CN112529867A - Method for identifying sealing element in waste steel material - Google Patents
Method for identifying sealing element in waste steel material Download PDFInfo
- Publication number
- CN112529867A CN112529867A CN202011433254.4A CN202011433254A CN112529867A CN 112529867 A CN112529867 A CN 112529867A CN 202011433254 A CN202011433254 A CN 202011433254A CN 112529867 A CN112529867 A CN 112529867A
- Authority
- CN
- China
- Prior art keywords
- sealing element
- connected domain
- identifying
- ellipse
- steel
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000007789 sealing Methods 0.000 title claims abstract description 63
- 229910000831 Steel Inorganic materials 0.000 title claims abstract description 51
- 239000010959 steel Substances 0.000 title claims abstract description 51
- 238000000034 method Methods 0.000 title claims abstract description 34
- 239000000463 material Substances 0.000 title claims abstract description 16
- 239000002699 waste material Substances 0.000 title description 10
- 238000001514 detection method Methods 0.000 claims abstract description 22
- 238000003709 image segmentation Methods 0.000 claims abstract description 9
- 238000007689 inspection Methods 0.000 claims abstract description 6
- 238000013135 deep learning Methods 0.000 claims description 5
- 238000007599 discharging Methods 0.000 claims description 4
- 238000012544 monitoring process Methods 0.000 claims description 3
- 239000002360 explosive Substances 0.000 abstract description 11
- 238000003723 Smelting Methods 0.000 description 6
- 238000009628 steelmaking Methods 0.000 description 6
- 238000004519 manufacturing process Methods 0.000 description 4
- 238000005520 cutting process Methods 0.000 description 3
- 238000011084 recovery Methods 0.000 description 3
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 description 2
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 2
- 238000005266 casting Methods 0.000 description 2
- 239000000571 coke Substances 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000004880 explosion Methods 0.000 description 2
- 238000004064 recycling Methods 0.000 description 2
- 229910001021 Ferroalloy Inorganic materials 0.000 description 1
- 235000019738 Limestone Nutrition 0.000 description 1
- 229910000805 Pig iron Inorganic materials 0.000 description 1
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000007664 blowing Methods 0.000 description 1
- 229910002092 carbon dioxide Inorganic materials 0.000 description 1
- 239000001569 carbon dioxide Substances 0.000 description 1
- 239000003245 coal Substances 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 239000012535 impurity Substances 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 229910052742 iron Inorganic materials 0.000 description 1
- 239000006028 limestone Substances 0.000 description 1
- 238000010309 melting process Methods 0.000 description 1
- 239000007769 metal material Substances 0.000 description 1
- 238000000643 oven drying Methods 0.000 description 1
- 229910052760 oxygen Inorganic materials 0.000 description 1
- 239000001301 oxygen Substances 0.000 description 1
- 239000004033 plastic Substances 0.000 description 1
- 229920003023 plastic Polymers 0.000 description 1
- 239000000843 powder Substances 0.000 description 1
- 238000009423 ventilation Methods 0.000 description 1
Images
Classifications
-
- 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
- G06T7/0008—Industrial image inspection checking presence/absence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- 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
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- 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)
- Quality & Reliability (AREA)
- Geometry (AREA)
- Processing Of Solid Wastes (AREA)
Abstract
The invention discloses a method for identifying a sealing element in a steel scrap material, and belongs to the technical field of detection and identification. The invention discloses a method for identifying a sealing element in a steel scrap material, which specifically comprises the following steps: step one, obtaining a picture of a material pile to be detected or a material pile detection range as a target picture; step two, using a unet model to detect the sealing element of the target picture, and obtaining a primary sealing element image segmentation result, namely a connected domain corresponding to the sealing element; step three, carrying out oval fitting on the obtained connected domains of the sealing elements respectively; step four, calculating the area difference value of the fitted ellipse and the corresponding connected domain; and step five, excluding the connected domain with a large area difference according to a set threshold value, thereby obtaining the sealing element corresponding to the connected domain which cannot be excluded, namely the sealing element judged to be actually existing. By adopting the technical scheme of the invention, sealing elements, explosives and the like in the scrap steel can be effectively identified, so that the problem of possible missing detection in the traditional manual quality inspection is solved.
