drawings without inventive effort for a person skilled in the art.
In the drawings: fig. 1 is a schematic structural diagram of an phosphogypsum impurity removal system based on a visual detection technology.
Fig. 2 is a schematic flow chart of a phosphogypsum impurity removal method based on a visual detection technology.
Fig. 3 is a schematic diagram of an impurity detection sub-model according to the present invention.
Fig. 4 is a schematic diagram of a parameter generation sub-model structure provided by the present invention.
Detailed Description
For the purpose of making apparent the objects, technical solutions and advantages of the present invention, the present invention will be further described in detail with reference to the following examples and the accompanying drawings, wherein the exemplary embodiments of the present invention and the descriptions thereof are for illustrating the present invention only and are not to be construed as limiting the present invention.
Example 1: the embodiment of the invention discloses an phosphogypsum impurity removal system based on a visual detection technology, which is shown by referring to fig. 1 and comprises the following components: phosphogypsum processing structure, impurity snatch parameter generation structure and impurity removal structure all set up near phosphogypsum processing structure, and impurity snatch parameter generation structure is connected with impurity removal structure.
(1) And the phosphogypsum treatment structure is used for screening the input target phosphogypsum material to obtain phosphogypsum to be treated containing target impurities, and carrying out dispersion treatment on the phosphogypsum to be treated.
Specifically, the target phosphogypsum material is the original phosphogypsum which needs to be subjected to harmful impurity removal, and because organic impurities in the phosphogypsum and organic flocculants introduced in the phosphoric acid production process can lead the phosphogypsum to contain a large amount of organic matters and eutectic phosphorus, the organic matters include but are not limited to ethylene glycol methyl ether acetate, isothiocyanato methane, 32 methoxy n-pentane, 22 ethyl 21 and 32 dioxolane, most of the organic matters are removed during phosphoric acid filtration, a small amount of the organic matters still remain in the phosphogypsum, the eutectic phosphorus is indissoluble phosphorus entering a phosphogypsum lattice, and the organic matters and the eutectic phosphorus serve as the harmful impurities of the phosphogypsum, so that the performance of the phosphogypsum can be degraded. The impurity removing device in phosphogypsum provided by the embodiment can remove the harmful impurities, and a person skilled in the art can select the harmful impurities to be removed according to actual demands, if the harmful impurities to be removed are organic impurities, namely organic matters, the corresponding target impurities are organic impurities, and the organic matters in phosphogypsum are selected as the impurities to be separated by the impurity removing structure; if the harmful impurities to be removed are eutectic phosphorus, the corresponding target impurities are eutectic phosphorus, and the impurity removing structure is to be sorted, namely the eutectic phosphorus in phosphogypsum; if the harmful impurities to be removed are organic matters and eutectic phosphorus, the corresponding target impurities are the organic matters and the eutectic phosphorus, and the impurity removing structure is to be sorted, namely the organic matters and the eutectic phosphorus in phosphogypsum.
In this embodiment, the screening process is aimed at screening phosphogypsum containing these target impurities, and generally, any a priori knowledge can be used to determine the characteristics of phosphogypsum containing the target impurities in advance, and then the screening process is performed by using the corresponding screening technology. For example, if the particle size of phosphogypsum is used as a screening standard, the preset particle size threshold of phosphogypsum containing target impurities is determined based on priori knowledge, and because the phosphogypsum contains harmful impurities and is easy to absorb water and form a lump, the particle size of the target impurities is usually larger than that of phosphogypsum which does not contain harmful impurities normally, phosphogypsum which is larger than the preset particle size threshold can be screened out as phosphogypsum to be treated containing the target impurities; if phosphogypsum shape or geometric characteristics are taken as screening standard, the target physical characteristics of phosphogypsum containing target impurities are determined based on priori knowledge, and phosphogypsum matched with the target physical characteristics can be taken as phosphogypsum to be treated containing target impurities by combining with an image processing technology. In order to facilitate the acquisition of phosphogypsum images, after phosphogypsum to be treated containing target impurities is obtained, the phosphogypsum to be treated can be subjected to dispersion treatment, and the best effect is flattening, for example, the phosphogypsum to be treated can be dispersed in a vibration mode, and can also be flattened by means of a mechanical arm, and the application effect of the embodiment is not affected no matter what dispersion mode is adopted.
