CN118794898B - Strip steel surface cleanliness detection method, detection device and system - Google Patents
Strip steel surface cleanliness detection method, detection device and system Download PDFInfo
- Publication number
- CN118794898B CN118794898B CN202411267711.5A CN202411267711A CN118794898B CN 118794898 B CN118794898 B CN 118794898B CN 202411267711 A CN202411267711 A CN 202411267711A CN 118794898 B CN118794898 B CN 118794898B
- Authority
- CN
- China
- Prior art keywords
- strip
- peak
- steel
- strip steel
- data points
- 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
- 229910000831 Steel Inorganic materials 0.000 title claims abstract description 327
- 239000010959 steel Substances 0.000 title claims abstract description 327
- 230000003749 cleanliness Effects 0.000 title claims abstract description 69
- 238000001514 detection method Methods 0.000 title claims abstract description 24
- 239000000126 substance Substances 0.000 claims abstract description 107
- 230000003595 spectral effect Effects 0.000 claims abstract description 43
- 238000000034 method Methods 0.000 claims abstract description 37
- 239000002436 steel type Substances 0.000 claims abstract description 36
- 238000001228 spectrum Methods 0.000 claims description 92
- 239000000463 material Substances 0.000 claims description 68
- 238000004140 cleaning Methods 0.000 claims description 23
- 230000006870 function Effects 0.000 claims description 16
- 238000004422 calculation algorithm Methods 0.000 claims description 13
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 claims description 11
- 238000009826 distribution Methods 0.000 claims description 11
- 239000010731 rolling oil Substances 0.000 claims description 10
- 229910052799 carbon Inorganic materials 0.000 claims description 9
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 claims description 8
- 230000001174 ascending effect Effects 0.000 claims description 8
- 230000011218 segmentation Effects 0.000 claims description 8
- 238000010521 absorption reaction Methods 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 7
- 239000006185 dispersion Substances 0.000 claims description 7
- 239000000839 emulsion Substances 0.000 claims description 6
- 238000010183 spectrum analysis Methods 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 11
- 238000005516 engineering process Methods 0.000 description 8
- UXFQFBNBSPQBJW-UHFFFAOYSA-N 2-amino-2-methylpropane-1,3-diol Chemical compound OCC(N)(C)CO UXFQFBNBSPQBJW-UHFFFAOYSA-N 0.000 description 4
- 101150035093 AMPD gene Proteins 0.000 description 4
- 239000011248 coating agent Substances 0.000 description 3
- 238000000576 coating method Methods 0.000 description 3
- 241000894007 species Species 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000000137 annealing Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000005096 rolling process Methods 0.000 description 2
- 206010039509 Scab Diseases 0.000 description 1
- 238000002835 absorbance Methods 0.000 description 1
- 239000000853 adhesive Substances 0.000 description 1
- 230000001070 adhesive effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 239000003344 environmental pollutant Substances 0.000 description 1
- 239000004519 grease Substances 0.000 description 1
- 239000012535 impurity Substances 0.000 description 1
- 239000012770 industrial material Substances 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- JEIPFZHSYJVQDO-UHFFFAOYSA-N iron(III) oxide Inorganic materials O=[Fe]O[Fe]=O JEIPFZHSYJVQDO-UHFFFAOYSA-N 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 229910052751 metal Inorganic materials 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000005554 pickling Methods 0.000 description 1
- 231100000719 pollutant Toxicity 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 238000005406 washing Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
- G01N21/27—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands using photo-electric detection ; circuits for computing concentration
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
- G01N21/27—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands using photo-electric detection ; circuits for computing concentration
- G01N21/274—Calibration, base line adjustment, drift correction
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/89—Investigating the presence of flaws or contamination in moving material, e.g. running paper or textiles
- G01N21/8914—Investigating the presence of flaws or contamination in moving material, e.g. running paper or textiles characterised by the material examined
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N2021/1765—Method using an image detector and processing of image signal
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/89—Investigating the presence of flaws or contamination in moving material, e.g. running paper or textiles
- G01N21/8914—Investigating the presence of flaws or contamination in moving material, e.g. running paper or textiles characterised by the material examined
- G01N2021/8918—Metal
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Textile Engineering (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Investigating And Analyzing Materials By Characteristic Methods (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
本申请涉及带钢光谱分析技术领域，具体涉及带钢表面清洁度检测方法、检测装置及系统，该方法包括：采集带钢表面的光谱数据图像，分别采集每种物质的光谱数据；基于各带钢数据点的峰值数量，将所有带钢数据点划分到各带钢种类集合；将各带钢种类集合中带钢数据点划分为各峰位相近类；基于任一峰位相近类中任两个带钢数据点之间在所有相同序号处的峰值差异，以及所述峰值差异的离散程度，确定所述任两个带钢数据点之间的峰值偏移程度；将各峰位相近类中的所有带钢数据点划分为各物质类；确定任一物质类中的带钢数据点是否由清洁影响物质形成；确定带钢的表面清洁度。本申请旨在提高对带钢表面清洁度进行检测的准确性。
The present application relates to the technical field of strip steel spectral analysis, and specifically to a strip steel surface cleanliness detection method, detection device and system, the method comprising: collecting a spectral data image of the strip steel surface, collecting spectral data of each substance respectively; dividing all the strip steel data points into each strip steel type set based on the number of peaks of each strip steel data point; dividing the strip steel data points in each strip steel type set into each peak position similar class; determining the peak deviation degree between any two strip steel data points in any peak position similar class based on the peak difference at all the same sequence numbers and the discrete degree of the peak difference; dividing all the strip steel data points in each peak position similar class into each substance class; determining whether the strip steel data points in any substance class are formed by a cleaning-affecting substance; and determining the surface cleanliness of the strip steel. The present application aims to improve the accuracy of detecting the surface cleanliness of the strip steel.