Description
Technical Field
The invention belongs to the technical field of detection and identification, and particularly relates to a method for identifying a sealing element in a steel scrap material.
Background
The scrap steel is an energy-saving renewable resource which can be recycled unlimitedly, one ton of scrap steel can be used more, 0.4 ton of coke or one ton of raw coal can be saved, the consumption of 1.7 ton of concentrate powder can be reduced, 4.3 ton of raw ore exploitation can be reduced, and the emission of 1.6 ton of carbon dioxide can be reduced. In recent years, with the formation of a large-scale industrial chain with high technical integration in the scrap steel recycling industry, the mode of scrap steel recovery and re-smelting is not lower than the traditional converter steelmaking mode in economic benefit.
Scrap steel is a resource and its source is also very extensive. Generally, scrap is mainly produced in steel-making, steel-casting or steel-processing plants, and in the manufacturing and processing of steel products, mainly in the cutting of steel material into ends, tails, chips, scrap, and the like. Waste equipment, waste parts, steel components, scrapped vehicles, and the like (also called depreciated steel scrap) are also important sources of steel scrap. In addition, waste steel such as cans (also called "social scrap" or "refuse scrap") in articles of daily use is also a scrap resource. However, because of its wide source, it is necessary to pay special attention to seals, explosives, etc. when scrap steel is used as a material for steel making and casting. The first place of peace is that there is nothing without safety. Even if the process is a non-smelting process, the pretreatment process of the scrap steel before smelting is afraid of cutting sealing elements and explosives, and the process usually adopts oxygen to cut the scrap steel; in the cutting process, the temperature of the sealing element and the explosive rises suddenly, the internal pressure is increased rapidly, and explosion can occur when the critical explosion condition is reached, so that major accidents are caused. Not to mention that the sealing elements and the explosives are put into the smelting furnace, the life safety of steel-making workers and the production safety of steel plants are endangered.
For example, in the chinese patent application No. 201410744398.X, a cupola melting process includes the steps of: 1) preparing materials: the method comprises the following steps of (1) including components of materials such as coke, pig iron, ferroalloy, limestone and the like, wherein the maximum size of a metal material is not more than one third of the furnace diameter near a charging opening, and scrap steel and return iron are seriously corroded and must be treated so as to be put into a furnace; various furnace charges are stacked in a classified mode, mixing is avoided, no dangerous objects such as untreated cartridge cases, waste guns and the like are mixed, and harmful impurities such as rubber, plastics and the like are prevented from being thrown into the furnace; 2) the formula is as follows: 3) repairing and baking the furnace; 4) after oven drying, adding firewood, igniting, and opening the air port cover for natural ventilation; 5) charging; 6) and (5) blowing air for smelting. The application indicates that various furnace charges are stacked in a classified mode and cannot be mixed when the materials are prepared, and dangerous objects such as untreated shells, waste guns and the like cannot be mixed, but the application does not specifically give out how to detect sealing elements, explosives and the like in the waste steel.
At present, in the existing processes of steel scrap recovery and re-smelting, except for relying on practitioners such as steel scrap recovery manufacturers and steel enterprises to improve safety awareness and self-customs control, arranging quality inspectors to customs control on a system, establishing a punishment system and the like, a more direct and effective method is not provided, and the problems can be fundamentally solved. Therefore, how to realize the efficient and accurate detection of sealing elements, explosives and the like in the scrap steel has important significance for recycling the scrap steel resources.
Disclosure of Invention
1. Technical problem to be solved by the invention
The invention aims to overcome the defect that sealing elements, explosives and the like in the waste steel are difficult to effectively identify in the prior art, so that great potential safety hazards are brought to the life safety of steel-making workers and the production safety of steel mills, and provides an identification method of the sealing elements in the waste steel. By adopting the technical scheme of the invention, sealing elements, explosives and the like in the scrap steel can be effectively identified, so that the problem of possible missing detection in the traditional manual quality inspection is solved.
