CN114414580B - Method for identifying and deducting impurities on scrap steel - Google Patents
Method for identifying and deducting impurities on scrap steel Download PDFInfo
- Publication number
- CN114414580B CN114414580B CN202210321060.8A CN202210321060A CN114414580B CN 114414580 B CN114414580 B CN 114414580B CN 202210321060 A CN202210321060 A CN 202210321060A CN 114414580 B CN114414580 B CN 114414580B
- Authority
- CN
- China
- Prior art keywords
- impurity
- determining
- splitting
- scrap steel
- scanning
- 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 173
- 229910000831 Steel Inorganic materials 0.000 title claims abstract description 132
- 239000010959 steel Substances 0.000 title claims abstract description 132
- 238000000034 method Methods 0.000 title claims abstract description 23
- 238000010191 image analysis Methods 0.000 claims abstract description 16
- 239000011800 void material Substances 0.000 claims abstract description 5
- 238000004220 aggregation Methods 0.000 claims description 29
- 230000002776 aggregation Effects 0.000 claims description 26
- 230000002093 peripheral effect Effects 0.000 claims description 20
- 238000004458 analytical method Methods 0.000 claims description 18
- 238000009826 distribution Methods 0.000 claims description 8
- 230000000877 morphologic effect Effects 0.000 claims description 6
- 239000011148 porous material Substances 0.000 claims description 6
- 238000012216 screening Methods 0.000 claims description 6
- 238000005259 measurement Methods 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 238000003860 storage Methods 0.000 claims description 3
- 238000005303 weighing Methods 0.000 claims description 3
- 230000009286 beneficial effect Effects 0.000 description 10
- 239000002689 soil Substances 0.000 description 5
- 238000010276 construction Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000011084 recovery Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 239000003518 caustics Substances 0.000 description 1
- 230000001149 cognitive effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 229920003023 plastic Polymers 0.000 description 1
- 239000004033 plastic Substances 0.000 description 1
- 238000012797 qualification Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
- 230000004580 weight loss Effects 0.000 description 1
Images
Classifications
-
- 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
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
-
- 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
- G06T7/60—Analysis of geometric attributes
- G06T7/66—Analysis of geometric attributes of image moments or centre of gravity
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30136—Metal
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Biochemistry (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Geometry (AREA)
- Signal Processing (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a method for identifying and deducting impurities on scrap steel, which comprises the following steps: determining the current placement state of the scrap steel in the placement area; when the single placing state exists in the current placing state, the single collection and identification are carried out, and when the mixed placing state exists in the current placing state, the integral scanning and identification are carried out; carrying out hierarchical division on the single acquisition and recognition result according to a hierarchical image analysis model to construct a first impurity list; calibrating the void points in the overall scanning recognition result, carrying out deep scanning recognition to obtain an overall scanning recognition result, carrying out recognition judgment according to a scanning recognition judgment model, and constructing a second impurity list; and obtaining an impurity deduction result of the scrap steel in the placement area based on the first impurity list and the second impurity list. Through adopting different modes, the steel scraps in different states are identified, the identification accuracy is improved, the effectiveness of the obtained impurity list is further ensured, and the stability of identification and deduction determination is ensured.
Description
Technical Field
The invention relates to the technical field of image recognition, in particular to a method for recognizing and deducting impurities on scrap steel.
Background
Generally, the judgment of the scrap steel is simple manual judgment to determine whether the scrap steel meets the recovery standard or whether the scrap steel meets the recovery standard, and the corresponding deduction weight is also manually discussed according to a certain numerical value;
since these are all done by manual operation, there may be effects of cognitive differences and the like, resulting in instability in scrap identification and weight loss determination, increasing the loss cost.
Therefore, the invention provides a method for identifying and deducting impurities on scrap steel.
Disclosure of Invention
The invention provides a method for identifying and deducting impurities on scrap steel, which is used for solving the technical problems.
The invention provides a method for identifying and deducting impurities on scrap steel, which comprises the following steps:
step 1: determining the current placement state of the scrap steel in the placement area;
step 2: when the single placing state exists in the current placing state, the scrap steel in the single placing state is separately collected and identified, and when the mixed placing state exists in the current placing state, the scrap steel in the mixed placing state is integrally scanned and identified;
and step 3: carrying out hierarchical division on the separately acquired identification results according to a hierarchical image analysis model, and constructing a first impurity list of the scrap steel in a separately placed state;
Step 4, calibrating the gap points in the overall scanning recognition result, carrying out deep scanning recognition based on the gap points to obtain a complete scanning recognition result, carrying out recognition judgment on the complete scanning recognition result according to a scanning recognition judgment model, and constructing a second impurity list of the scrap steel in a mixed placement state;
and 5: and obtaining an impurity deduction result of the scrap steel in the placement area based on the first impurity list and the second impurity list.
Preferably, the determining of the placement state of the scrap steel in the placement area comprises:
shooting four directions, namely, shooting the front direction, the rear direction, the left direction and the right direction of a pre-planned placing area, and identifying a gap line of each direction image;
determining the current line characteristics of the gap line according to the recognition result;
if the current line characteristics meet preset classification constraints, determining a first position sequence corresponding to the gap line;
if not, determining a second position sequence corresponding to the gap line;
constructing a gap topology based on a first position sequence of four directions, performing connection topology assistance on the gap topology based on a second position sequence to form a complete topology, and searching for an independent gap surface;
According to the independent gap surface, performing area division on the scrap steel in the placement area, and judging whether a piece of scrap steel exists in each divided area;
if a piece of scrap steel exists, the scrap steel is regarded as being in an independent placing state;
and if a plurality of pieces of scrap steel exist, the scrap steel is regarded as a mixed placing state.
Preferably, the independent collection and identification of the scrap steel in the independent placing state comprises the following steps:
determining the placement information and the form information of the scrap steel in the independent placement state;
according to the placement information, contact points with a placement area are determined, and meanwhile, based on morphological information, the point distribution of the contact points based on the scrap steel in the independent placement state is determined;
determining an initial azimuth and a shooting order of the n1 azimuths based on the initial azimuth according to the morphological information and the point distribution;
based on the shooting sequence, the scrap steel in the single placing state is shot in n1+1 directions, and n1+1 first images are obtained.