Description
Technical Field
The application relates to the technical field of band steel spectrum analysis, in particular to a band steel surface cleanliness detection method, a band steel surface cleanliness detection device and a band steel surface cleanliness detection system.
Background
Strip steel is an important industrial material, and is specially produced by various steel rolling enterprises in order to meet the diversified demands of different industrial departments on metal or mechanical products in industrial production. The cleanliness of the surface of the strip steel refers to the adhesion degree of pollutants such as grease, dirt, rust, oxide scale and the like on the surface of the strip steel, and is one of important indexes for measuring the surface quality of the strip steel. If the surface cleanliness of the strip steel does not reach the standard, defects such as grain forming, scab forming and the like can be formed in the production process, particularly when a coated plate is produced, the residual carbon quantity after annealing is too high, the surface quality of a finished product can be influenced, and the smooth adhesion is caused.
With the rapid development of spectroscopic technology, the spectroscopic technology has increasingly outstanding excellent performance in terms of object composition detection, and particularly, the spectroscopic technology has shown rapid and accurate detection capability when analyzing residues on the surface of an object. However, in practical applications, materials such as strip steel are usually located on a conveyor belt, and the movement of the conveyor belt can cause vibration of the strip steel, so that spectral lines acquired by a spectrometer shift, and errors occur when substances are identified through spectral data, thereby affecting the accuracy of the detection result of the cleanliness of the strip steel surface.
Disclosure of Invention
In view of the foregoing, it is necessary to provide a method, an apparatus and a system for detecting the surface cleanliness of a strip steel, which improve the accuracy of detecting the surface cleanliness of the strip steel compared with the conventional method for detecting the surface cleanliness of the strip steel:
in a first aspect, an embodiment of the present application provides a method for detecting surface cleanliness of a strip steel, including the steps of:
Collecting spectrum data images of the surface of the strip steel under a plurality of preset wavelengths, wherein one pixel is recorded as a strip steel data point, and spectrum data of each substance under the plurality of preset wavelengths are respectively collected, wherein cleaning influencing substances exist;
Acquiring peaks of spectrum data corresponding to all wavelengths of each strip steel data point; dividing all strip steel data points into strip steel type sets based on the number of the peak values of each strip steel data point;
calculating the average level of the wavelength difference between any two strip steel data points in each strip steel type set at all the peak points with the same serial numbers, and dividing the strip steel data points in each strip steel type set into similar types of peak positions based on the distribution condition of the strip steel type set corresponding to all the average levels;
determining the peak deviation degree between any two strip steel data points based on the peak difference between any two strip steel data points in any one similar class of peak positions at all the same serial numbers and the discrete degree of the peak difference;
Obtaining each cluster of the peak position similar class corresponding to all the peak value offset degrees, and dividing all strip steel data points in each peak position similar class into each material class by combining the difference between the peak value offset degrees in different clusters of each peak position similar class and the fluctuation degree of the peak value offset degree in each cluster;
Determining whether strip steel data points in any material class are formed from cleaning-affecting materials based on similarity between all strip steel data points in the material class and spectral data of each material at the same wavelength;
The surface cleanliness of the strip steel is determined based on the duty ratio of strip steel data points belonging to cleaning influencing substances in the spectrum data image.
In one embodiment, the method for dividing all strip steel data points into each strip steel type set comprises the following steps: dividing all the strip steel data points with the same peak value number into the same strip steel type set.
In one embodiment, the process of classifying the strip steel data points in each strip steel type set into each peak position similar type is as follows:
The wavelengths of all the strip steel data points at all the peak points are arranged in ascending order to form a spectrum peak position sequence of each strip steel;
taking the average value of the differences of all the same position elements in the strip steel spectrum peak position sequences of any two strip steel data points as a strip steel vibration offset value between any two strip steel data points;
threshold segmentation is carried out on any band steel class set corresponding to all band steel vibration offset values, and all band steel vibration offset values smaller than or equal to a segmentation threshold value in any band steel class set form a parity band steel vibration offset set;
And taking two strip steel data points corresponding to each strip steel vibration offset value in any synchronous strip steel vibration offset set as one class, combining at least one class with the same strip steel data point, and taking the combined class as the similar class of each peak position.
In one embodiment, the determining of the peak offset degree is:
arranging all peaks of each strip steel data point according to the ascending order of the wavelength to form each strip steel spectrum peak value sequence;
The differences among all the same position elements in the strip steel spectrum peak value sequences of any two strip steel data points are formed into a strip steel peak value deviation sequence according to the element positions;
the expression of the peak deviation degree between any two strip steel data points is as follows:
; in the formula, Representing the peak deviation degree between the xth and the y strip steel data points in any peak position similar class;、 respectively representing the strip steel spectrum peak sequences of the x-th strip steel data point and the y-th strip steel data point in any similar peak position class; Representing a strip steel peak deviation sequence between the xth strip steel data point and the y strip steel data point in any peak position similar class; cos () represents a cosine similarity function; MAD () represents a discrete degree function; Indicating that a value greater than 0 is preset.