2. Technical scheme
In order to achieve the purpose, the technical scheme provided by the invention is as follows:
the invention discloses a method for identifying a sealing element in a steel scrap, which is used for detecting and identifying the sealing element in the steel scrap based on a deep learning unet image segmentation model and an ellipse detection algorithm, and specifically comprises the following steps:
step one, obtaining a picture of a material pile to be detected or a material pile detection range as a target picture;
step two, using a unet model to detect the sealing element of the target picture, and obtaining a primary sealing element image segmentation result, namely a connected domain corresponding to the sealing element;
step three, carrying out oval fitting on the obtained connected domains of the sealing elements respectively;
step four, calculating the area difference value of the fitted ellipse and the corresponding connected domain;
and step five, excluding the connected domain with a large area difference according to a set threshold value, thereby obtaining the sealing element corresponding to the connected domain which cannot be excluded, namely the sealing element judged to be actually existing.
Furthermore, in the first step, a picture of a stockpile viewing range or an overlook picture of a carriage of a discharging vehicle on the spot in the discharging process is taken as a target picture by using a scrap yard network video monitoring system.
Furthermore, if there are multiple seals in the target picture, an output graph corresponding to multiple connected domains is output in step two.
Furthermore, in the third step, a fit ellipse fitting algorithm in opencv is called to perform ellipse fitting processing on the pixel points of each connected domain respectively through a least square method, and then the ellipse regions corresponding to the connected domains are obtained.
Furthermore, when the connected domains are subjected to ellipse fitting, the ellipse regions corresponding to the connected domains are circled, and the pixel points of the connected domains are located on the ellipse as much as possible.
Furthermore, in the fourth step, the area of the connected domain corresponding to the ellipse is calculated according to all the pixel points in the elliptical area corresponding to the connected domain, and meanwhile, the area of each connected domain is calculated according to all the pixel points of the connected domain.
3. Advantageous effects
Compared with the prior art, the technical scheme provided by the invention has the following remarkable effects:
(1) the method for identifying the sealing element in the steel scrap is based on the image segmentation model and the ellipse detection algorithm, so that AI identification can be performed on the sealing element in the steel scrap, columnar explosives such as aeronautical shells, cannonballs and the like, and the occurrence of the missing detection phenomenon can be effectively reduced compared with the conventional method of mainly performing quality detection manually, so that the life safety of steel-making workers and the production safety of steel plants can be ensured.
(2) According to the method for identifying the sealing element in the steel scrap, disclosed by the invention, through the combination of the deep learning unet model and the ellipse detection algorithm, the deep learning unet model is utilized to detect the sealing element on a target picture, a primary sealing element image segmentation result is obtained, then the connected domain of each sealing element is subjected to ellipse fitting, the area difference value between the fitted ellipse and the corresponding connected domain is calculated, and the connected domain with a larger area difference value is eliminated according to the set threshold value, so that the detection and identification precision of the sealing element in the steel scrap and the like can be effectively improved, and the detection omission and false positive false detection can be further avoided.
Drawings
FIG. 1 is a schematic flow chart illustrating the identification of a sealing member in a scrap according to the present invention;
fig. 2 is a schematic diagram of a connected domain output diagram of a corresponding seal member output after detection by a unet model.
Detailed Description
Based on the problems that quality inspection is mainly carried out on sealing elements in the existing steel scrap manually, the phenomenon of missing inspection is easy to occur, and the detection efficiency is low, the invention provides an automatic identification method for the sealing elements in the steel scrap. It should be noted that, because the characteristics of the sealing element in the steel scrap are not obvious, the selection of the image segmentation model is crucial to the detection accuracy, and the detection accuracy of the sealing element in the steel scrap can be effectively improved by selecting the deep learning unet model in the invention compared with other target object detection models.