The embodiment provides a simple and efficient phosphogypsum treatment structure in the practical application process, which can comprise a feeding port, a transmission device, a screening device, a flattening transmission device and a discharging port. Wherein the feed inlet is the position for inputting the target phosphogypsum material; the conveying device conveys the target phosphogypsum material input from the feeding port to the screening device; the screening device divides phosphogypsum into two parts according to a preset particle size threshold value, the phosphogypsum with the particle size larger than the preset particle size threshold value is used as phosphogypsum to be treated containing target impurities and is conveyed to the flattening conveying device, and the phosphogypsum with the particle size smaller than or equal to the preset particle size threshold value is conveyed to the discharge port; the flattening and conveying device disperses phosphogypsum to be treated in a vibration mode.
Preferably, the screening device can adopt a screen with the mesh number of a preset particle size threshold value, for example, the preset particle size threshold value is 25mm, the screening device is a 25mm screen, the target phosphogypsum material is subjected to vibration screening by using the 25mm screen, fine phosphogypsum passing through the 25mm screen enters an undersize belt, and the undersize belt is directly connected to the discharge port; the phosphogypsum to be treated which is left on the 25mm screen is conveyed to a flattening conveying device for further treatment. As a simple structure easy to realize the dispersing function, the flattening and conveying device can adopt a vibrating belt, phosphogypsum to be treated can be flattened through the vibrating function of the vibrating belt, and the length of the vibrating belt is not less than 5m.
(2) The impurity grabbing parameter generating structure is used for acquiring the dispersed image information of the phosphogypsum to be processed in real time, analyzing and processing the image information to obtain target impurity position information, generating grabbing parameters according to the target impurity position information, and transmitting the grabbing parameters to the impurity removing structure; the grabbing parameters comprise preset grabbing positions and grabbing gesture information of target impurities.
The embodiment provides an impurity grabbing parameter generating structure, which comprises a first control module, an image collector, an image processor, a visual detection module and a first communication module. The first control module is respectively connected with the image collector, the image processor, the visual detection module and the first communication module; the image collector is connected with the image processor; the image processor is connected with the visual detection module; the first communication module is connected with the visual detection module and the impurity removing structure respectively.
The image acquisition device is used for acquiring the dispersed image information of the phosphogypsum to be processed in real time and transmitting the image information to the image processor. To facilitate image acquisition, the image acquisition device may be disposed around a device for performing decentralized processing on phosphogypsum to be processed, and in this embodiment, the image acquisition device may be disposed at a flattening transmission device.
Preferably, the image collector comprises at least one camera of any type including, but not limited to, an RGB camera, an infrared camera, etc., and a camera mounting structure, which may be any one of the camera holders of the related art. For example, the camera mounting mechanism may include a camera mounting plate, a camera mount, an optical axis, and a fixed element; the camera support is arranged on the camera mounting plate and is connected with the optical axis; the camera is slidably arranged on the optical axis, the position of the camera on the optical axis can be adjusted in real time according to the position of phosphogypsum to be processed, and the optical axis is a mechanical part for supporting a rotating part or serving as the rotating part, and plays roles of transmitting motion, torque and the like; the fixing element is used for fixing the camera on the optical axis and can be a screw, a nut and a bolt, riveting, a clip or a clamp, or a bayonet. In order to improve the image quality, a light source can be arranged on the head of the camera, or a light barrier can be arranged.
In this embodiment, the image processor may be an upper computer, a PC, a server, a microprocessor, or a computing chip, and is configured to perform preprocessing on the received image information, and transmit the preprocessed image information to the visual detection module.
The vision detection module is used for analyzing and processing the preprocessed image information by using the trained grabbing parameter generation model to obtain target impurity position information, and generating grabbing parameters according to the target impurity position information; the grabbing parameters comprise preset grabbing positions and grabbing gesture information of target impurities. The training process of grabbing the parametric generative model is as described above and will not be repeated here.
And the first communication module is used for transmitting the grabbing parameters to the impurity removing structure.
The first control module is used for controlling the normal operation of each sub-module in the impurity grabbing parameter generating structure.
(3) And the impurity removing structure is used for grabbing target impurities from the phosphogypsum to be treated according to the grabbing parameters to obtain phosphogypsum from which the target impurities are removed.