In one embodiment, the process of classifying all the strip steel data points in each similar peak position class into each material class is as follows:
obtaining all cluster clusters corresponding to the peak value offset degree by adopting a clustering algorithm, and obtaining a cluster with the maximum peak value offset degree mean value and a cluster with the minimum mean value, which are respectively marked as a maximum cluster and a minimum cluster;
The expression of the substance type judgment value of any similar peak position type is as follows:
; wherein GF represents a substance type determination value of any of the similar peak positions; p, Q respectively represent the maximum cluster and the minimum cluster of any similar peak position class; 、 The discrete degree of the peak value offset degree in the maximum cluster and the minimum cluster of any similar peak position class is respectively represented; max (), min () respectively represent a maximum function and a minimum function;
when the material type judgment value of any one of the similar peak positions is less than or equal to 0, taking any one of the similar peak positions as one material type; otherwise, dividing two strip steel data points corresponding to each peak value deviation degree in the maximum cluster into one class, merging the classes with at least one identical strip steel data point, and marking each class obtained after merging as a substance class.
In one embodiment, the determining whether any of the substance classes corresponds to a cleaning affecting substance is:
calculating the average value of spectrum data of all strip steel data points in any material class at the same wavelength, and arranging the average value according to the ascending order of the wavelengths to form a material waiting sequence;
The spectrum data of each substance are arranged according to the ascending order of the wavelength to form each spectrum sequence;
And calculating the similarity between the substance undetermined sequence of any substance class and the spectrum sequences of various substances, and judging that the strip steel data points in any substance class are formed by the cleaning influence substances when the substances corresponding to the spectrum sequences with the maximum similarity are the cleaning influence substances.
In one embodiment, the surface cleanliness is expressed as:
; wherein B represents the surface cleanliness of the strip steel; m represents the number of band steel data points in the spectrum data image; n represents the number of strip data points belonging to the cleaning-influencing substance in the spectral data image.
In one embodiment, when the surface cleanliness of the strip steel is greater than a preset cleanliness threshold value, judging that the quality of the strip steel is qualified; otherwise, judging that the quality of the strip steel is unqualified.
In a second aspect, an embodiment of the present application further provides a device for detecting surface cleanliness of a strip steel, where the device includes:
The device comprises a spectrum data acquisition module, a data acquisition module and a data processing module, wherein the spectrum data acquisition module is used for acquiring spectrum data images of the surface of the strip steel at a plurality of preset wavelengths, wherein one pixel is recorded as a strip steel data point, and spectrum data of each substance at the plurality of preset wavelengths are respectively acquired, wherein cleaning influencing substances exist;
The spectrum data classification module is used for acquiring peaks of spectrum data corresponding to all the wavelengths of each strip steel data point; dividing all strip steel data points into strip steel type sets based on the number of the peak values of each strip steel data point;
calculating the average level of the wavelength difference between any two strip steel data points in each strip steel type set at all the peak points with the same serial numbers, and dividing the strip steel data points in each strip steel type set into similar types of peak positions based on the distribution condition of the strip steel type set corresponding to all the average levels;
determining the peak deviation degree between any two strip steel data points based on the peak difference between any two strip steel data points in any one similar class of peak positions at all the same serial numbers and the discrete degree of the peak difference;
Obtaining each cluster of the peak position similar class corresponding to all the peak value offset degrees, and dividing all strip steel data points in each peak position similar class into each material class by combining the difference between the peak value offset degrees in different clusters of each peak position similar class and the fluctuation degree of the peak value offset degree in each cluster;
A cleanliness detection module for determining whether strip steel data points in any material class are formed by cleaning-affecting substances based on similarity between all strip steel data points in the any material class and spectral data of each material at the same wavelength;
The surface cleanliness of the strip steel is determined based on the duty ratio of strip steel data points belonging to cleaning influencing substances in the spectrum data image.
In a third aspect, an embodiment of the present application further provides a system for detecting the cleanliness of a strip steel surface, including a memory, a processor, and a computer program stored in the memory and running on the processor, where the processor implements the steps of any one of the above methods for detecting the cleanliness of a strip steel surface when executing the computer program.
The application has at least the following beneficial effects:
according to the method, the number of peaks in the spectrum data of the strip steel data points is analyzed, the strip steel spectrum data points are classified for the first time, the peak position difference is utilized to classify the first classification result for the second time, the secondary classification result is combined with the difference between the peaks to classify the second classification result, each material class is obtained, the strip steel data points in each material class are formed by the same material, and whether the strip steel data points in each material class are formed by cleaning influencing materials is judged according to the average level of the spectrum data of the strip steel data points in each material class relative to the similarity of the spectrum data of each material class, so that the influence of spectral line deviation caused by vibration of a conveyor belt on the material identification result is eliminated, and the accuracy of detecting the strip steel surface cleanliness is improved.
Drawings
In order to more clearly illustrate the embodiments of the application or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the application, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart showing steps of a method for detecting the cleanliness of a strip steel surface according to an embodiment of the present application;
FIG. 2 is a schematic diagram of the detection of the AMPD algorithm;
FIG. 3 is a schematic diagram of the synthesis of similar classes of peak positions;
fig. 4 is a schematic diagram of an acquisition flow of a peak position similarity class.