Preferably, carry out whole scanning discernment to the steel scrap of mixing the state of putting, include:
determining a first level placing profile of the scrap steel in the mixed placing state, which is positioned on the horizontal plane, and a first central point of each level, and determining a point, which is farthest away from the corresponding first central point, of each first level placing profile;
Performing center fitting on the first central point to obtain a first fitting axis, determining a first distance between each farthest point and the first fitting axis, constructing a first aggregation scanning set obtained based on the horizontal plane according to the first distance, and screening first sub-aggregation level points in the first aggregation scanning set;
determining a second level placing profile of the scrap steel in the mixed placing state on the vertical plane and a second central point of each level, and determining a point with the closest distance between each second level placing profile and the corresponding second central point;
performing center fitting on the second central point to obtain a second fitting axis, determining a second distance between each nearest point and the second fitting axis, constructing a second aggregation scanning set obtained based on the vertical plane according to the size of the second distance, and screening second aggregation grade points in the second aggregation scanning set;
acquiring a first salient point closest to the first center from a maximum first level laying contour, constructing a first circle according to the distance between the first salient point and a corresponding first center point, acquiring a second salient point closest to a second center point from a maximum second level laying contour, and constructing a second circle according to the distance between the second salient point and a corresponding second center point;
Determining the position structures of the first circle and the second circle, acquiring a corresponding point determination rule from a structure-point rule according to the position structures, and acquiring a third central point formed by the first circle and the second circle according to the point determination rule;
determining a first expansion surface of the third central point based on a horizontal plane and a second expansion surface of the third central point based on a horizontal plane, and acquiring an intersection point of the first expansion surface and the most lateral part of the scrap steel in the mixed placement state and an intersection point of the second expansion surface and the most top part of the scrap steel in the mixed placement state;
and based on the most lateral intersection point and the most top intersection point as initial scanning positions, and according to the first aggregation grade point and the second aggregation grade point, realizing the overall scanning identification of the scrap steel in a mixed placing state.
Preferably, the hierarchy division is performed on the single acquisition identification result according to a hierarchical image analysis model, and includes:
acquiring n1+1 first images formed based on the single acquisition identification result;
based on the hierarchical image analysis model, respectively carrying out hierarchical splitting on the corresponding first images according to the shooting direction of each first image;
Determining the number of layer of hierarchy splitting, respectively measuring the splitting width of each layer, and constructing a splitting rule corresponding to the first image according to the measurement result;
and comparing the splitting rule with the splitting standard corresponding to the hierarchical image analysis model, and determining that the corresponding first image is qualified for splitting if the splitting rule is completely consistent with the splitting standard corresponding to the hierarchical image analysis model.
Preferably, if the splitting rule is inconsistent with the corresponding splitting standard, obtaining difference information between the splitting rule and the splitting standard;
determining a layer position of the splitting width corresponding to the maximum width difference value based on the difference information, if the layer position belongs to a key layer of the splitting hierarchy corresponding to the shooting azimuth, at this time, determining that the adjustability in the corresponding shooting azimuth is smaller than a preset adjustability, and when the shooting weight of the shooting azimuth is larger than the preset weight, constructing an offset adjustment function;
wherein,
representing the shooting orientation; n represents the total number of layers of unqualified splitting widths;
the unqualified splitting width of the ith layer is represented;
indicating the standard width of the ith layer; exp () represents an exponential function; f1 denotes preset adjustability; f1 denotes adjustability; f2 denotes a preset weight; f2 denotes a photographing weight; 
Representing an offset adjustment function; m represents the total number of layers of the qualified splitting width;
representing the maximum width difference in the n unqualified splitting levels;
matching corresponding offset adjustment information from an offset-attribute database according to the offset adjustment function and the azimuth attribute corresponding to the shooting azimuth;
carrying out offset adjustment on the shooting direction according to the offset adjustment information to obtain a new first image for hierarchical division;
when the shooting weight of the shooting azimuth is not more than the preset weight, acquiring a new first image again based on the corresponding shooting azimuth for hierarchical division;
and when the adjustability is not less than the preset adjustability, respectively fine-tuning the splitting width of the corresponding layer according to the difference information and the level weight of the unqualified splitting level to obtain a first image with qualified splitting.
Preferably, the first impurity list of the scrap steel in the individual arrangement state is constructed, and comprises the following steps:
constructing a splitting set based on a first image after the splitting is qualified, wherein the splitting set comprises a plurality of hierarchical images;
based on the impurity analysis model, performing identification analysis on each level image to determine the existing impurities;
dividing the types of the impurities, and sequencing the division results according to the identification purpose of the scrap steel;
And determining the layout of the impurities of the same type based on the splitting level based on the sorting result, respectively storing the layout information and the impurity information into corresponding storage units, and constructing to obtain a first impurity list.
Preferably, calibrating a gap point in the overall scanning recognition result, and performing deep scanning recognition based on the gap point to obtain a complete scanning recognition result, including:
obtaining a calibrated integral scanning identification result to construct a peripheral structure;
performing pixel analysis on the peripheral structure, performing unit slice splitting on the peripheral structure based on a pixel analysis result, and determining a peripheral edge line of each split unit slice;
determining whether the adjacent peripheral edge lines are completely overlapped, and if not, calibrating the adjacent non-overlapped lines;
respectively determining corresponding calibration areas and calibration shapes based on all independent calibration results, and further constructing a calibration list;
respectively allocating a scanning mode to each independent calibration result in the calibration list based on a calibration-scanning allocation database, and performing deep scanning identification in the pore points corresponding to the independent calibration results according to the scanning mode;
and obtaining a complete scanning identification result based on the depth scanning identification result and the integral scanning identification result.
Preferably, according to the scanning identification judgment model, the complete scanning identification result is subjected to identification judgment, and a second impurity list of the scrap steel in a mixed placing state is constructed, including:
constructing and obtaining an integral three-dimensional structure based on the complete scanning recognition result, and carrying out substructure splitting on the integral three-dimensional structure based on a gap line formed by gap points;
and judging and analyzing each split substructure according to the scanning recognition judgment model and the three-dimensional judgment model, determining impurity information of each single substructure, and constructing to obtain a second impurity list.