The embodiment provides a full-automatic impurity removing structure, which can comprise a plurality of sorting robots, wherein each sorting robot comprises a sorting module, a second control module and a second communication module; the second control module is respectively connected with the sorting module and the second control module; the second communication module is connected with the first communication module.
The second communication module is used for transmitting the received grabbing parameters to the second control module;
And the second control module is used for controlling the sorting module to work according to the grabbing parameters, namely grabbing target impurities from phosphogypsum to be treated according to the matched grabbing postures at the preset grabbing positions, and placing the grabbed target impurities into the preset storage positions. In order to facilitate management, the impurity removing structure can be further provided with a waste storage bin, wherein the waste storage bin is a preset storage position, and each sorting robot places the grabbed target impurities into the waste storage bin.
In order to further ensure that the target impurities are effectively sorted out, the sorting robot can also automatically control the treatment of phosphogypsum to be treated entering the working area thereof, comprising: each sorting robot monitors the dynamic position of phosphogypsum to be processed on the vibrating conveyor belt, and determines the current conveying speed of the vibrating conveyor belt according to the working speed and the quantity of phosphogypsum to be processed entering a preset grabbing area so as to dynamically adjust the conveying speed of the vibrating conveyor belt; the preset grabbing area is used for limiting the moving area range of the corresponding sorting robot on the vibrating conveyor belt.
In order to further improve the impurity removal accuracy, the impurity removal structure may further include a force sensor; the force sensor is respectively connected with the sorting module and the second control module. The force sensor is used for collecting force information of the sorting module and transmitting the force information to the second control module; the second control module judges and processes the force information to obtain a grabbing result, and the grabbing result is transmitted to the first communication module through the second communication module.
In addition, the impurity grabbing parameter generating structure further comprises a result display module; the result display module is respectively connected with the first control module, the visual detection module and the first communication module. The result display module is used for displaying a grabbing result; meanwhile, the vision detection module adjusts grabbing parameters according to grabbing results and related loss functions. In other words, the accuracy of the grabbing parameters can be verified by the impurity grabbing results of the sorting robots, if the sorting robots cannot grab target impurities successfully according to the grabbing parameters, the fact that the accuracy of the grabbing parameters generated at present is poor is indicated, the impurity grabbing parameter generating structure needs to be adjusted and optimized, and the accuracy of grabbing parameters is improved.
Notably, given the extremely fine phosphogypsum particles, which are prone to agglomerating, the agglomerated phosphogypsum can become a weak site for compacting the formed phosphogypsum base layer, greatly shortening the strength and durability of the phosphogypsum base layer. In order to effectively detect the agglomerated components in phosphogypsum, the agglomerated components can be crushed so as to improve the performance of the phosphogypsum mixture.
In this embodiment, the system may further comprise a comminution structure; the crushing structure is arranged behind the impurity removing structure and in front of the discharge hole; and the flattening and conveying device is used for conveying the phosphogypsum with the target impurities removed as agglomerated phosphogypsum to the crushing structure for crushing treatment, and conveying the crushed agglomerated phosphogypsum to the discharge port.
Example 2: the embodiment of the invention discloses a phosphogypsum impurity removal method based on a visual detection technology, which is shown by referring to fig. 2 and comprises the following steps of: s1, constructing a grabbing parameter generation model based on a deep neural network model frame and a neural radiation field, and training the grabbing parameter generation model.
S2, acquiring image information of phosphogypsum to be processed in real time. The image information of the phosphogypsum to be processed can be obtained through a camera, a video camera, a scanner or other equipment with a photographing function, and the image information can be picture information or video information, which is not particularly limited herein.
S3, inputting the image information into a trained grabbing parameter generation model for analysis processing to obtain target impurity position information, and generating grabbing parameters according to the target impurity position information; the grabbing parameters comprise preset grabbing positions and grabbing gesture information of target impurities.
Specifically, image information of phosphogypsum to be processed is obtained in real time, a pre-trained grabbing parameter generation model is called to detect impurity components of the image information and position target impurities, grabbing parameters are generated based on target impurity position information, and the grabbing parameters comprise grabbing gesture information of preset grabbing positions and target impurities.