Detailed Description
In describing embodiments of the present application, words such as "exemplary," "or," "such as," and the like are used to mean serving as an example, instance, or illustration. Any embodiment or design described herein as "exemplary" or "for example" should not be construed as preferred or advantageous over other embodiments or designs. Rather, the use of words such as "exemplary," "or," "such as," and the like are intended to present related concepts in a concrete fashion.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. It is to be understood that, unless otherwise indicated, a "/" means or.
It should be further noted that the terms "first" and "second" are used herein to distinguish similar objects from each other and are not used to describe a particular order or sequence.
The following specifically describes a specific scheme of the method, the device and the system for detecting the surface cleanliness of the strip steel provided by the application with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of steps of a method for detecting the cleanliness of a strip steel surface according to an embodiment of the application is shown, the method includes the following steps:
step 1, collecting spectrum data images of the surface of the strip steel under a plurality of preset wavelengths, wherein one pixel is recorded as a strip steel data point, and spectrum data of each substance under the plurality of preset wavelengths are respectively collected, wherein cleaning influencing substances exist.
And a multispectral imager is vertically arranged right above a conveying belt for conveying the strip steel, so that the shooting angle of the multispectral imager is ensured to form 90 degrees with the horizontal plane, and the multispectral imager is directly aligned to the surface of the strip steel, so that accurate spectrum data are acquired. Because the main residues on the surface of the strip steel are rolling oil and carbon elements, the two substances have strong reaction within the wavelength range of 200-1000 nm, so that the spectrum absorptivity data of the surface of the strip steel within the wavelength range of 200-1000 nm are acquired through a multispectral imager and recorded as spectrum data, and one pixel is recorded as a strip steel data point for the acquired spectrum data image.
And respectively acquiring spectral data of each substance within the wavelength range of 200-1000 nm by adopting a spectral imager. The materials comprise rolling oil, carbon element, emulsion, iron powder and strip steel, wherein the rolling oil, the carbon element, the emulsion and the iron powder are cleaning influencing materials.
Because the spectrum data is likely to be lost in the process of acquisition or transmission, a missing value filling method is adopted to fill the missing spectrum data, and all spectrum data of each strip steel data point are respectively arranged into each spectrum sequence according to the sequence from small to large of the wavelength for the full spectrum data; all spectrum data of various substances are respectively arranged in order of wavelengths from small to large to form each spectrum sequence.
In this embodiment, the mean value filling method is adopted for filling, and as other embodiments, on the basis of realizing data filling, an implementer may adopt other existing technologies for data filling, for example, a median filling method, an interpolation method, and the like, and the present application is not limited in particular.
Step 2, obtaining peak values of spectrum data corresponding to all the wavelengths of each strip steel data point; all strip data points are divided into each strip class set based on the number of peaks of each strip data point.
The main production process flow of the strip steel comprises pickling, rolling, alkaline washing, annealing, coating and slitting. Before coating the strip steel, the surface cleaning is required to be ensured so as to ensure the quality and the adhesive force of the coating and produce a strip steel product with higher quality. Because the strip steel products are large in size and mass and cannot be carried manually, the strip steel products are generally carried by a conveyor belt. When the strip steel is conveyed to a region to be detected through the conveyor belt, the spectral line of the strip steel spectral data acquired by the multispectral imager is shifted due to the vibration of the conveyor belt, so that the characteristic peak of the strip steel spectral data is shifted. However, since the spectral data is determined by the absorption of a substance by wavelength, the order of the characteristic peaks of one substance does not change although the characteristic peaks are shifted.
And respectively taking the spectrum sequence of each strip steel data point as the input of an automatic multi-scale peak value searching (Automatic multiscale-based peak detection, AMPD) algorithm, outputting the peak value and the wavelength corresponding to the peak value in the spectrum sequence of each strip steel data point, and arranging the wavelengths corresponding to all the peak values in the spectrum sequence of each strip steel data point in order from small to large to form each strip steel spectrum peak position sequence. The AMPD algorithm is a well-known technique, and the present application will not be described in detail. A schematic diagram of the detection of the AMPD algorithm is shown in fig. 2, in which the horizontal axis represents wavelength and the vertical axis represents absorbance.
Because different substances generate unique peak combinations in a specific wave band due to the difference of absorption characteristics, substances with different peak numbers are generated in a certain wave band, and are different substances. Therefore, aiming at all the strip steel data points, classifying is carried out according to the lengths of the strip steel spectrum peak position sequences, and all the strip steel data points with the same strip steel spectrum peak position sequences are classified into the same strip steel type set.
And step 3, calculating the average level of the wavelength difference between any two strip steel data points in each strip steel type set at all the peak points with the same serial numbers, and dividing the strip steel data points in each strip steel type set into similar types of peak positions based on the distribution condition of each strip steel type set corresponding to all the average levels.
Taking any set of strip types as an example, the peaks of the spectral data of the same species are closely spaced, although the spectral lines are shifted due to the vibration of the conveyor belt.