Preferably, the obtaining of the impurity deduction weight result of the scrap steel in the placement area based on the first impurity list and the second impurity list comprises:
determining a first layout of the same type of impurities and impurity information of the same type of impurities based on the first impurity list, and determining an impurity deduction weight G1 of the same type of impurities;
wherein M1 represents the current weight of the scrap in the individual placement state determined based on the impurity information; m2 represents the initial weight of the scrap steel in the determined individual placement state;
a first layout representing impurities of the same type based on all impurities; 
Denotes all types of impurities
The second layout of (1);
is expressed for the same type of impuritiesAn error discrimination factor of the layout, and the value range is [0,0.2 ]]；
Determining each individual substructure and impurity information based on the second list of impurities, thereby determining an impurity weight G2 for the individual substructure;
wherein Z represents the total number of individual substructures;
representing the density of the z1 th impurity on the corresponding individual substructure determined based on the impurity information;
representing a volume of z1 th impurity on the corresponding individual substructure determined based on the impurity information;
a weight of influence representing additional variation of z1 th impurity on the corresponding individual substructure determined based on the impurity information based on the adjacent individual substructure;
and obtaining an impurity deduction result of the scrap steel in the placement area based on all the impurity deduction weights G1 and G2, and outputting and displaying the result.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by the practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and drawings.
The technical solution of the present invention is further described in detail by the accompanying drawings and embodiments.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings:
FIG. 1 is a flow chart of a method for identifying and weighing impurities on scrap steel according to an embodiment of the present invention;
FIG. 2 is a block diagram of a placement area in an embodiment of the present invention;
FIG. 3 is a block diagram of a hierarchy and points according to an embodiment of the present invention.
Detailed Description
The preferred embodiments of the present invention will be described in conjunction with the accompanying drawings, and it will be understood that they are described herein for the purpose of illustration and explanation and not limitation.
Example 1:
the invention provides a method for identifying and deducting impurities on scrap steel, which comprises the following steps of:
step 1: determining the current placement state of the scrap steel in the placement area;
step 2: when the single placing state exists in the current placing state, the scrap steel in the single placing state is separately collected and identified, and when the mixed placing state exists in the current placing state, the scrap steel in the mixed placing state is integrally scanned and identified;
And 3, step 3: carrying out hierarchical division on the separately acquired and identified results according to a hierarchical image analysis model, and constructing a first impurity list of the scrap steel in a separately placed state;
step 4, calibrating the void points in the overall scanning recognition result, performing deep scanning recognition based on the void points to obtain a complete scanning recognition result, performing recognition determination on the complete scanning recognition result according to a scanning recognition determination model, and constructing a second impurity list of the scrap steel in a mixed placement state;
and 5: and obtaining an impurity deduction weight result of the scrap steel in the placement area based on the first impurity list and the second impurity list.
In this embodiment, as shown in fig. 2, in the placement area a, there are scrap steel 1 in an individual placement state and scrap steel 2 in a mixed placement state, at this time, the scrap steel 1 is separately identified and collected, the scrap steel 2 is integrally scanned and identified, and the individual placement state can be regarded as only one piece of scrap steel, and the mixed placement state means that a plurality of pieces of scrap steel are placed next to each other, and there is a contact portion.
In this embodiment, the single collection and identification is to photograph the scrap steel, and to photograph only one scrap steel in a single placement state in a plurality of directions, so as to obtain the identification result of the scrap steel, and the identification result is obtained based on the two-dimensional image, and the integral scanning and identification is to obtain the external three-dimensional structure of the scrap steel in the mixed placement state.
In this embodiment, the hierarchical image analysis model is trained in advance, and is obtained by performing hierarchical splitting on scrap steel of various shapes as a sample, for example, the scrap steel is square, and one face of the scrap steel is split into 5 parts in the transverse direction, and the width of each part may be different.
In this embodiment, the first impurity list is obtained information on impurities on different surfaces of the scrap steel in a single placement state, and the first impurity list is further constructed, and the first impurity and the second impurity may be any factors that are not related to the steel itself but affect the steel itself, such as soil, water, plastics, and corrosive substances on the scrap steel.
In this embodiment, the construction of the scrap steel corresponding to the three-dimensional structure is structured, and a non-contact area 3 is inevitably present in the structure, as shown in fig. 2, the area 3 can be regarded as a pore point.
In this embodiment, the scan recognition determination model is trained in advance, and is obtained by training using various scan results as samples, and the scan results are related to the results of the steel scrap itself and the impurities on the steel scrap.
In this embodiment, the second impurity list includes a three-dimensional structure in a mixed-placement state and impurity information included in the structure.
In this example, the result of the impurity deduction is the total weight of the impurities on the scrap in the deposit area, which weight is to be removed in the process of determining the steel quality of the scrap itself.
The beneficial effects of the above technical scheme are: through adopting different modes, the steel scraps in different states are identified, the identification accuracy is improved, the effectiveness of the obtained impurity list is further ensured, and the stability of identification and deduction determination is ensured.
Example 2:
based on embodiment 1, the method for determining the placement state of the scrap steel in the placement area comprises the following steps:
shooting four directions, namely, shooting the front direction, the rear direction, the left direction and the right direction of a pre-planned placing area, and identifying a gap line of each direction image;
determining the current line characteristics of the gap line according to the recognition result;
if the current line characteristics meet preset classification constraints, determining a first position sequence corresponding to the gap line;
if not, determining a second position sequence corresponding to the gap line;
constructing a gap topology based on a first position sequence of four directions, performing connection topology assistance on the gap topology based on a second position sequence to form a complete topology, and searching for an independent gap surface;
According to the independent gap surface, performing region division on the scrap steel in the placement region, and judging whether a piece of scrap steel exists in each divided region;
if a piece of scrap steel exists, the scrap steel is regarded as being in an independent placing state;
and if a plurality of pieces of scrap steel exist, the scrap steel is regarded as a mixed placing state.
In this embodiment, before determining the individual placement state and the mixed placement state, it is necessary to determine the states that may exist in the placement area, and therefore, the situation of the current placement area is determined by photographing from four orientations.
In this embodiment, a plurality of gap lines may exist in each position, and the gap lines refer to spatial position lines formed by different scrap placement positions, for example, lines of positions where scrap does not contact in a space, and whether a preset classification constraint is satisfied is determined by determining a current line feature, such as a position feature or a spatial feature, of the spatial position lines, where the preset classification constraint refers to whether a corresponding gap line is a broken line, that is, whether the gap line can penetrate from top to bottom in an image acquired from the position, if so, it is determined that the preset classification constraint is satisfied, and then a first position sequence is determined, otherwise, it is determined that the preset classification constraint is not satisfied, and then a second position sequence is determined.
In the embodiment, based on the lines which can be completely penetrated and the line positions, a gap topology is constructed, and the lines corresponding to the second position sequence are used for assisting to obtain a complete topology, wherein the topology is of a three-dimensional structure, whether surfaces which can completely and independently open scrap steel at multiple positions exist is determined through the topology, namely independent gap surfaces exist, and if the surfaces exist, region division is performed according to the independent gap surfaces, so that the number of the scrap steel in each region is determined.