Specifically, with reference to fig. 1, the method for identifying a sealing member in scrap steel according to the present invention specifically comprises the following steps:
the method is characterized in that a scrap yard network video monitoring system is used for taking a picture of an inspection range of a stockpile or an overhead picture of a carriage of a vehicle for unloading on the site, and the sealing member in the scrap steel is inspected for quality and the shape of the sealing member is close to that of a common columnar explosive (such as aerobomb, cannonball and the like).
Firstly, taking the photo as a target picture, and detecting a sealing element on the target picture by using a unet model to obtain a primary sealing element image segmentation result, namely a connected domain corresponding to the sealing element; if a plurality of sealing elements exist in the target picture, outputting an output graph corresponding to a plurality of connected domains, as shown in fig. 2;
then, ellipse fitting is performed on each connected domain: for the connected domain, calling a fit ellipse fitting algorithm in opencv to fit an ellipse by a least square method so that the pixel point of the connected domain is on the ellipse as much as possible, and respectively performing ellipse fitting processing on the pixel point of the connected domain; an oval area corresponding to the connected domain is circled, namely the oval area corresponding to the connected domain;
after the ellipse fitting is completed, calculating the area of the connected domain corresponding to the ellipse according to all pixel points in the elliptical region corresponding to the connected domain, and meanwhile calculating the area of the connected domain according to the pixel points of the connected domain;
and calculating the area difference value of the fitted ellipse and the connected domain, and eliminating the connected domain with larger difference value and the sealing element corresponding to the connected domain which cannot be eliminated through a set threshold value, wherein the sealing element is judged to be a real sealing element if no false positive problem exists.
Claims (6)
1. The method for identifying the sealing element in the steel scrap is characterized by detecting and identifying the sealing element in the steel scrap based on a deep learning unet image segmentation model and an ellipse detection algorithm, and specifically comprises the following steps:
step one, obtaining a picture of a material pile to be detected or a material pile detection range as a target picture;
step two, using a unet model to detect the sealing element of the target picture, and obtaining a primary sealing element image segmentation result, namely a connected domain corresponding to the sealing element;
step three, carrying out oval fitting on the obtained connected domains of the sealing elements respectively;
step four, calculating the area difference value of the fitted ellipse and the corresponding connected domain;
and step five, excluding the connected domain with a large area difference according to a set threshold value, thereby obtaining the sealing element corresponding to the connected domain which cannot be excluded, namely the sealing element judged to be actually existing.
2. The method for identifying the sealing element in the steel scrap according to claim 1, wherein the method comprises the following steps: in the first step, a picture of a stockpile inspection range or an overlook picture of a carriage of a discharging vehicle on the spot in the discharging process is taken as a target picture by using a scrap yard network video monitoring system.
3. The method for identifying the sealing element in the steel scrap according to claim 1, wherein the method comprises the following steps: and if a plurality of sealing pieces exist in the target picture, outputting an output graph corresponding to a plurality of connected domains in the second step.
4. The method for identifying the sealing element in the steel scrap according to claim 1, wherein the method comprises the following steps: and calling a fit ellipse fitting algorithm in opencv to perform ellipse fitting processing on the pixel points of each connected domain respectively through a least square method in the third step, so as to obtain an ellipse region corresponding to each connected domain.
5. The method for identifying the sealing element in the steel scrap according to claim 4, wherein the method comprises the following steps: and when the connected domains are subjected to ellipse fitting processing, the ellipse regions corresponding to the connected domains are circled, and the pixel points of the connected domains are positioned on the ellipse as much as possible.