Based on the difference between the band steel spectrum peak position sequences of any two band steel data points in the band steel type set, determining a band steel vibration offset value between the any two band steel data points, wherein the expression is as follows:
; in the formula, Representing a strip vibration offset value between the ith and jth strip data points in the strip class set; n represents the length of a strip steel spectrum peak position sequence of any strip steel data point in the strip steel type set;、 Respectively representing the kth element in the band steel spectrum peak position sequence of the ith and jth band steel data points in the band steel type set.
It should be noted that: in an ideal case, for two strip data points of the same material, if the spectral data is not shifted, the elements at the same position in the strip spectral peak sequences of the two strip data points are identical, so that the absolute values of the differences of the elements in the two strip spectral peak sequencesThe value of (2) is 0, and the vibration offset value of the strip steel between strip steel data points is 0; when the spectrum data are shifted, compared with different substances, the element phase difference in the band steel spectrum peak position sequence of the same substance is smaller, namelySmaller, so that the strip vibration offset value between strip data pointsWhen all the vibration offset values of the strip steel are smaller, the strip steel data points in the strip steel type collection correspond to the same substance, and if the substance is a substance which influences the strip steel cleanliness, such as rolling oil, carbon powder or iron powder, the more the number of elements in the strip steel type collection is, the lower the strip steel surface cleanliness is.
If the strip steel type set is composed of strip steel data points of different substances, the strip steel vibration offset values among the strip steel data points of different substances are larger, and the strip steel vibration offset values among the strip steel data points of the same substance are smaller, so that all strip steel vibration offset values are used as the input of a threshold segmentation algorithm, the segmentation threshold of the strip steel vibration offset values is output, all strip steel vibration offset values smaller than or equal to the segmentation threshold form a peer strip steel vibration offset set, and all strip steel vibration offset values larger than the segmentation threshold form a different strip steel vibration offset set.
In this embodiment, the division threshold is obtained by using an oxford threshold division algorithm, and as other embodiments, on the basis that the division threshold can be obtained, an implementer may obtain the division threshold by using other existing technologies, for example, global threshold division, iterative threshold division, and the like, and the present application is not limited in particular.
And taking two strip steel data points corresponding to each strip steel vibration offset value in the same-position strip steel vibration offset set as one class, combining at least one class with the same strip steel data point, and taking the combined class as the similar class of each peak position.
The synthetic schematic diagram of the similar peak position class is shown in fig. 3, for example, the corresponding binary groups of each element in the vibration offset set of the co-located strip steel are (1, 2), (2, 5), (5, 8), (3, 7), (9, 3) and (7, 9), and one binary group consists of two strip steel data points and is regarded as a class. Since (1, 2), (2, 5) have the same strip data point 2, the two classes are combined into (1, 2, 5), and since (5, 8) also has strip data point 5, (1, 2, 5) is combined with (5, 8), resulting in (1, 2,5, 8). And (3, 7), (9, 3), (7, 9) are combined into (3, 7, 9) in the same way. Then (1, 2,5, 8) and (3, 7, 9) are both of similar peak position classes. A schematic diagram of the acquisition flow of the peak position similar class is shown in FIG. 4.
And 4, determining the peak deviation degree between any two strip steel data points based on the peak difference between any two strip steel data points in any similar peak position class at all the same serial numbers and the discrete degree of the peak difference.
Taking any similar peak position as an example, when calculating the cleanliness of the surface of the strip steel, substances such as rolling oil, carbon powder, iron powder and the like are not uniformly distributed on the surface of the strip steel, and may have different concentrations in different areas. Since the absorption rate of different concentrations of the same substance on the same wavelength is different, the higher the concentration of the substance is, the stronger the absorption capacity of the substance on the wavelength is, whereas the lower the concentration is, the lower the absorption capacity of the substance on the wavelength is, and the property exists for all wavelengths, therefore, under the different concentrations of the same substance, the peak value of the spectrum characteristic peak has a certain difference, so that errors are easy to occur when the substance is identified through spectrum data, and further errors occur when the cleanliness of the strip steel surface is calculated according to the identification result of the substance. Thus, the problem of material identification errors of strip steel data points caused by concentration differences needs to be eliminated.
And arranging all peaks in the spectrum sequence of each strip steel data point according to the order of the wavelengths from small to large to form each strip steel spectrum peak sequence.
Based on the analysis, calculating the difference between the same position elements in the strip steel spectrum peak value sequence of any two strip steel data points in the similar peak position class, and marking the difference as strip steel peak value deviation; when two strip spectrum peak sequences are formed by the same substance and the same concentration, the substances have the same absorptivity at the same wavelength, so that the difference between the same position elements of the two strip spectrum peak sequences is smaller, namely the value of the strip peak deviation is smaller.
In this embodiment, the difference between the same position elements in the strip spectrum peak sequence is the absolute value of the difference, and as other embodiments, on the basis of being able to measure the difference between the same position elements in the strip spectrum peak sequence, an operator may measure the difference between the same position elements in the strip spectrum peak sequence by using other calculation methods, for example, the ratio, the square of the difference, etc., and the application is not limited in particular.
And forming a strip steel peak value deviation sequence according to the element positions by using the strip steel peak value deviations among all the elements at the same position in the strip steel spectrum peak value sequence of any two strip steel data points.
Although the concentration differences cause differences between peaks, the differences between peaks exhibit a regular change, i.e., the position of the peaks shift with concentration changes without changing the peak shape or introducing new peaks.