The beneficial effects of the above technical scheme are: the method comprises the steps of preliminarily obtaining gap lines from multiple directions to determine line characteristics, classifying lines according to preset classification constraints, constructing gap topology and complete topology, subsequently carrying out region division by searching for independent gap surfaces, further obtaining regions corresponding to independent placing states and mixed placing states, and providing a basis for subsequently adopting different modes to carry out refined identification.
Example 3:
based on embodiment 1, carry out the discernment of separately gathering to the scrap steel of state of putting alone, include:
determining the placement information and the form information of the scrap steel in the independent placement state;
according to the placement information, contact points with a placement area are determined, and meanwhile, based on morphological information, the point distribution of the contact points based on the scrap steel in the single placement state is determined;
Determining an initial azimuth and a shooting order of the n1 azimuths based on the initial azimuth according to the morphological information and the point distribution;
based on the shooting sequence, the scrap steel in the single placing state is shot in n1+1 directions, and n1+1 first images are obtained.
In this embodiment, the placement information refers to a placement position, and the shape information refers to a contact state with a placement area and a shape of the scrap itself.
In this embodiment, the contact point refers to a contact point between the scrap and the area, and the distribution of the points is determined according to the form of the scrap, for example, the contact point between the lower right side of the scrap and the area is too many, at this time, the initial orientation is set to be the lower right side, and then the remaining n1 orientations are determined sequentially according to the instantaneous direction and the rotation of the preset angle, so as to acquire the first image in the n1+1 orientation.
In this embodiment, the first image refers to an image of the scrap itself.
The beneficial effects of the above technical scheme are: based on the placement information and the form information, the initial direction can be conveniently and effectively determined, the shooting sequence is further obtained, a plurality of first images are obtained, the effectiveness of obtaining the images through shooting is guaranteed, and a basis is provided for subsequent identification.
Example 4:
based on embodiment 1, carry out whole scanning discernment to the steel scrap of mixing the state of putting, include:
determining a first level placing profile of the scrap steel in the mixed placing state on the horizontal plane and a first central point of each level, and determining a point of each first level placing profile which is farthest away from the corresponding first central point;
performing center fitting on the first central point to obtain a first fitting axis, determining a first distance between each farthest point and the first fitting axis, constructing a first aggregation scanning set obtained based on the horizontal plane according to the size of the first distance, and screening first sub-aggregation grade points in the first aggregation scanning set;
determining a second level placing contour of the scrap steel in the mixed placing state on the vertical plane and a second central point of each level, and determining a point of each second level placing contour, which is closest to the corresponding second central point;
performing center fitting on the second central point to obtain a second fitting axis, determining a second distance between each nearest point and the second fitting axis, constructing a second aggregation scanning set obtained based on the vertical plane according to the size of the second distance, and screening second aggregation grade points in the second aggregation scanning set;
Acquiring a first protruding point closest to the first center from a maximum first-level placing contour, constructing a first circle according to the distance between the first protruding point and a corresponding first center point, acquiring a second protruding point closest to a second center point from a maximum second-level placing contour, and constructing a second circle according to the distance between the second protruding point and a corresponding second center point;
determining the position structures of the first circle and the second circle, acquiring a corresponding point determination rule from a structure-point rule according to the position structures, and acquiring a third central point formed by the first circle and the second circle according to the point determination rule;
determining a first expansion surface of the third central point based on a horizontal plane and a second expansion surface of the third central point based on a horizontal plane, and acquiring an intersection point of the first expansion surface and the most lateral part of the scrap steel in the mixed placement state and an intersection point of the second expansion surface and the most top part of the scrap steel in the mixed placement state;
and based on the most lateral intersection point and the most top intersection point as initial scanning positions, and according to the first aggregation grade point and the second aggregation grade point, realizing the overall scanning identification of the scrap steel in a mixed placing state.
In this embodiment, as shown in fig. 3, for example, each first level is obtained by plotting points on a vertical line perpendicular to the horizontal plane one by one, so as to obtain a plurality of first levels, and each first level is an outermost contour of the acquired scrap steel in a mixed state on the corresponding horizontal plane, so as to determine a first center point of the outermost contour, and determine a farthest point of the contour through the contour and the points.
For example, the first level is a1, the corresponding vertical line is a, the corresponding outermost contour is a2, the first center point of the corresponding outermost contour is A3, the corresponding farthest point is a4, and the first protruding point is a 5.
In this embodiment, since the contour shapes are different, the corresponding first central points may not be on a straight line, so that by fitting, a first fitted curve may be determined, and by determining a first distance between the farthest point and the first fitted axis, where the first distance refers to an intersection point of a horizontal plane where the farthest point is located and the first fitted axis, and further determining a distance between the intersection point and the farthest point, and regarding the distance as the first distance.
In this embodiment, the determination of the first distance is based on the sequence of the scrap in the mixed placement state, and the sequence is, for example: 1122212222444, therefore, one point can be obtained from "11", "222", "1", "2222", and "444", respectively, as an aggregation level point.
In this embodiment, the principle of the second-level placement and the work content of the vertical plane is similar to that of the first-level placement and the work content of the horizontal plane, and the description thereof is omitted here.
In this embodiment, for example, the position results in that there is an intersection portion between the first circle and the second circle, and in this case, the obtained point determination rule may be that the right center of the intersection portion is regarded as the third center point.
In this embodiment, the initial scanning position is the intersection point described above, the initial scanning position is used as an initial scanning point, and the overall scanning identification is performed sequentially according to the first partial aggregation level point and the second partial aggregation level point according to the up-down scanning order or the left-right scanning order.
In this embodiment, the closer the distance corresponding to the first distance is, the fewer the corresponding aggregation results are, the fewer the corresponding acquired grade points are, and vice versa. The method mainly comprises the step of effectively analyzing corresponding points according to the aggregation analysis of the distance size to obtain a more complete scanning result.
The beneficial effects of the above technical scheme are: through horizontal plane and perpendicular, confirm different level's profile and central point, and then through the fit, size comparison, obtain the diversity grade point, and through confirming the salient point, establish the circle of horizontal plane and perpendicular, and then confirm the rule according to position result and point, confirm the third central point, and then through face extension, obtain initial scanning position, the combination with the above point at last, realize whole scanning, guarantee regular scanning, obtain the integrality of data, avoid random scanning, the scanning data that leads to acquireing is incomplete, acquire for follow-up impurity information, effective basis is provided.