In the embodiment, the grabbing parameter generation model is a network model which is based on any deep learning algorithm, trains by utilizing a training sample data set and can play a corresponding role; the grabbing parameter generation model has the functions of detecting impurity components of phosphogypsum images, positioning target impurities, and generating grabbing parameters according to target impurity positioning data, wherein the grabbing parameters are what gesture to grab to a certain position. The preset grabbing position is a preset grabbing area.
In this embodiment, the grasping parameter generating model includes an impurity detecting sub-model and a parameter generating sub-model. The impurity detection sub-model is used for detecting impurity components of input phosphogypsum image information to obtain a target impurity image marked with a target impurity position, and transmitting the target impurity image to the parameter generation sub-model; and the parameter generation sub-model is used for generating a three-dimensional model of the target impurity at least one view angle at the dispersion treatment position of the phosphogypsum treatment structure according to the target impurity image, and determining grabbing parameters based on the three-dimensional model.
The training process of the grabbing parameter generation model comprises the following steps:
preprocessing the image information of phosphogypsum to be processed to obtain an impurity grabbing sample data set; the impurity grabbing sample data set comprises a plurality of phosphogypsum images, each phosphogypsum image comprises an impurity image with impurity positions marked with different component types, the impurity image is provided with a corresponding impurity sampling data label, and the impurity sampling data sub-label comprises sampling position information and sampling visual angle information of impurities;
inputting the impurity grabbing sample data set into a deep neural network model frame, outputting a target impurity image marked with the position of the target impurity, and continuously carrying out iterative updating on the deep neural network model according to the target impurity image and the related loss until the first preset model training ending condition is met;
Inputting the target impurity image and the impurity sampling data label corresponding to the target impurity image into a nerve radiation field to obtain an initial nerve radiation field;
constructing an initial variant sub-model for learning the geometric characteristics of the target impurity image on the nerve radiation field, and determining a plurality of visual angles for training the initial nerve radiation field;
And training the initial nerve radiation field for multiple times in an iterative mode according to the target impurity image and the projection area corresponding to each view angle by adopting a mode of training one view angle of the multiple view angles until the training ending condition of the second preset model is met.
The trained deep neural network model framework is an impurity detection sub-model, and the trained nerve radiation field is a parameter generation sub-model. The iterative updating of the grabbing parameter generating model is that the model parameters of the grabbing parameter generating model are updated, that is, the model parameters of the deep neural network model frame and the neural radiation field are updated, for example, a batch random gradient descent method can be used for training the grabbing parameter generating model until a preset model training ending condition is adopted, the preset model training ending condition is a first preset model training ending condition and a second preset model training ending condition, the preset model training ending condition can be that the iteration number reaches a preset value, for example, the first preset model training ending condition can be that the iteration number of the impurity submodels reaches a preset number of times, the second preset model training ending condition can be that the iteration number of the parameter generating submodels reaches a preset value, the preset model training ending condition can also be that the model converges, that is, the impurity detecting submodels converge, the preset model training ending condition can also be that the model precision reaches a preset precision threshold, that is, that the precision of the impurity detecting submodels and the precision of the parameter generating submodels reach corresponding preset precision threshold, which does not affect the implementation of the invention. In addition, before the gradient update iteration, the model needs to initialize a gradient descent algorithm, set epoch (training period), batch_size (batch size), weight update period t, and iteration number iteration. For example, the total number of samples included in the impurity capture sample data set may be 6 ten thousand, the capture parameter generation model is trained for at least 100 training periods, and one training period refers to that model parameters of the neural network are not repeatedly updated by using all training samples in the training set, and a small batch (mini-batch) of data is taken for updating model parameters of the capture parameter generation model each time, so that a training process is completed. In the gradient update iteration process, 500 samples are used per iteration update, and these 500 problem-image samples are referred to as one small batch (mini-batch) of data, i.e., batch_size number of samples. The iteration number iteration refers to the number of training using batch_size samples, and the iteration number iteration=60000/500=120 for one epoch is completed. The weight updating period refers to updating the weight once every iteration t times when the grabbing parameter generating model is trained. And when the preset model training ending condition is reached, the grabbing parameter generating model is the trained grabbing parameter generating model.