From this, based on the similarity of the strip spectrum peak sequence between any two strip data points and the degree of dispersion of the strip peak bias in the strip peak bias sequence therebetween, the degree of peak bias between any two strip data points is determined, with the expression:
; in the formula, Representing the peak deviation degree between the xth and the yh strip steel data points in the similar peak position class;、 Respectively representing the strip steel spectrum peak sequences of the x-th strip steel data point and the y-th strip steel data point in the similar peak positions; representing a strip steel peak value deviation sequence between the xth strip steel data point and the yh strip steel data point in the similar peak position class; cos () represents a cosine similarity function; MAD () represents a discrete degree function; Representing a preset value greater than 0, in order to avoid a denominator of 0, The value of (2) is preset by human, and can be set by the implementer, in this embodimentThe value is 1.
In this embodiment, the degree of dispersion of the band steel peak deviation in the band steel peak deviation sequence is an average absolute deviation, and as other embodiments, on the basis of being able to measure the degree of non-uniformity of the band steel peak deviation distribution, an implementer may measure the degree of non-uniformity of the band steel peak deviation distribution by using other existing technologies, such as variance, standard deviation, variation coefficient, etc., and the present application is not limited in particular.
It should be noted that: when two strip steel data points are formed by the same substance, such as rolling oil and carbon, although the peak value in the strip steel spectrum peak value sequence of the two strip steel data points is not at the same position possibly due to different concentrations of the substances, the peak value position is shifted integrally along with the change of the concentration, so that the cosine similarity between the two strip steel spectrum peak value sequences is achievedLarger, close to 1; the difference between the peak deviations of the strip steel at different positions of the two strip steel spectrum peak sequences is smaller, namely the degree of dispersion is smaller, so that the peak deviation degree is causedThe value of (2) is larger.
And 5, obtaining each cluster of the peak position similar class corresponding to all the peak value offset degrees, and dividing all strip steel data points in each peak position similar class into each material class by combining the difference between the peak value offset degrees in different clusters of each peak position similar class and the fluctuation degree of the peak value offset degree in each cluster.
If a strip data point is formed from a cleaning-affecting substance, the greater the degree of peak shift from the strip data point, the more likely it is for impurity substances to form. To distinguish which species a strip data point is formed from, it is necessary to classify strip data points in similar peak categories to determine the area of strip surface species.
Based on the analysis, the peak value deviation degree obtained for the same material is larger and closer, the distribution of the values has a certain concentration characteristic, and if all strip steel data points in the similar peak position are formed by the same material on the strip steel surface, all the peak value deviation degrees have the same fluctuation characteristic. Therefore, all peak deviation degrees corresponding to the similar peak position classes are used as the input of a clustering algorithm, and each cluster is output. And calculating the average value of all peak value offset degrees in each cluster, and marking the cluster with the largest average value as the largest cluster and the cluster with the smallest average value as the smallest cluster.
In this embodiment, the clustering is performed by using a K-means (K-means clustering algorithm) clustering algorithm, and as other embodiments, on the basis that the clustering of the peak offset degree can be achieved, an implementer may use other existing technologies to cluster the peak offset degree, for example, a peak clustering algorithm, a DBSCAN (sensitivity-Based Spatial Clustering of Applications with Noise) algorithm, and the like, and the present application is not limited in particular.
Combining the difference between the peak value offset degree in the maximum cluster and the minimum cluster of the similar peak value, and the fluctuation degree of the peak value offset degree in the maximum cluster and the minimum cluster, determining a substance type judgment value of the similar peak value, wherein the expression is as follows:
; wherein GF represents a substance type determination value of a substance type having a similar peak position; p, Q respectively represent the maximum cluster and the minimum cluster of similar peak positions; 、 the discrete degrees of the peak value offset degree in the maximum cluster and the minimum cluster of the similar peak position class are respectively represented; max (), min (), respectively represent maximum value taking function, minimum value taking function.
In this embodiment, the degree of dispersion of the peak deviation degree is a standard deviation, and as other embodiments, on the basis that the degree of non-uniformity of the peak deviation degree distribution can be measured, the practitioner can measure the degree of non-uniformity of the peak deviation degree distribution by using other existing technologies, for example, variance, coefficient of variation, etc., and the application is not limited in particular.
It should be noted that: when the strip steel data points in the similar peak positions are formed by the same substance on the strip steel surface, the fluctuation degree of the peak positions corresponding to all the peak value deviation degrees is smaller, namely the difference between all the peak value deviation degrees is smaller, so that the minimum value in the maximum cluster after classifying the peak value deviation degreesWith the maximum in the smallest clusterShould be less than the maximum value of the degree of fluctuation of the degree of peak shift in the two clusters; Therefore, when the material type determination value is smaller, the probability that the strip data points in the similar peak type are formed of the same material is greater.
When the material type determination value of the similar peak position class is less than or equal to 0, the strip steel data points in the similar peak position class are formed by the same material, and the similar peak position class is used as a material class; if the material type judgment value of the similar peak position class is larger than 0, the strip steel data points in the similar peak position class are formed by different materials. Because the peak value of the shift degree between different substances is smaller, two strip steel data points corresponding to each peak value shift degree in the cluster P are divided into one class, the class with at least one identical strip steel data point is merged into the same class according to the merging mode shown in fig. 3, and each class obtained after merging is recorded as one substance class.