6. A method of identifying seals in scrap according to claims 1 to 5, in which: and in the fourth step, the area of the connected domain corresponding to the ellipse is calculated according to all the pixel points in the elliptical area corresponding to the connected domain, and meanwhile, the area of each connected domain is calculated according to all the pixel points of the connected domain.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011433254.4A CN112529867A (en) | 2020-12-10 | 2020-12-10 | Method for identifying sealing element in waste steel material |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011433254.4A CN112529867A (en) | 2020-12-10 | 2020-12-10 | Method for identifying sealing element in waste steel material |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112529867A true CN112529867A (en) | 2021-03-19 |
Family
ID=74999928
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011433254.4A Pending CN112529867A (en) | 2020-12-10 | 2020-12-10 | Method for identifying sealing element in waste steel material |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112529867A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113810605A (en) * | 2021-08-17 | 2021-12-17 | 阿里巴巴达摩院(杭州)科技有限公司 | Target object processing method and device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109948510A (en) * | 2019-03-14 | 2019-06-28 | 北京易道博识科技有限公司 | A kind of file and picture example dividing method and device |
| WO2019149071A1 (en) * | 2018-01-30 | 2019-08-08 | 华为技术有限公司 | Target detection method, device, and system |
| CN111178173A (en) * | 2019-12-14 | 2020-05-19 | 杭州电子科技大学 | Target colony growth characteristic identification method |
-
2020
- 2020-12-10 CN CN202011433254.4A patent/CN112529867A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019149071A1 (en) * | 2018-01-30 | 2019-08-08 | 华为技术有限公司 | Target detection method, device, and system |
| CN109948510A (en) * | 2019-03-14 | 2019-06-28 | 北京易道博识科技有限公司 | A kind of file and picture example dividing method and device |
| CN111178173A (en) * | 2019-12-14 | 2020-05-19 | 杭州电子科技大学 | Target colony growth characteristic identification method |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113810605A (en) * | 2021-08-17 | 2021-12-17 | 阿里巴巴达摩院(杭州)科技有限公司 | Target object processing method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Lv et al. | Life cycle energy consumption and greenhouse gas emissions of iron pelletizing process in China, a case study | |
| CN112529867A (en) | Method for identifying sealing element in waste steel material | |
| CN103160640A (en) | Method of dynamically detecting contents of manganese, phosphorus and sulphur of slag in converter steelmaking process | |
| CN104392213B (en) | A kind of image information state recognition system suitable for fusion process | |
| CN115330780A (en) | Rapid detection method for slag inclusion defect of metal welding | |
| Wallace | Production of secondary aluminium | |
| CN110738434A (en) | Quality tracing management method for carbon anode production whole process | |
| CN103695666B (en) | Utilize the method for metallic zinc in molten point modified with reduction stove enrichment industrial solid castoff | |
| CN108595383A (en) | A kind of residual heat resources analysis method and system | |
| CN103447480A (en) | Smelting and casting operation method for producing automobile engine cylinder by adopting composite iron casting technology | |
| CN111112287A (en) | Hazardous waste recycling comprehensive treatment system adopting electric arc furnace and treatment method thereof | |
| Javaid et al. | Final report on scrap management, sorting and classification of steel | |
| CN113516416B (en) | Evaluation method, device, equipment and medium for cement kiln waste treatment complete cycle | |
| CN106169100A (en) | Automobile product scraps recovery method | |
| Lis et al. | Options of utilizing steelmaking dust in a non-metallurgical industry | |
| Chizhikova | Best available techniques in the blast-furnace production | |
| Wei et al. | Analysis on pollution prevention and control of waste lead battery recycling process | |
| CN110883070A (en) | Comprehensive treatment system and method for recycling hazardous waste by adopting converter | |
| Ayres et al. | Measurement of thickness of oxygen lance skull in LD converters using artificial vision | |
| CN104649564A (en) | Hot-wind charging device applied to glass furnace in furnace heating-up period | |
| Radović et al. | Cleaner metallurgical industry in Serbia: A road to the sustainable development | |
| Yingjun et al. | A case study of LCA for environmental protection in steel company | |
| CN111139371A (en) | Preparation method and equipment of green low-cost regenerated aluminum alloy | |
| CN211275859U (en) | Hazardous waste recycling comprehensive treatment system adopting intermediate frequency furnace | |
| CN108319210B (en) | Analytic hierarchy process-based green early warning method for sand casting process |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210319 |