Referring to fig. 3, the impurity detection submodel is an end-to-end convolutional neural network, which can realize feature extraction and classification of an input phosphogypsum image, and the network structure comprises 7 layers, including 5 convolutional layers and 2 full-connection layers, wherein the largest pooling layer is connected behind 3 convolutional layers, and finally 2 full-connection layers. In order to prevent overfitting and improve generalization capability, dropout (random discard) can be used in the training stage of the impurity detection submodel to randomly ignore a part of neurons, so that the overfitting phenomenon of the neural network can be relieved. In addition, dropout can be used in the following 2 full connection layers, and some neurons of the input layer and the middle layer are randomly zeroed, so that the training process converges more slowly, but the obtained network model is more robust.
Further, the training process of the deep neural network model framework (i.e., the impurity detection sub-model) may include: acquiring an phosphogypsum image dataset, and carrying out data enhancement on image data in the dataset, including overturning, noise reduction and the like, so as to increase the data volume in the dataset, wherein the number of samples of the dataset is as much as possible, and at least 20000 sheets are ensured; dividing the phosphogypsum image data set into a test set and a training set according to a certain ratio, for example, 3:7, and performing super-parameter optimization on the basis of the training set to obtain the optimal network configuration; verifying the classification capacity of the optimized deep convolutional neural network model by using a test set, if the classification capacity does not meet the test precision requirement of 90%, adjusting model parameters, returning to a training set for retraining until the test precision is met, and obtaining an impurity detection sub-model at the moment; and identifying the input phosphogypsum image by using the impurity detection submodel network after training and testing to obtain a target impurity image of the target impurity position.
Referring to fig. 4, the parameter generation sub-model is implemented by using a neural radiation field model, and a specific training process is as follows: the pose of all three-dimensional points forming the radiation field, namely the 3-dimensional position and orientation of each three-dimensional point, is acquired, is input into a neural network to obtain the radiation field, and according to the camera pose of each three-dimensional point, a volume rendering technology is utilized to obtain a photographing of a three-dimensional scene hidden in the radiation field at the pose T. In practice, the neural network inputs a set of poses (x, y, z, theta, fai) at a time, which corresponds to a set of radiation field values (r, g, b, delta). That is, each pose is input once into the neural network to obtain a complete radiation field. After the complete radiation field is obtained, the photographing can be synthesized. However, during the actual training, the radiation field values of all three-dimensional points may not be generated, and only the radiation field values of the three-dimensional points used for calculating the MSE loss may be generated. The MSE loss function is defined based on the error between the actual value and the predicted value, so that the prediction capability and the error source of the model can be known through the MSE loss function.
In addition, after inputting the target impurity image and the impurity sampling data label corresponding to the target impurity image into the nerve radiation field, the method further comprises the following steps: impurity component information, color characteristic values and symbol distance function values corresponding to the sampling position information are obtained according to the output of the nerve radiation field; for each sampling position information, determining at least two sub-sets of sampling points based on a symbol distance function value corresponding to the current sampling position information; and rendering according to the impurity component information and the color characteristic value of the sampling points in each sampling point subset to generate an impurity component object corresponding to each sampling point subset, and generating an impurity component object based on each impurity component object.
S4, grabbing target impurities from the phosphogypsum to be treated according to grabbing parameters to obtain phosphogypsum with the target impurities removed.
Those of ordinary skill in the art will appreciate that implementing all or part of the above facts and methods may be accomplished by a program to instruct related hardware, the program involved or the program may be stored in a computer readable storage medium, the program when executed comprising the steps of: the corresponding method steps are introduced at this time, and the storage medium may be a ROM/RAM, a magnetic disk, an optical disk, or the like.
The foregoing description of the embodiments has been provided for the purpose of illustrating the general principles of the invention, and is not meant to limit the scope of the invention, but to limit the invention to the particular embodiments, and any modifications, equivalents, improvements, etc. that fall within the spirit and principles of the invention are intended to be included within the scope of the invention.