Example 5:
based on the embodiment 1, the method carries out hierarchical division on the single acquisition and recognition result according to a hierarchical image analysis model, and comprises the following steps:
acquiring n1+1 first images formed based on the single acquisition identification result;
based on the hierarchical image analysis model, carrying out hierarchical splitting on the corresponding first images according to the shooting orientation of each first image;
determining the number of layer of hierarchy splitting, respectively measuring the splitting width of each layer, and constructing a splitting rule corresponding to the first image according to the measurement result;
and comparing the splitting rule with the splitting standard corresponding to the hierarchical image analysis model, and determining that the corresponding first image is qualified for splitting if the splitting rule is completely consistent with the splitting standard corresponding to the hierarchical image analysis model.
In this embodiment, the measured splitting width is based on the width after the actual splitting of the model, and the splitting standard of the splitting rule is the width after the splitting according to the model standard.
The beneficial effects of the above technical scheme are: by comparing the actual image with the standard image, whether the image is qualified or not can be effectively determined, and an effective acquisition basis is provided for the follow-up acquisition of impurity information.
Example 6:
based on embodiment 5, if the splitting rule is inconsistent with the corresponding splitting standard, obtaining difference information between the splitting rule and the splitting standard;
Determining a layer position of the splitting width corresponding to the maximum width difference value based on the difference information, if the layer position belongs to a key layer of the splitting hierarchy corresponding to the shooting azimuth, at this time, determining that the adjustability in the corresponding shooting azimuth is smaller than a preset adjustability, and when the shooting weight of the shooting azimuth is larger than the preset weight, constructing an offset adjustment function;
wherein,
representing the shooting orientation; n represents the total number of layers of unqualified splitting widths;
the splitting width of the i-th layer which is unqualified is represented;
indicating the standard width of the ith layer; exp () represents an exponential function; f1 denotes preset adjustability; f1 denotes adjustability; f2 denotes a preset weight; f2 denotes a photographing weight;
representing an offset adjustment function; m represents the total number of layers of the qualified splitting width;
representing the maximum width difference in the n unqualified splitting levels;
matching corresponding offset adjustment information from an offset-attribute database according to the offset adjustment function and the azimuth attribute corresponding to the shooting azimuth;
carrying out offset adjustment on the shooting direction according to the offset adjustment information to obtain a new first image for hierarchical division;
when the shooting weight of the shooting azimuth is not more than the preset weight, acquiring a new first image again based on the corresponding shooting azimuth for hierarchical division;
And when the adjustability is not less than the preset adjustability, respectively fine-tuning the splitting width of the corresponding layer according to the difference information and the level weight of the unqualified splitting level to obtain a first image with qualified splitting.
In this embodiment, when there is a difference, for example, the maximum width difference is 2, and the corresponding layer is set as the position 1, and at this time, the position 1 is the key layer, so at this time, the adjustability is considered to be smaller than the preset adjustability, and the weight of the azimuth shooting at this time is greater than the preset weight, so as to construct a function, and by using this function, the offset adjustment information can be effectively determined, and the offset adjustment of the azimuth is realized, for example, the initial azimuth is 45 degrees at the top left, and is 43 degrees at the top left after the adjustment.
In this embodiment, when the key layer is located at position 1, a qualified image can be obtained only by performing fine adjustment on the layer width of the unqualified split level.
In this embodiment, the orientation attribute is, for example, direction information, and the offset adjustment information is obtained by the direction information and an offset function.
In this embodiment, the number of n and m is both greater than 1, and n + m is greater than 10.
In this embodiment, the preset adjustability and the preset weight are preset, for example, the preset weight is 0.5, and the shooting weight is less than 1. For example, if the corresponding adjustability is 0 and the corresponding preset adjustability is 0.5, the corresponding key layer is determined to be the key layer.
In this embodiment, the offset-attribute database includes offset information and corresponding attribute information, and is constructed in advance.
The beneficial effects of the above technical scheme are: whether an offset adjustment function is constructed or not can be determined through analysis of difference information and analysis of layer positions, when construction is carried out, offset adjustment information is obtained through the function, initial directions are adjusted, when construction is not needed, shooting is determined again or width is micro-adjusted according to comparison of weight sizes, accordingly, qualification of images is guaranteed, a basis is provided for obtaining impurities, and identification stability and stability of impurity deduction and obtaining are guaranteed.
Example 7:
based on the embodiment 1, a first impurity list of the scrap steel in an individual placing state is constructed, and the first impurity list comprises the following components:
constructing a splitting set based on a first image after the splitting is qualified, wherein the splitting set comprises a plurality of hierarchical images;
based on the impurity analysis model, performing identification analysis on each level image to determine the existing impurities;
dividing the types of the impurities, and sequencing the division results according to the identification purpose of the scrap steel;
And determining the layout of the impurities of the same type based on the splitting level based on the sorting result, respectively storing the layout information and the impurity information into corresponding storage units, and constructing to obtain a first impurity list.
In this embodiment, the impurity analysis model is trained in advance, and the training samples are various impurity images, so that impurities of images of different levels can be determined by performing impurity identification through the model, and the integrity of identification can be ensured by hierarchical division.
In this embodiment, for identification purposes, for example, it is determined that soil exists on the scrap steel, at this time, the division results are sorted, the levels where soil exists may be regarded as one type, and the remaining levels are regarded as one type, at this time, a layout of soil based on split levels is obtained, for example, there are levels 1, 2, 3, and 4, at this time, levels 1 and 2 are occupied, and the distribution of soil on corresponding levels 1 and 2 is randomly distributed, so that impurity information is obtained and stored.
The beneficial effects of the above technical scheme are: through carrying out hierarchical image recognition, guarantee the integrality of discernment, and through impurity type division, can confirm the impurity type of paying attention to, make things convenient for follow-up reasonable analysis to this type impurity.
Example 8:
based on the embodiment 1, calibrating the gap points in the overall scanning recognition result, and performing deep scanning recognition based on the gap points to obtain a complete scanning recognition result, including:
obtaining a calibrated integral scanning identification result to construct a peripheral structure;
performing pixel analysis on the peripheral structure, performing unit piece splitting on the peripheral structure based on a pixel analysis result, and determining a peripheral edge line of each split unit piece;
determining whether the adjacent peripheral edge lines are completely overlapped, if not, calibrating the adjacent non-overlapped lines;
respectively determining corresponding calibration areas and calibration shapes based on all independent calibration results, and further constructing a calibration list;
respectively allocating a scanning mode to each independent calibration result in the calibration list based on a calibration-scanning allocation database, and performing deep scanning identification in the pore points corresponding to the independent calibration results according to the scanning mode;
and obtaining a complete scanning identification result based on the depth scanning identification result and the integral scanning identification result.