For example, a, B, and c are strip data points of the same substance, and group A, d, e, and f are strip data points of the same substance, and group B. The degree of peak shift between the steel-in-class data points will be clustered into cluster P, as will the degree of peak shift between the steel-in-class data points in cluster P. The degree of peak offset between the steel-in-class a data points and the steel-in-class B data points will be clustered into cluster Q. After the peak value offset degrees corresponding to the class A and the class B are clustered, the class consisting of the strip steel data points corresponding to each peak value offset degree in the cluster P is ab, ac, bc, de, df and ef respectively, and two substance classes abc and def are obtained after the classes with the same strip steel data points are combined.
Step 6, determining whether the strip steel data points in any material class are formed by cleaning influencing materials based on the similarity between all strip steel data points in the material class and the spectrum data of each material at the same wavelength.
Calculating the average value of spectrum data of all strip steel data points in any material class at the same wavelength, and arranging the average value according to the order of the wavelengths from small to large to form a material waiting sequence.
And calculating cosine similarity between the substance undetermined sequence of any substance class and the spectrum sequences of various substances, and judging that the strip steel data points in any substance class are formed by the cleaning influence substances when the substance corresponding to the spectrum sequence with the largest cosine similarity is the cleaning influence substance.
And 7, determining the surface cleanliness of the strip steel based on the duty ratio of strip steel data points belonging to cleaning influencing substances in the spectrum data image.
Determining the surface cleanliness of the strip steel based on the duty ratio of strip steel data points belonging to cleaning influencing substances in the spectrum data image, wherein the expression is as follows:
; wherein B represents the surface cleanliness of the strip steel; m represents the number of band steel data points in the spectrum data image; n represents the number of strip data points belonging to the cleaning-influencing substance in the spectral data image.
Further, a cleanliness threshold is set, and whether the quality of the strip steel is qualified or not is detected by comparing the surface cleanliness of the strip steel with the cleanliness threshold, and the specific method is as follows:
when the surface cleanliness of the strip steel is larger than a cleanliness threshold value, judging that the quality of the strip steel is qualified; when the surface cleanliness of the strip steel is smaller than the cleanliness threshold value, the quality of the strip steel is judged to be unqualified, and the strip steel needs to be treated again.
In this embodiment, the value of the cleanliness threshold is 67%, the value of the cleanliness threshold is preset by human, and the operator can set the value according to the actual situation, so that the application is not particularly limited.
Based on the same inventive concept as the method, the embodiment of the application also provides a strip steel surface cleanliness detection device, which comprises:
The device comprises a spectrum data acquisition module, a data acquisition module and a data processing module, wherein the spectrum data acquisition module is used for acquiring spectrum data images of the surface of the strip steel at a plurality of preset wavelengths, wherein one pixel is recorded as a strip steel data point, and spectrum data of each substance at the plurality of preset wavelengths are respectively acquired, wherein cleaning influencing substances exist;
The spectrum data classification module is used for acquiring peaks of spectrum data corresponding to all the wavelengths of each strip steel data point; dividing all strip steel data points into strip steel type sets based on the number of the peak values of each strip steel data point;
calculating the average level of the wavelength difference between any two strip steel data points in each strip steel type set at all the peak points with the same serial numbers, and dividing the strip steel data points in each strip steel type set into similar types of peak positions based on the distribution condition of the strip steel type set corresponding to all the average levels;
determining the peak deviation degree between any two strip steel data points based on the peak difference between any two strip steel data points in any one similar class of peak positions at all the same serial numbers and the discrete degree of the peak difference;
Obtaining each cluster of the peak position similar class corresponding to all the peak value offset degrees, and dividing all strip steel data points in each peak position similar class into each material class by combining the difference between the peak value offset degrees in different clusters of each peak position similar class and the fluctuation degree of the peak value offset degree in each cluster;
A cleanliness detection module for determining whether strip steel data points in any material class are formed by cleaning-affecting substances based on similarity between all strip steel data points in the any material class and spectral data of each material at the same wavelength;
The surface cleanliness of the strip steel is determined based on the duty ratio of strip steel data points belonging to cleaning influencing substances in the spectrum data image.
Based on the same inventive concept as the above method, the embodiment of the application further provides a strip steel surface cleanliness detection system, which comprises a memory, a processor and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to realize the steps of any one of the above strip steel surface cleanliness detection methods.
In summary, the method and the device perform first classification on the band steel spectrum data points by analyzing the number of peaks in the band steel spectrum data points, perform second classification on the first classification result by utilizing the peak position difference, and classify the second classification result by combining the difference between the peaks to obtain each material class, wherein the band steel data points in each material class are formed by the same material, and further judge whether the band steel data points in each material class are formed by cleaning influencing materials or not according to the average level of the spectrum data of the band steel data points in each material class relative to the similarity of the spectrum data of each material, thereby eliminating the influence of spectral line deviation caused by vibration of a conveyor belt on the material identification result and further improving the accuracy of detecting the band steel surface cleanliness.
The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. In the description corresponding to the flowcharts and block diagrams in the figures, operations or steps corresponding to different blocks may also occur in different orders than that disclosed in the description, and sometimes no specific order exists between different operations or steps. For example, two consecutive operations or steps may actually be performed substantially in parallel, they may sometimes be performed in reverse order, which may be dependent on the functions involved. Each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
It will be evident to those skilled in the art that the application is not limited to the details of the foregoing illustrative embodiments, and that the present application may be embodied in other specific forms without departing from the essential characteristics thereof. The above-described embodiments of the application should therefore be regarded as illustrative in all respects and not restrictive.