Claims (9)
1. The phosphogypsum impurity removal method based on the visual detection technology is characterized by comprising the following steps of:
Constructing a grabbing parameter generation model based on a deep neural network model frame and a neural radiation field, and training the grabbing parameter generation model; the trained deep neural network model framework is an impurity detection sub-model, and the trained nerve radiation field is a parameter generation sub-model;
Acquiring image information of phosphogypsum to be processed in real time;
inputting the image information into a trained grabbing parameter generation model for analysis processing to obtain target impurity position information, and generating grabbing parameters according to the target impurity position information; the grabbing parameters comprise preset grabbing positions and grabbing gesture information of target impurities;
grabbing target impurities from phosphogypsum to be treated according to the grabbing parameters to obtain phosphogypsum with the target impurities removed;
training the grabbing parameter generation model, wherein the specific process is as follows:
preprocessing the image information of phosphogypsum to be processed to obtain an impurity grabbing sample data set;
Inputting the impurity grabbing sample data set into a deep neural network model frame, outputting a target impurity image marked with the position of a target impurity, and continuously carrying out iterative updating on the deep neural network model frame according to the target impurity image and related loss until a first preset model training ending condition is met;
inputting the target impurity image into a nerve radiation field to obtain an initial nerve radiation field;
Constructing an initial variant sub-model for learning geometric features of the target impurity image on the nerve radiation field, and determining a plurality of visual angles for training the initial nerve radiation field;
Training the initial nerve radiation field according to the target impurity image and the projection area corresponding to each view angle until a second preset model training ending condition is met;
after inputting the target impurity image and the impurity sampling data label corresponding to the target impurity image into the nerve radiation field, the method further comprises the following steps:
impurity component information, color characteristic values and symbol distance function values corresponding to the sampling position information are obtained according to the output of the nerve radiation field;
for each sampling position information, determining at least two sub-sets of sampling points based on a symbol distance function value corresponding to the current sampling position information;
and rendering according to the impurity component information and the color characteristic value of the sampling points in each sampling point subset to generate an impurity component object corresponding to each sampling point subset, and generating an impurity component object based on each impurity component object.
2. Phosphogypsum impurity removal system based on visual detection technology, characterized by comprising: phosphogypsum treatment structure, impurity grabbing parameter generation structure and impurity removal structure;
the impurity grabbing parameter generating structure and the impurity removing structure are arranged near the phosphogypsum processing structure;
the impurity grabbing parameter generating structure is connected with the impurity removing structure;
The phosphogypsum treatment structure is used for screening an input target phosphogypsum material to obtain phosphogypsum to be treated containing target impurities, and carrying out dispersion treatment on the phosphogypsum to be treated;
The impurity grabbing parameter generating structure is used for acquiring the dispersed image information of phosphogypsum to be processed in real time, analyzing and processing the image information to obtain target impurity position information, generating grabbing parameters according to the target impurity position information, and transmitting the grabbing parameters to the impurity removing structure; the grabbing parameters comprise preset grabbing positions and grabbing gesture information of target impurities;
And the impurity removing structure is used for grabbing target impurities from the phosphogypsum to be treated according to the grabbing parameters to obtain phosphogypsum from which the target impurities are removed.
3. The phosphogypsum impurity removal system based on the visual inspection technology according to claim 2, wherein the phosphogypsum treatment structure comprises a feed inlet, a transmission module, a screening module, a flattening transmission module and a discharge outlet;
the feeding port, the transmission module, the screening module and the flattening transmission module are sequentially connected;
the screening module is connected with the discharge port;
the feeding port is used for inputting target phosphogypsum materials;
the transmission module is used for conveying the target phosphogypsum material to the screening module;
The screening module is used for screening the target phosphogypsum material according to a preset rule to obtain phosphogypsum to be treated and phosphogypsum which is not required to be treated, conveying the phosphogypsum to the flattening transmission module, and conveying the phosphogypsum which is not required to be treated to the discharge port;
and the flattening transmission module is used for carrying out dispersion treatment on the phosphogypsum to be treated.
4. The phosphogypsum impurity removal system based on the visual detection technology according to claim 3, wherein the impurity grabbing parameter generation structure comprises a first control module, an image collector, an image processor, a visual detection module and a first communication module;
The first control module is respectively connected with the image collector, the image processor, the visual detection module and the first communication module;
The image collector is connected with the image processor;
the image processor is connected with the visual detection module;
The first communication module is respectively connected with the visual detection module and the impurity removal structure;
The image acquisition device is used for acquiring the dispersed image information of phosphogypsum to be processed in real time and transmitting the image information to the image processor;
The image processor is used for preprocessing the received image information to obtain preprocessed image information and transmitting the preprocessed image information to the visual detection module;
The vision detection module analyzes and processes the preprocessed image information by using a trained grabbing parameter generation model to obtain target impurity position information, and generates grabbing parameters according to the target impurity position information; the grabbing parameters comprise preset grabbing positions and grabbing gesture information of target impurities;
The first communication module is used for transmitting the grabbing parameters to the impurity removing structure;
The first control module is used for controlling the normal operation of each sub-module in the impurity grabbing parameter generating structure.