In this embodiment, the peripheral structure is a three-dimensional structure of the constructed appearance, and through pixel analysis, a continuous region and a discontinuous region existing on the appearance structure may be determined, and then the unit pieces are split according to the continuous region and the discontinuous region, so as to obtain a peripheral edge line of adjacent regions.
In this embodiment, the calibration is performed by determining whether the peripheral edge lines are completely overlapped, and the calibration list is determined by determining the calibration area and the shape.
In the embodiment, the scanning mode can be effectively distributed to the pore points through the area and the shape to carry out deep scanning;
in this embodiment, the larger the area, the simpler the shape, and the simpler the scanning manner of the corresponding assignment.
The beneficial effects of the above technical scheme are: whether the peripheral edge lines are completely overlapped or not is determined by obtaining the peripheral edge lines, a calibration list is built, and the pore points are deeply scanned by matching of scanning modes, so that the completeness of the obtained scanning result is ensured, and the stability of deduction and acquisition is ensured.
Example 9:
on the basis of the embodiment 8, according to the scanning recognition judgment model, the complete scanning recognition result is subjected to recognition judgment, and a second impurity list of the scrap steel in a mixed placing state is constructed, including:
constructing and obtaining an integral three-dimensional structure based on the complete scanning recognition result, and performing substructure splitting on the integral three-dimensional structure based on a gap line formed by gap points;
and judging and analyzing each split substructure according to the scanning identification judgment model and the three-dimensional judgment model, determining impurity information of each single substructure, and constructing to obtain a second impurity list.
The beneficial effects of the above technical scheme are: and splitting the substructure of the three-dimensional structure to obtain the impurity information of the single substructure so as to obtain an impurity list.
Example 10:
based on the embodiment 1, the method for obtaining the impurity deduction weight of the scrap steel in the placement area based on the first impurity list and the second impurity list comprises the following steps:
determining a first layout of the impurities of the same type and impurity information of the impurities of the same type based on the first impurity list, and determining an impurity deduction G1 of the impurities of the same type;
wherein M1 represents the current weight of the scrap in the individual placement state determined based on the impurity information; m2 represents the initial weight of the scrap steel in the determined individual placement state;
a first layout representing impurities of the same type based on all impurities;
denotes all types of impurities
The second layout of (1);
the error discrimination factor of the first layout of the same type of impurities is represented, and the value range is [0,0.2 ]]；
Determining each individual substructure and impurity information based on the second list of impurities, thereby determining an impurity weight G2 for the individual substructure;
wherein Z represents the total number of individual substructures;
representing the density of the z1 th impurity on the corresponding individual substructure determined based on the impurity information; 
Representing a volume of z1 th impurity on the corresponding individual substructure determined based on the impurity information;
a weight of influence representing additional variation of z1 th impurity on the corresponding individual substructure determined based on the impurity information based on the adjacent individual substructure;
and obtaining an impurity deduction result of the scrap steel in the placement area based on all the impurity deduction weights G1 and G2, and outputting and displaying the result.
In this embodiment of the present invention,
the first layout is based on the occupation ratio of the second layout, and the value is less than 1.
Are densities determined in advance based on the type of impurity.
In this embodiment, the additional variation affecting the weight means that, for example, when determining an individual substructure, a part of the impurities belonging to a substructure 1 is planned onto an adjacent substructure 2, but this part of the impurities is relatively less heavy and this is relatively likely to occur because, for example, the provision of a separate substructure
The value tends to be 0.
The beneficial effects of the above technical scheme are: through obtaining the impurity knot weight of putting the independent substructure of state alone and putting the impurity knot weight of the independent substructure of state alone of the same type in the state alone, can effectively obtain this total impurity knot weight of placing the region, guarantee the accuracy of acquireing, avoid artificial estimation, the irrationality of bringing.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (7)
1. A method for identifying and deducting impurities on scrap steel is characterized by comprising the following steps:
step 1: determining the current placement state of the scrap steel in the placement area;
step 2: when the single placing state exists in the current placing state, the scrap steel in the single placing state is separately collected and identified, and when the mixed placing state exists in the current placing state, the scrap steel in the mixed placing state is integrally scanned and identified;
and step 3: carrying out hierarchical division on the separately acquired identification results according to a hierarchical image analysis model, and constructing a first impurity list of the scrap steel in a separately placed state;
and 4, step 4: calibrating the gap points in the overall scanning recognition result, carrying out deep scanning recognition based on the gap points to obtain a complete scanning recognition result, carrying out recognition judgment on the complete scanning recognition result according to a scanning recognition judgment model, and constructing a second impurity list of the scrap steel in a mixed placing state;
And 5: obtaining an impurity deduction weight result of the scrap steel in the placement area based on the first impurity list and the second impurity list;
wherein, carry out the discernment of gathering alone to the scrap steel of putting the state alone, include:
determining the placement information and the form information of the scrap steel in the independent placement state;
according to the placement information, contact points with a placement area are determined, and meanwhile, based on morphological information, the point distribution of the contact points based on the scrap steel in the independent placement state is determined;
determining an initial azimuth and a shooting sequence of n1 azimuths based on the initial azimuth according to the morphological information and the point distribution;
based on the shooting sequence, carrying out n1+1 azimuth shooting on the scrap steel in the single placement state, and obtaining n1+1 first images;
wherein, carry out whole scanning discernment to the scrap steel of mixing the state of putting, include:
determining a first level placing profile of the scrap steel in the mixed placing state, which is positioned on the horizontal plane, and a first central point of each level, and determining a point, which is farthest away from the corresponding first central point, of each first level placing profile;
performing center fitting on the first central point to obtain a first fitting axis, determining a first distance between each farthest point and the first fitting axis, constructing a first aggregation scanning set obtained based on the horizontal plane according to the first distance, and screening first sub-aggregation level points in the first aggregation scanning set;
Determining a second level placing profile of the scrap steel in the mixed placing state on the vertical plane and a second central point of each level, and determining a point with the closest distance between each second level placing profile and the corresponding second central point;
performing center fitting on the second central point to obtain a second fitting axis, determining a second distance between each nearest point and the second fitting axis, constructing a second aggregation scanning set obtained based on the vertical plane according to the size of the second distance, and screening second aggregation grade points in the second aggregation scanning set;
acquiring a first salient point closest to the first central point from a maximum first level laying contour, constructing a first circle according to the distance between the first salient point and the corresponding first central point, acquiring a second salient point closest to the second central point from a maximum second level laying contour, and constructing a second circle according to the distance between the second salient point and the corresponding second central point;
determining the position structures of the first circle and the second circle, acquiring a corresponding point determination rule from a structure-point rule according to the position structures, and acquiring a third central point formed by the first circle and the second circle according to the point determination rule;
Determining a first expansion surface of which the third central point is based on a horizontal plane and a second expansion surface of which the third central point is based on a vertical plane, and acquiring an intersection point of the first expansion surface and the most side part of the scrap steel in the mixed putting state and an intersection point of the second expansion surface and the most top part of the scrap steel in the mixed putting state;
based on the most lateral intersection point and the most top intersection point as initial scanning positions, and according to the first aggregation grade point and the second aggregation grade point, realizing integral scanning identification of the scrap steel in a mixed placing state;
obtaining an impurity deduction weight result of the scrap steel in the placement area based on the first impurity list and the second impurity list, wherein the method comprises the following steps:
determining a first layout of the same type of impurities and impurity information of the same type of impurities based on the first impurity list, and determining an impurity deduction weight G1 of the same type of impurities;
wherein M1 represents the current weight of the scrap in the individual placement state determined based on the impurity information; m2 represents the initial weight of the scrap in the determined individual placement state;
a first layout representing impurities of the same type based on all impurities;
denotes all types of impurities 
The second layout of (1);
the error discrimination factor of the first layout of the same type of impurities is represented, and the value range is [0,0.2 ]]；
Determining each individual substructure and impurity information based on the second list of impurities, thereby determining an impurity weight G2 for the individual substructure;
wherein Z represents the total number of individual substructures;
representing the density of the z1 th impurity on the corresponding individual substructure determined based on the impurity information;
representing a volume of z1 th impurity on the corresponding individual substructure determined based on the impurity information;
a weight of influence representing additional variation of z1 th impurity on the corresponding individual substructure determined based on the impurity information based on the adjacent individual substructure;
and obtaining the impurity deduction result of the scrap steel in the placement area based on all the impurity deduction weights G1 and the impurity deduction weight G2, and outputting and displaying the result.
2. The method of claim 1, wherein determining the current placement state of the scrap in the placement area comprises:
shooting four directions, namely, shooting the front direction, the rear direction, the left direction and the right direction of a pre-planned placing area, and identifying a gap line of each direction image;
determining the current line characteristics of the gap line according to the recognition result;
If the current line characteristics meet preset classification constraints, determining a first position sequence corresponding to the gap line;
if not, determining a second position sequence corresponding to the gap line;
constructing a gap topology based on a first position sequence of four directions, performing connection topology assistance on the gap topology based on a second position sequence to form a complete topology, and searching for an independent gap surface;
according to the independent gap surface, performing area division on the scrap steel in the placement area, and judging whether a piece of scrap steel exists in each divided area;
if a piece of scrap steel exists, the scrap steel is regarded as being in an independent placing state;
and if a plurality of pieces of scrap steel exist, the scrap steel is regarded as a mixed placing state.
3. The method for identifying and deducting impurities on steel scrap according to claim 1, wherein the step of hierarchically dividing the individually acquired identification results according to a hierarchical image analysis model comprises the steps of:
acquiring n1+1 first images formed based on the single acquisition identification result;
based on the hierarchical image analysis model, respectively carrying out hierarchical splitting on the corresponding first images according to the shooting direction of each first image;
determining the number of layers of hierarchy splitting, respectively measuring the splitting width of each layer, and constructing a splitting rule corresponding to the first image according to the measurement result;
And comparing the splitting rule with the splitting standard corresponding to the hierarchical image analysis model, and if the splitting rule is completely consistent with the splitting standard, determining that the corresponding first image is qualified for splitting.
4. The method of claim 3, wherein the identification and deduction of impurities from the scrap steel is performed by using a counter,
if the splitting rule is inconsistent with the corresponding splitting standard, acquiring the difference information of the splitting rule and the splitting standard;
determining a layer position of the splitting width corresponding to the maximum width difference value based on the difference information, if the layer position belongs to a key layer of the splitting hierarchy corresponding to the shooting azimuth, at this time, determining that the adjustability in the corresponding shooting azimuth is smaller than a preset adjustability, and when the shooting weight of the shooting azimuth is larger than the preset weight, constructing an offset adjustment function;
wherein,
representing the shooting orientation; n represents the total number of layers of unqualified splitting widths;
the splitting width of the i-th layer which is unqualified is represented;
indicating the standard width of the ith layer; exp () represents an exponential functionCounting; f1 denotes preset adjustability; f1 denotes adjustability; f2 denotes a preset weight; f2 denotes a photographing weight;
representing an offset adjustment function; m represents the total number of layers of the qualified splitting width; 
Representing the maximum width difference in the n unqualified splitting levels;
matching corresponding offset adjustment information from an offset-attribute database according to the offset adjustment function and the azimuth attribute corresponding to the shooting azimuth;
carrying out offset adjustment on the shooting direction according to the offset adjustment information to obtain a new first image for hierarchical division;
when the shooting weight of the shooting orientation is not more than the preset weight, acquiring a new first image again based on the corresponding shooting orientation for hierarchical division;
and when the adjustability is not less than the preset adjustability, respectively fine-tuning the splitting width of the corresponding layer according to the difference information and the level weight of the unqualified splitting level to obtain a first image with qualified splitting.
5. The method for identifying and weighing impurities on steel scraps according to claim 3, wherein the step of constructing a first impurity list of the steel scraps in an individually placed state comprises the steps of:
constructing a splitting set based on a first image after the splitting is qualified, wherein the splitting set comprises a plurality of hierarchical images;
based on the impurity analysis model, performing identification analysis on each level image to determine the existing impurities;
dividing the types of the impurities, and sequencing the division results according to the identification purpose of the scrap steel;
And determining the layout of the impurities of the same type based on the splitting level based on the sorting result, respectively storing the layout information and the impurity information into corresponding storage units, and constructing to obtain a first impurity list.
6. The method for identifying and weighing impurities on the scrap steel according to claim 1, wherein the steps of calibrating the void points in the overall scanning identification result and performing deep scanning identification based on the void points to obtain the complete scanning identification result comprise:
obtaining a calibrated integral scanning identification result to construct a peripheral structure;
performing pixel analysis on the peripheral structure, performing unit piece splitting on the peripheral structure based on a pixel analysis result, and determining a peripheral edge line of each split unit piece;
determining whether the adjacent peripheral edge lines are completely overlapped, and if not, calibrating the adjacent non-overlapped lines;
respectively determining corresponding calibration areas and calibration shapes based on all independent calibration results, and further constructing a calibration list;
respectively allocating a scanning mode to each independent calibration result in the calibration list based on a calibration-scanning allocation database, and carrying out deep scanning identification on the pore points in the corresponding independent calibration results according to the scanning mode;
And obtaining a complete scanning identification result based on the depth scanning identification result and the integral scanning identification result.
7. The method for identifying and deducting impurities on steel scraps according to claim 6, wherein the step of identifying and judging the complete scanning identification result according to a scanning identification judgment model to construct a second impurity list of the steel scraps in a mixed placing state comprises the following steps:
constructing and obtaining an integral three-dimensional structure based on the complete scanning recognition result, and performing substructure splitting on the integral three-dimensional structure based on a gap line formed by gap points;
and judging and analyzing each split substructure according to the scanning identification judgment model and the three-dimensional judgment model, determining impurity information of each single substructure, and constructing to obtain a second impurity list.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210321060.8A CN114414580B (en) | 2022-03-30 | 2022-03-30 | Method for identifying and deducting impurities on scrap steel |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210321060.8A CN114414580B (en) | 2022-03-30 | 2022-03-30 | Method for identifying and deducting impurities on scrap steel |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114414580A CN114414580A (en) | 2022-04-29 |
| CN114414580B true CN114414580B (en) | 2022-06-14 |
Family
ID=81263124
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210321060.8A Active CN114414580B (en) | 2022-03-30 | 2022-03-30 | Method for identifying and deducting impurities on scrap steel |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114414580B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114998318B (en) * | 2022-07-18 | 2022-10-25 | 聊城一明五金科技有限公司 | Scrap steel grade identification method used in scrap steel treatment process |
Family Cites Families (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5705818A (en) * | 1996-02-29 | 1998-01-06 | Bethlehem Steel Corporation | Method and apparatus for detecting radioactive contamination in steel scrap |
| KR100245367B1 (en) * | 1996-12-20 | 2000-03-02 | 구마모토 마사히로 | Method for re-utilizing scrap steels |
| CN109919243A (en) * | 2019-03-15 | 2019-06-21 | 天津拾起卖科技有限公司 | A kind of scrap iron and steel type automatic identifying method and device based on CNN |
| CN111292026A (en) * | 2020-04-27 | 2020-06-16 | 江苏金恒信息科技股份有限公司 | Scrap steel grading method and device based on neural network model fusion |
| CN112348791B (en) * | 2020-11-04 | 2023-03-14 | 中冶赛迪信息技术(重庆)有限公司 | Intelligent scrap steel detecting and judging method, system, medium and terminal based on machine vision |
| CN113743210A (en) * | 2021-07-30 | 2021-12-03 | 阿里巴巴达摩院(杭州)科技有限公司 | Image recognition method and scrap grade recognition method |
| CN215466225U (en) * | 2021-08-13 | 2022-01-11 | 山西晋南钢铁集团有限公司 | Intelligent grading device for scrap steel |
| CN113627830A (en) * | 2021-09-08 | 2021-11-09 | 西安智者云集云计算有限公司 | Scrap steel grading system and method |
| CN114037987A (en) * | 2021-11-05 | 2022-02-11 | 中冶赛迪上海工程技术有限公司 | Intelligent identification method, device, medium and equipment for scrap steel |
| CN114078126B (en) * | 2022-01-19 | 2022-04-26 | 江苏金恒信息科技股份有限公司 | Scrap steel grading method and device based on machine learning |
-
2022
- 2022-03-30 CN CN202210321060.8A patent/CN114414580B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN114414580A (en) | 2022-04-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2020233152A1 (en) | Urban building space data-based built-up area boundary identification method and device | |
| Zhu et al. | Seed point selection method for triangle constrained image matching propagation | |
| CN107248159A (en) | A kind of metal works defect inspection method based on binocular vision | |
| CN114414580B (en) | Method for identifying and deducting impurities on scrap steel | |
| CN109961470B (en) | Living standing tree She Shuxing accurate estimation method based on laser point cloud | |
| EP4290470A1 (en) | Rock identification method, system and apparatus, terminal, and readable storage medium | |
| CN107392929A (en) | A kind of intelligent target detection and dimension measurement method based on human vision model | |
| CN106845559A (en) | Take the ground mulching verification method and system of POI data special heterogeneity into account | |
| CN110363299A (en) | Space reasoning by cases method towards delamination-terrane of appearing | |
| CN111626321A (en) | Image data clustering method and device | |
| CN108830317A (en) | The quick fine obtaining value method of open mine side slope ROCK MASS JOINT occurrence based on digital photogrammetry | |
| CN102506753B (en) | Fourteen-point spherical wavelet transformation-based shape difference detection method for irregular parts | |
| CN103116183B (en) | Method of oil earthquake collection surface element covering degree property body slicing mapping | |
| CN114037835B (en) | Grain quality estimation method, device, equipment and storage medium | |
| CN117173364B (en) | Slicing and plotting method and system based on building three-dimensional plotting | |
| CN113469968B (en) | Method and device for detecting height difference of box seams of cigarette packaging box and electronic equipment | |
| CN117765006A (en) | Multi-level dense crown segmentation method based on unmanned aerial vehicle image and laser point cloud | |
| CN111275753B (en) | Method for measuring volume of sand and stone in carriage | |
| CN111985823A (en) | Crystal bar quality evaluation method for roller mill orientation instrument | |
| CN114937215B (en) | Method and device for identifying urban functional area | |
| CN110427961A (en) | A kind of rule-based and samples fusion architecture information extracting method and system | |
| CN109948205A (en) | A kind of nterface roughness calculation method based on three-dimensional configuration description | |
| US20230066348A1 (en) | A System for Evaluating the Scan Quality of a Scanner and a Method Thereof | |
| Von Saldern et al. | Parameterization of Arctic sea-ice surface roughness for application in ice type classification | |
| JP2008178605A (en) | Back shape classifying and determining method and back shape classifying and determining apparatus |
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 |