Claims (10)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411267711.5A CN118794898B (en) | 2024-09-11 | 2024-09-11 | Strip steel surface cleanliness detection method, detection device and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411267711.5A CN118794898B (en) | 2024-09-11 | 2024-09-11 | Strip steel surface cleanliness detection method, detection device and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118794898A CN118794898A (en) | 2024-10-18 |
| CN118794898B true CN118794898B (en) | 2024-11-15 |
Family
ID=93025570
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202411267711.5A Active CN118794898B (en) | 2024-09-11 | 2024-09-11 | Strip steel surface cleanliness detection method, detection device and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118794898B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118964966B (en) * | 2024-10-21 | 2025-03-21 | 青岛海关技术中心 | Multispectral data prediction method and system based on deep learning |
| CN119643532B (en) * | 2025-02-13 | 2025-05-06 | 大连胜光科技发展有限公司 | High-precision determination method and system for aluminum content in zinc liquid component |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101151623A (en) * | 2005-01-27 | 2008-03-26 | 剑桥研究和仪器设备股份有限公司 | Classifying image features |
| CN108460342A (en) * | 2018-02-05 | 2018-08-28 | 西安电子科技大学 | Hyperspectral image classification method based on convolution net and Recognition with Recurrent Neural Network |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10898928B2 (en) * | 2018-03-27 | 2021-01-26 | Huron Valley Steel Corporation | Vision and analog sensing scrap sorting system and method |
| DE102021114351A1 (en) * | 2021-06-02 | 2022-12-08 | Carl Zeiss Microscopy Gmbh | Method and device for capturing microscopy objects in image data |
| US11579044B1 (en) * | 2021-12-02 | 2023-02-14 | Aktiebolaget Skf | System and method for bearing defect auto-detection |
| CN115311292B (en) * | 2022-10-12 | 2023-01-17 | 南通创铭伊诺机械有限公司 | Strip steel surface defect detection method and system based on image processing |
| CN116660248A (en) * | 2023-07-04 | 2023-08-29 | 国能南京煤炭质量监督检验有限公司 | Portable LIBS coal quality rapid detection device and method |
| CN117907337A (en) * | 2024-01-10 | 2024-04-19 | 陕西科技大学 | A strip steel surface cleanliness detection device and method |
| CN118583069B (en) * | 2024-08-06 | 2024-10-29 | 长沙韶光芯材科技有限公司 | Method and system for detecting uniformity of film coating thickness of glass substrate |
-
2024
- 2024-09-11 CN CN202411267711.5A patent/CN118794898B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101151623A (en) * | 2005-01-27 | 2008-03-26 | 剑桥研究和仪器设备股份有限公司 | Classifying image features |
| CN108460342A (en) * | 2018-02-05 | 2018-08-28 | 西安电子科技大学 | Hyperspectral image classification method based on convolution net and Recognition with Recurrent Neural Network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118794898A (en) | 2024-10-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN118794898B (en) | Strip steel surface cleanliness detection method, detection device and system | |
| CN116611748B (en) | Titanium alloy furniture production quality monitoring system | |
| Hu et al. | An adaptive surface filter for airborne laser scanning point clouds by means of regularization and bending energy | |
| CN110243806B (en) | Mixture component identification method based on similarity under Raman spectrum | |
| US7181365B2 (en) | Diagnostic data detection and control | |
| CN107179310B (en) | Raman spectrum characteristic peak recognition methods based on robust noise variance evaluation | |
| US20140169639A1 (en) | Image Detection Method and Device | |
| CN118134539B (en) | User behavior prediction method based on intelligent kitchen multi-source data fusion | |
| CN105913062A (en) | Big data trend curve local characteristics based matching method and terminal | |
| CN116385450A (en) | PS sheet wear resistance detection method based on image processing | |
| Han et al. | A shape–size index extraction for classification of high resolution multispectral satellite images | |
| CN104978738A (en) | Method of detection of points of interest in digital image | |
| CN116309565A (en) | High-strength conveyor belt deviation detection method based on computer vision | |
| CN117152149A (en) | Edge smoothness detection method and system for wafer dicing | |
| CN110866931A (en) | Image segmentation model training method and classification-based enhanced image segmentation method | |
| CN113702328A (en) | Method, device, equipment and storage medium for analyzing properties of product oil | |
| CN117874430A (en) | Heating underground pipe well operation environment monitoring system | |
| CN117541626A (en) | Remote sensing image change detection method based on deep learning | |
| Gruszczyński | Prediction of soil properties with machine learning models based on the spectral response of soil samples in the near infrared range | |
| CN117523324B (en) | Image processing method, image sample classification method, device and storage medium | |
| CN117541832B (en) | Abnormality detection method, abnormality detection system, electronic device, and storage medium | |
| CN116188510B (en) | Enterprise emission data acquisition system based on multiple sensors | |
| Gauthier et al. | Digitized 3D mesh segmentation based on curvature analysis | |
| CN118865267B (en) | Intelligent processing method for infusion monitoring data | |
| CN118521836B (en) | Data intelligent processing method of spectrum analysis platform |
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 |