5. The phosphogypsum impurity removal system based on visual inspection technology of claim 4, wherein the impurity removal structure comprises a plurality of sorting robots;
The sorting robot comprises a sorting module, a second control module and a second communication module;
The second control module is respectively connected with the sorting module and the second control module;
the second communication module is connected with the first communication module;
The second communication module is used for transmitting the received grabbing parameters to the second control module;
And the second control module is used for controlling the sorting module to grasp target impurities from the phosphogypsum to be treated according to the grasping parameters.
6. The phosphogypsum impurity removal system based on visual inspection technology of claim 5, wherein the impurity removal structure further comprises a force sensor;
the force sensor is respectively connected with the sorting module and the second control module;
the force sensor is used for collecting force information of the sorting module and transmitting the force information to the second control module; the second control module judges and processes the force information to obtain a grabbing result, and the grabbing result is transmitted to the first communication module through the second communication module.
7. The phosphogypsum impurity removal system based on the visual inspection technology of claim 6, wherein the impurity grabbing parameter generation structure further comprises a result display module;
The result display module is respectively connected with the first control module, the visual detection module and the first communication module;
the result display module is used for displaying the grabbing result; and the vision detection module adjusts the grabbing parameters according to the grabbing results and the related loss function.
8. The phosphogypsum impurity removal system based on visual inspection technology of claim 7, further comprising a comminution structure;
The crushing structure is arranged between the flattening transmission module and the discharge port, and is used for crushing phosphogypsum for removing target impurities and conveying the crushed phosphogypsum to the discharge port.
9. A computer readable storage medium having stored therein at least one program code loaded and executed by a computer processor to implement a phosphogypsum impurity removal method based on visual inspection techniques of claim 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410627938.XA CN118204275B (en) | 2024-05-21 | 2024-05-21 | Phosphogypsum impurity removal method and phosphogypsum impurity removal system based on visual detection technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410627938.XA CN118204275B (en) | 2024-05-21 | 2024-05-21 | Phosphogypsum impurity removal method and phosphogypsum impurity removal system based on visual detection technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118204275A CN118204275A (en) | 2024-06-18 |
| CN118204275B true CN118204275B (en) | 2024-08-20 |
Family
ID=91450770
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410627938.XA Active CN118204275B (en) | 2024-05-21 | 2024-05-21 | Phosphogypsum impurity removal method and phosphogypsum impurity removal system based on visual detection technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118204275B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113289925A (en) * | 2021-05-21 | 2021-08-24 | 中国矿业大学(北京) | Intelligent sorting system and method for sundries in coal based on deep learning |
| CN114972632A (en) * | 2022-04-21 | 2022-08-30 | 阿里巴巴达摩院(杭州)科技有限公司 | Image processing method and device based on nerve radiation field |
| CN116543086A (en) * | 2023-05-04 | 2023-08-04 | 阿里巴巴达摩院(杭州)科技有限公司 | Nerve radiation field processing method and device and electronic equipment |
| CN116984269A (en) * | 2023-06-14 | 2023-11-03 | 沈阳工业大学 | Gangue grabbing method and system based on image recognition |
| CN117325170A (en) * | 2023-10-27 | 2024-01-02 | 武汉工程大学 | Method for grabbing hard disk rack based on depth vision guiding mechanical arm |
Family Cites Families (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8396247B2 (en) * | 2008-07-31 | 2013-03-12 | Microsoft Corporation | Recognizing actions of animate objects in video |
| CN206296207U (en) * | 2016-12-19 | 2017-07-04 | 江西康缘桔都药业有限公司 | A kind of full automatic lamp detecting machine |
| CN110298879B (en) * | 2018-03-21 | 2021-08-27 | 北京猎户星空科技有限公司 | Method and device for determining pose of object to be grabbed and electronic equipment |
| CN117413300A (en) * | 2021-03-26 | 2024-01-16 | 上海科技大学 | Method and system for training quantized nerve radiation field |
| CN113920433A (en) * | 2021-10-22 | 2022-01-11 | Oppo广东移动通信有限公司 | Method and apparatus for analyzing surface material of object |
| EP4392935A1 (en) * | 2021-11-15 | 2024-07-03 | Google LLC | Robustifying nerf model novel view synthesis to sparse data |
| US20240123620A1 (en) * | 2022-09-29 | 2024-04-18 | Nvidia Corporation | Grasp pose prediction |
| CN117095128A (en) * | 2023-09-04 | 2023-11-21 | 清华大学深圳国际研究生院 | Priori-free multi-view human body clothes editing method |
| CN117563766A (en) * | 2023-12-14 | 2024-02-20 | 北京建工资源循环利用有限公司 | Sorting method and device for construction waste |
| CN117710583A (en) * | 2023-12-18 | 2024-03-15 | 中铁第四勘察设计院集团有限公司 | Space-to-ground image three-dimensional reconstruction method, system and equipment based on nerve radiation field |
-
2024
- 2024-05-21 CN CN202410627938.XA patent/CN118204275B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113289925A (en) * | 2021-05-21 | 2021-08-24 | 中国矿业大学(北京) | Intelligent sorting system and method for sundries in coal based on deep learning |
| CN114972632A (en) * | 2022-04-21 | 2022-08-30 | 阿里巴巴达摩院(杭州)科技有限公司 | Image processing method and device based on nerve radiation field |
| CN116543086A (en) * | 2023-05-04 | 2023-08-04 | 阿里巴巴达摩院(杭州)科技有限公司 | Nerve radiation field processing method and device and electronic equipment |
| CN116984269A (en) * | 2023-06-14 | 2023-11-03 | 沈阳工业大学 | Gangue grabbing method and system based on image recognition |
| CN117325170A (en) * | 2023-10-27 | 2024-01-02 | 武汉工程大学 | Method for grabbing hard disk rack based on depth vision guiding mechanical arm |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118204275A (en) | 2024-06-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108277720B (en) | Asphalt mixing station aggregate grading online detection and anti-overflow control method and system | |
| CA3118067C (en) | Apparatus for analyzing composition of electronic and electrical device part scraps, device for processing electronic and electrical device part scraps, and method for processing electronic and electrical device part scraps | |
| CN206139527U (en) | Panoramic vision potato is selected separately and defect detecting device | |
| CN109916826B (en) | Solid waste online identification system and method based on hyperspectral detection | |
| CN109785378B (en) | Online ore granularity detection equipment based on atlas image algorithm analysis technology | |
| CN104867144B (en) | IC element welding point defect detection methods based on mixed Gauss model | |
| US5519793A (en) | Apparatus and method for computer vision measurements | |
| CN107999405A (en) | A kind of building waste on-line sorting system and method for sorting | |
| CN112024424A (en) | Man-machine cooperation type garbage sorting system | |
| CN112191355B (en) | Ore processing method, device, electronic equipment and computer readable storage medium | |
| CN103344583A (en) | Praseodymium-neodymium (Pr/Nd) component content detection system and method based on machine vision | |
| CN116637902B (en) | Intelligent multifunctional recycling treatment method and system for waste residue soil | |
| CN207742057U (en) | A kind of aggregate on-Line Monitor Device | |
| CN108108678B (en) | Tungsten ore identification and separation method | |
| CN116493290A (en) | Intelligent sorting method and system for plastic particle production | |
| CN118204275B (en) | Phosphogypsum impurity removal method and phosphogypsum impurity removal system based on visual detection technology | |
| CN115095384A (en) | Intelligent ore positioning and crushing control method and device | |
| CN117611961A (en) | Ore granularity identification method and system based on particle size distribution characteristics of crushed products | |
| CN117181600A (en) | Aluminum and aluminum alloy sorting method and device based on pulsating airflow sorting | |
| CN1488119A (en) | Resolution enhancement by nearest neighbor classified filtering | |
| CN113112482A (en) | PCB defect detection method based on attention mechanism network | |
| CN115818166B (en) | Unmanned automatic control method and system for continuous system of wheel bucket | |
| CN115953568A (en) | Intelligent mineral sorting method based on lightweight target detection framework | |
| CN115452736A (en) | Control method for vanadium ore separation | |
| AU2021105929A4 (en) | A Multi-Parameter Prediction Model For Coal Sample Hardiness |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |