CN116012371A - Machine learning-based mechanical equipment oil leakage detection method and system - Google Patents
Machine learning-based mechanical equipment oil leakage detection method and system Download PDFInfo
- Publication number
- CN116012371A CN116012371A CN202310155923.3A CN202310155923A CN116012371A CN 116012371 A CN116012371 A CN 116012371A CN 202310155923 A CN202310155923 A CN 202310155923A CN 116012371 A CN116012371 A CN 116012371A
- Authority
- CN
- China
- Prior art keywords
- oil
- image
- area
- mechanical equipment
- receiving disc
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 28
- 238000010801 machine learning Methods 0.000 title claims abstract description 24
- 238000000034 method Methods 0.000 claims abstract description 28
- 238000006243 chemical reaction Methods 0.000 claims abstract description 17
- 238000012216 screening Methods 0.000 claims abstract description 12
- 238000004519 manufacturing process Methods 0.000 claims abstract description 11
- 238000012545 processing Methods 0.000 claims description 25
- 238000001914 filtration Methods 0.000 claims description 18
- 238000004458 analytical method Methods 0.000 claims description 9
- 238000000605 extraction Methods 0.000 claims description 8
- 230000006870 function Effects 0.000 claims description 6
- 238000003384 imaging method Methods 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 238000012544 monitoring process Methods 0.000 abstract description 5
- 238000007689 inspection Methods 0.000 abstract description 4
- 238000004364 calculation method Methods 0.000 description 5
- 239000004973 liquid crystal related substance Substances 0.000 description 4
- 238000013135 deep learning Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 101100025813 Caenorhabditis elegans dpy-31 gene Proteins 0.000 description 1
- 101100348003 Caenorhabditis elegans toh-1 gene Proteins 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000004040 coloring Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A10/00—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE at coastal zones; at river basins
- Y02A10/40—Controlling or monitoring, e.g. of flood or hurricane; Forecasting, e.g. risk assessment or mapping
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a machine learning-based mechanical equipment oil leakage detection method and a system thereof, wherein the method comprises the following steps: s1, acquiring an image of an oil receiving disc area of mechanical equipment to be tested as an original image; s2, taking an original image as input, and manufacturing a mask; s3, dividing the original image according to the mask, and extracting an oil receiving disc area to obtain an oil receiving disc image with interference removed; s4, extracting oil drops in the oil receiving disc image through HSV color space conversion; s5, obtaining oil drop positions and oil drop areas through screening of the connected domains and hole positioning; s6, judging the oil leakage state according to the size of the oil drop area. According to the invention, the machine learning method is adopted to model the oil leakage algorithm of the mechanical equipment, so that the oil leakage monitoring of the mechanical equipment can be performed in real time, the problems of detection omission, high cost and incapability of real-time monitoring in manual inspection are solved, and the feasibility is high.
Description
Technical Field
The invention relates to the technical field of oil leakage detection, in particular to a machine learning-based mechanical equipment oil leakage detection method and a machine learning-based mechanical equipment oil leakage detection system.
Background
The mechanical equipment leaks oil, so that the oil is wasted, the production environment of a factory is polluted, and potential safety hazards such as fire disaster are brought. The oil leakage phenomenon of mechanical equipment is rare, but the mechanical equipment needs to be strictly stopped, and once the fire disaster problem occurs, serious consequences are brought. Therefore, it is a necessary task to perform oil leakage detection on the mechanical equipment to prevent the mechanical equipment from being damaged.
At present, an oil leakage condition of equipment is usually detected by adopting a manual regular inspection mode, however, the mode has the problems of missed detection, high cost and incapability of real-time monitoring. In the prior art, automatic detection equipment is adopted to replace manual inspection to detect oil leakage, but a deep learning method is used for modeling an oil leakage algorithm of mechanical equipment, and because oil leakage data are less and most mechanical equipment cannot be stopped in actual production, sample types are seriously unbalanced, a serious overfitting problem can be caused by adopting the deep learning-based method, and the feasibility is low.
Disclosure of Invention
The invention aims to provide a machine learning-based mechanical equipment oil leakage detection method and a machine learning-based mechanical equipment oil leakage detection system, which are capable of carrying out real-time oil leakage monitoring on mechanical equipment by adopting a machine learning method to model an oil leakage algorithm of the mechanical equipment, solving the problems of detection omission, high cost and incapability of real-time monitoring in manual inspection and the over-fitting problem caused by serious unbalance of oil leakage sample types, and have higher feasibility.
In order to achieve the above object, the present invention provides a machine learning-based oil leakage detection method for a mechanical device, comprising the steps of:
s1, acquiring an image of an oil receiving disc area of mechanical equipment to be tested as an original image;
s2, taking the original image as input, and manufacturing a mask;
s3, dividing the original image according to the mask, and extracting an oil receiving disc area to obtain an oil receiving disc image with interference removed;
s4, extracting oil drops in the oil receiving disc image through HSV color space conversion;
s5, obtaining oil drop positions and oil drop areas through screening of the connected domains and hole positioning;
s6, judging the oil leakage state according to the size of the oil drop area.
Further, the specific operation of making the mask comprises the following steps:
s21, converting the original image from an RGB color space to an HSV color space according to a first chroma threshold, filtering, performing binarization processing, and generating a binarized image as a preliminary mask to finish rough extraction of an oil receiving disc area;
s22, carrying out connected domain processing on the binarized image, selecting a maximum connected domain area, and further reducing an oil receiving disc area;
s23, filling holes in the largest area of the connected domain;
s24, generating a target mask.
Further, the specific operation of step S21 includes: inputting the original RGB three-channel imageIConverting to HSV format imageI hsv The method comprises the steps of carrying out a first treatment on the surface of the By limitingI hsv Each channel of the imageH、S、VExtracting the color region of the oil receiving tray to obtain an image after color extractionI color The method comprises the steps of carrying out a first treatment on the surface of the Extracting the color from the imageI color Converting into gray level imageI gray The method comprises the steps of carrying out a first treatment on the surface of the Through threshold valueth 1 Binarization filtering to obtain binarized imageI binary 。
Further, the specific operation of step S22 includes: the binarized image is subjected to a connectiedcomponents-WithStats function using opencvI binary Performing connected domain processing, filtering the binarized image after contour analysisI binary The area of the medium connected domain is smaller, the largest area is left as the oil receiving disc area, and the screened image of the connected domain only containing the largest area is obtainedI cont-max 。
Further, in the step S23, the hole filling isFilling holes existing in the maximum area region of the connected domain through a pan Hong Suanfa, and screening the connected domain to obtain an imageI cont-max Holes in the middle oil receiving disc area are filled to obtain binary 0, 255 pixel value single-channel imagesI fillhole 。
Further, the specific operation of step S3 includes: converting the binary 0, 255 pixel value single-channel image into 0, 1 representation, and extracting and dividing the binary 0, 255 pixel value single-channel image as mask to obtain an input RGB imageISpecific color area of oil receiving disc to obtain image of oil receiving discI img-mask 。
Further, the specific operation of step S4 includes: imaging the drip tray according to a second chromaticity thresholdI img-mask HSV color space conversion to obtain an imageI hsv-mask The method comprises the steps of carrying out a first treatment on the surface of the By combining the imagesI hsv-mask Each channelH、S、VThreshold filtering color to remove darker oil drop region to obtain imageI color-hole The method comprises the steps of carrying out a first treatment on the surface of the The image is processedI color-hole Conversion to grey-scale drawingsI gray-hole The method comprises the steps of carrying out a first treatment on the surface of the By binarizing threshold valuesth 2 Filtering to obtain a binarized single-channel imageI binary-hole For the binarized single-channel imageI binary-hole Black-white inversion is carried out to obtain an image with a cavity, namely an oil leakage positionI oil-hole 。
Further, the image is processed using the connectidcomponentsWithStats function of opencvI oil-hole Processing the connected domain to obtain a final oil leakage result imageI oil Through the final oil leakage result imageI oil And positioning the oil drop position coordinates to obtain the pixel area of an oil drop region, and judging the oil leakage state according to the size of the pixel area of the oil drop region.
The invention also provides a machine learning-based mechanical equipment oil leakage detection system, which comprises:
the image acquisition unit is used for acquiring images of the oil receiving disc area of the mechanical equipment to be tested;
the preprocessing unit is used for manufacturing a mask and extracting an oil receiving disc image;
the image processing unit filters and eliminates oil drops in the oil receiving disc image through HSV color space conversion, and obtains oil drop positions and oil drop areas through connected domain screening and hole positioning;
and the analysis processing unit is used for judging the oil leakage state according to the size of the oil drop area.
Further, the device also comprises an oil leakage alarm unit, and when the analysis processing unit judges that oil leaks, the oil leakage alarm unit gives an alarm.
Compared with the prior art, the invention has the following beneficial effects:
according to the invention, the mask is arranged to automatically extract the oil receiving disc area, other interference areas are removed, an oil receiving disc image is obtained, oil drops in the oil receiving disc image are filtered and removed through HSV color space conversion, the positions of the oil drops and the oil drop areas are obtained through screening of the communicating areas and positioning of holes, and the oil leakage state of the mechanical equipment is judged according to the sizes of the oil drop areas. The method can automatically identify whether the mechanical equipment leaks oil or not, and solves the safety problem in the production environment.
According to the invention, the original image is converted from the RGB color space to the HSV color space, then the binarization processing component is used for carrying out preliminary mask, the connected domain processing is carried out on the binarized image, the largest connected domain area is selected, the oil receiving disc area is further reduced, holes in the largest connected domain area are filled, the target mask is calculated and generated to manufacture the mask, the problem of over-fitting caused by serious unbalance of the oil leakage sample type is solved, and the feasibility is high.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a machine learning-based oil leakage detection method for a mechanical device according to an embodiment of the present invention.
Fig. 2 is a flowchart of mask making according to an embodiment of the present invention.
Detailed Description
In order to better understand the technical solutions in the present application, the following description will clearly and completely describe the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are within the scope of the present disclosure.
Throughout the specification and claims, unless explicitly stated otherwise, the term "comprise" or variations thereof such as "comprises" or "comprising", etc. will be understood to include the stated element or component without excluding other elements or components.
1-2, an embodiment of the invention provides a machine learning-based mechanical equipment oil leakage detection method, which comprises the following steps:
s1, acquiring an image of an oil receiving disc area of mechanical equipment to be tested as an original image;
s2, taking the original image as input, and manufacturing a mask;
s3, dividing the original image according to the mask, and extracting an oil receiving disc area to obtain an oil receiving disc image with interference removed;
s4, filtering and removing oil drops in the oil receiving disc image through HSV color space conversion;
s5, obtaining oil drop positions and oil drop areas through screening of the connected domains and hole positioning;
s6, judging the oil leakage state according to the size of the oil drop area.
In general, from the viewpoints of safety and pollution, the oil receiving disc is placed at a position where oil is easily leaked by the mechanical equipment, and therefore, whether the oil is leaked by the mechanical equipment can be judged by detecting the oil receiving disc. In this embodiment, the device for collecting the image is an industrial camera, which can collect a clear and distinguishable image of the oil leakage area of the mechanical device, and obtain the oil leakage state of the mechanical device.
As shown in fig. 2, the specific operations for making the mask include the following steps:
s21, converting the original image from an RGB color space to an HSV color space according to a first chroma threshold, filtering, performing binarization, and generating a binarized image as a preliminary mask to finish rough extraction of an oil receiving disc area.
HSV is an intuitive color model, can intuitively express the tone, vividness and brightness of colors, and is convenient for color comparison. In the HSV color space, objects of a certain color are more readily available. When converting the original image from RGB color space to HSV color space, defining tone #H) Saturation ofS) And brightness%V) The upper and lower limits of (1) are used to obtain the color region of the drip pan based on the threshold, in this embodiment, the first color threshold of HSV color space is set toH1∈[0,30]，S1∈[43,255]，V1∈[46,255]。
In the step, binarization processing is carried out through formulas (1) - (4), and a preliminary mask is constructed, wherein the concrete calculation process is as follows:
I hsv =RGB2HSV(I) (1)
I color =I hsv [H,S,V](2)
I gray= GRAY(I color ) (3)
I binary =I gray [th 1 ](4)
wherein, the liquid crystal display device comprises a liquid crystal display device,Iis an original image in the RGB format,I hsv is an HSV format image of the image,I color for the image after the color extraction,I gray in the form of a gray scale map,I binary is a binarized image of the image to be processed,th 1 is the corresponding binarization threshold.
Specifically, equation (1) is to input the original RGB three-channel imageIConverting to HSV format imageI hsv The method comprises the steps of carrying out a first treatment on the surface of the Equation (2) is by way of limitationI hsv Each channel of the imageH、S、VExtracting the color region of the oil receiving tray to obtain an image after color extractionI color The method comprises the steps of carrying out a first treatment on the surface of the Equation (3) is an image after extracting the colorI color Converting into gray level imageI gray The method comprises the steps of carrying out a first treatment on the surface of the Equation (4) is the pass thresholdth 1 Binarization filtering to obtain binarized imageI binary 。
S22, carrying out connected domain processing on the binarized image, selecting a maximum connected domain area, and further reducing an oil receiving disc area.
The specific operations in step S22 include: performing connected domain processing on the binarized image by using a connected components-WithStats function of opencv, filtering a region with smaller connected domain area in the binarized image after contour analysis, and leaving a maximum area region as an oil receiving disc region to obtain a connected domain filtered image only containing the maximum area regionI cont-max The calculation process is as follows:
I cont-max ,stats=connectedComponentsWithStats(I binary ) (5)
wherein, the formula (5) is obtained by binarizingI binary Obtaining a connected domain screened image only containing the largest area region by screening the area of the image connected domainI cont-max Stats contains the maximum area connected domain coordinates and connected domain area size information.
S23, filling holes in the largest area of the connected domain.
Holes are formed by missing pixels at partial positions of the mask in color extraction, the holes are filled through flooding Hong Suanfa, and the flooding algorithm starts from a pixel point, so that coloring is expanded to surrounding pixel points until the boundary of the image, and the holes in the mask are filled.
The hole filling is to fill holes existing in the area with the largest area of the connected domain through flooding Hong Suanfa, and the calculation formula of the flooding algorithm is as follows:
I fillhole =fillhole(I cont-max ) (6)
equation (6) is an image obtained after screening the connected domainI cont-max Filling the hollow in the middle oil receiving disc area to obtainI fillhole Binarized 0, 255 pixel value single channel imageI fillhole 。
S24, generating a target mask.
In the step S3, the original image is segmented according to a mask, and an oil receiving disc area is extracted to obtain an oil receiving disc imageI img-mask The calculation process is as follows:
I img-mask =I*(I fillhole /255) (7)
* Multiplying the corresponding positions of the images. The formula (7) is to convert the binary 0, 255 pixel value single-channel image into 0, 1 representation and then extract and divide the binary 0, 255 pixel value single-channel image into an input RGB image as a maskISpecific color area of oil receiving disc to obtain image of oil receiving discI img-mask 。
Further, the specific operation of step S4 includes: imaging the drip tray according to a second chromaticity thresholdI img-mask HSV color space conversion to obtain an imageI hsv-mask The method comprises the steps of carrying out a first treatment on the surface of the By combining the imagesI hsv-mask Each channelH、S、VThreshold filtering color to remove darker oil drop region to obtain imageI color-hole The method comprises the steps of carrying out a first treatment on the surface of the The image is processedI color-hole Conversion to grey-scale drawingsI gray-hole The method comprises the steps of carrying out a first treatment on the surface of the By binarizing threshold valuesth 2 Filtering to obtain a binarized single-channel imageI binary-hole For the binarized single-channel imageI binary-hole Black-white inversion is carried out to obtain an image with a cavity, namely an oil leakage positionI oil-hole The calculation process is as follows:
I hsv-mask =RGB2HSV(I img-mask ) (8)
I color-hole =I hsv-mask [H,S,V](9)
I gray-hole= GRAY(I color-hole ) (10)
I binary-hole =I gray-hole [th 2 ](11)
I oil-hole =255-I binary-hole (12)
wherein, the liquid crystal display device comprises a liquid crystal display device,th 2 is the corresponding binarization threshold.
Specifically, the formula (8) is an image of the drip pan obtained in the formula (7)I img-mask Conversion to HSV color space to obtain an imageI hsv-mask The method comprises the steps of carrying out a first treatment on the surface of the Equation (9) is by combining the imagesI hsv-mask Each channelH、S、VThreshold filtering color to remove darker oil drop regionI color-hole The method comprises the steps of carrying out a first treatment on the surface of the Equation (10) is to obtain gray-scale image by image gray-scale image processingI gray-hole The method comprises the steps of carrying out a first treatment on the surface of the Equation (11) is a threshold value by binarizationth 2 Filtering to obtain a binarized single-channel imageI binary-hole At this time, the pixel value of the oil drop area is 0, and an image with the pixel value of the oil drop area being 255 and the pixel value of the other oil receiving pan area being 0 is obtained by image inversionI oil-hole So as to determine the oil leakage position and the like through the communicating region.
The oil droplets appear brown and are imaged against the oil panI img-mask Proceeding withHSV color space conversion to imageI hsv-mask The second chrominance threshold of the HSV color space is set toH2∈[6,30]，S2∈[43,255]，V2∈[65,255]Image is formedI hsv-mask Conversion to grey-scale drawingsI gray-hole Then removing darker oil drops, reserving the oil receiving disc area, wherein the oil drop area is provided with holes, and binarizing the imageI oil-binary Black-white inversion to obtain image with cavity, i.e. oil leakage positionI oil-hole 。
Further, the image is processed using the connectidcomponentsWithStats function of opencvI oil-hole Processing the connected domain to obtain a final oil leakage result imageI oil Through the final oil leakage result imageI oil And positioning oil drop position coordinates to obtain the pixel area of an oil drop region, judging the oil leakage state according to the size of the pixel area of the oil drop region, and calculating as follows:
I oil ，stats=connectedComponentsWithStats(I oil-hole )(13)
equation (13) is for processing an image by a connected region functionI oil-hole Obtaining stats which comprise oil drop coordinates and area size, and obtaining a final oil leakage result diagram with oil leakage area pixel value of 255 and other positions pixel value of zeroI oil 。
The invention also provides a machine learning-based mechanical equipment oil leakage detection system, which comprises:
the image acquisition unit is used for acquiring images of the oil receiving disc area of the mechanical equipment to be tested;
the preprocessing unit is used for manufacturing a mask and extracting an oil receiving disc image;
the image processing unit filters and eliminates oil drops in the oil receiving disc image through HSV color space conversion, and obtains oil drop positions and oil drop areas through connected domain screening and hole positioning;
and the analysis processing unit is used for judging the oil leakage state according to the size of the oil drop area.
Further, the system also comprises an oil leakage alarm unit, and the oil leakage alarm unit gives an alarm when the analysis processing unit judges that oil leaks.
The foregoing descriptions of specific exemplary embodiments of the present invention are presented for purposes of illustration and description. It is not intended to limit the invention to the precise form disclosed, and obviously many modifications and variations are possible in light of the above teaching. The exemplary embodiments were chosen and described in order to explain the specific principles of the invention and its practical application to thereby enable one skilled in the art to make and utilize the invention in various exemplary embodiments and with various modifications as are suited to the particular use contemplated. It is intended that the scope of the invention be defined by the claims and their equivalents.
Claims (10)
1. The machine learning-based mechanical equipment oil leakage detection method is characterized by comprising the following steps of:
s1, acquiring an image of an oil receiving disc area of mechanical equipment to be tested as an original image;
s2, taking the original image as input, and manufacturing a mask;
s3, dividing the original image according to the mask, and extracting an oil receiving disc area to obtain an oil receiving disc image with interference removed;
s4, extracting oil drops in the oil receiving disc image through HSV color space conversion;
s5, obtaining oil drop positions and oil drop areas through screening of the connected domains and hole positioning;
s6, judging the oil leakage state according to the size of the oil drop area.
2. The machine learning based mechanical equipment oil leakage detection method of claim 1, wherein the specific operations of making a mask comprise the steps of:
s21, converting the original image from an RGB color space to an HSV color space according to a first chroma threshold, filtering, performing binarization processing, and generating a binarized image as a preliminary mask to finish rough extraction of an oil receiving disc area;
s22, carrying out connected domain processing on the binarized image, selecting a maximum connected domain area, and further reducing an oil receiving disc area;
s23, filling holes in the largest area of the connected domain;
s24, generating a target mask.
3. The machine learning based mechanical equipment oil leakage detection method according to claim 2, wherein the specific operations of step S21 include: inputting the original RGB three-channel imageIConverting to HSV format imageI hsv The method comprises the steps of carrying out a first treatment on the surface of the By limitingI hsv Each channel of the imageH、S、VExtracting the color region of the oil receiving tray to obtain an image after color extractionI color The method comprises the steps of carrying out a first treatment on the surface of the Extracting the color from the imageI color Converting into gray level imageI gray ；
Through threshold valueth 1 Binarization filtering to obtain binarized imageI binary 。
4. The machine learning based mechanical equipment oil leakage detection method according to claim 3, wherein the specific operations of step S22 include: the binarized image is subjected to a connectiedcomponents-WithStats function using opencvI binary Performing connected domain processing, filtering the binarized image after contour analysisI binary The area of the medium connected domain is smaller, the largest area is left as the oil receiving disc area, and the screened image of the connected domain only containing the largest area is obtainedI cont-max 。
5. The machine learning based mechanical equipment oil leakage detection method according to claim 4, wherein in the step S23, the hole filling is performed by filling the hole through a pan Hong SuanfaHoles in the largest area region of the connected domain, and screening the connected domain to obtain an imageI cont-max Holes in the middle oil receiving disc area are filled to obtain binary 0, 255 pixel value single-channel imagesI fillhole 。
6. The machine learning based mechanical equipment oil leakage detection method according to claim 5, wherein the specific operations of step S3 include: converting the binary 0, 255 pixel value single-channel image into 0, 1 representation, and extracting and dividing the binary 0, 255 pixel value single-channel image as mask to obtain an input RGB imageISpecific color area of oil receiving disc to obtain image of oil receiving discI img-mask 。
7. The machine learning based mechanical equipment oil leakage detection method according to claim 6, wherein the specific operations of step S4 include: imaging the drip tray according to a second chromaticity thresholdI img-mask HSV color space conversion to obtain an imageI hsv-mask The method comprises the steps of carrying out a first treatment on the surface of the By combining the imagesI hsv-mask Each channelH、S、VThreshold filtering color to remove darker oil drop region to obtain imageI color-hole The method comprises the steps of carrying out a first treatment on the surface of the The image is processedI color-hole Conversion to grey-scale drawingsI gray-hole The method comprises the steps of carrying out a first treatment on the surface of the By binarizing threshold valuesth 2 Filtering to obtain a binarized single-channel imageI binary-hole For the binarized single-channel imageI binary-hole Black-white inversion is carried out to obtain an image with a cavity, namely an oil leakage positionI oil-hole 。
8. The machine learning based mechanical equipment oil leak detection method of claim 7 wherein the image is processed using a connectiedcomponents wittstats function of opencvI oil-hole Processing the connected domain to obtain a final oil leakage result imageI oil Through the final leakageOil result imageI oil And positioning the oil drop position coordinates to obtain the pixel area of an oil drop region, and judging the oil leakage state according to the size of the pixel area of the oil drop region.
9. A machine learning-based mechanical equipment oil leak detection system, comprising:
the image acquisition unit is used for acquiring images of the oil receiving disc area of the mechanical equipment to be tested;
the preprocessing unit is used for manufacturing a mask and extracting an oil receiving disc image;
the image processing unit filters and eliminates oil drops in the oil receiving disc image through HSV color space conversion, and obtains oil drop positions and oil drop areas through connected domain screening and hole positioning;
and the analysis processing unit is used for judging the oil leakage state according to the size of the oil drop area.
10. The machine learning based mechanical equipment oil leak detection system of claim 9, further comprising an oil leak alarm unit that issues an alarm when the analysis processing unit determines that oil is leaking.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310155923.3A CN116012371A (en) | 2023-02-23 | 2023-02-23 | Machine learning-based mechanical equipment oil leakage detection method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310155923.3A CN116012371A (en) | 2023-02-23 | 2023-02-23 | Machine learning-based mechanical equipment oil leakage detection method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116012371A true CN116012371A (en) | 2023-04-25 |
Family
ID=86032137
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310155923.3A Pending CN116012371A (en) | 2023-02-23 | 2023-02-23 | Machine learning-based mechanical equipment oil leakage detection method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116012371A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109241851A (en) * | 2018-08-07 | 2019-01-18 | 电子科技大学 | A kind of the Oil Leakage Detecting method and its system of view-based access control model image |
| CN113034624A (en) * | 2021-05-06 | 2021-06-25 | 湖州云电笔智能科技有限公司 | Temperature early warning image identification method, system, equipment and storage medium based on temperature sensing color-changing adhesive tape |
| CN114519698A (en) * | 2022-01-14 | 2022-05-20 | 深圳市华付信息技术有限公司 | Equipment oil leakage detection method, device, equipment and storage medium in dark environment |
| CN115457297A (en) * | 2022-08-23 | 2022-12-09 | 中国航空油料集团有限公司 | Method and device for detecting oil leakage of aviation oil depot and aviation oil safety operation and maintenance system |
-
2023
- 2023-02-23 CN CN202310155923.3A patent/CN116012371A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109241851A (en) * | 2018-08-07 | 2019-01-18 | 电子科技大学 | A kind of the Oil Leakage Detecting method and its system of view-based access control model image |
| CN113034624A (en) * | 2021-05-06 | 2021-06-25 | 湖州云电笔智能科技有限公司 | Temperature early warning image identification method, system, equipment and storage medium based on temperature sensing color-changing adhesive tape |
| CN114519698A (en) * | 2022-01-14 | 2022-05-20 | 深圳市华付信息技术有限公司 | Equipment oil leakage detection method, device, equipment and storage medium in dark environment |
| CN115457297A (en) * | 2022-08-23 | 2022-12-09 | 中国航空油料集团有限公司 | Method and device for detecting oil leakage of aviation oil depot and aviation oil safety operation and maintenance system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107705288B (en) | Infrared video detection method for dangerous gas leakage under strong interference of pseudo-target motion | |
| CN111209876B (en) | Oil leakage defect detection method and system | |
| CN108550159B (en) | Flue gas concentration identification method based on image three-color segmentation | |
| CN105139386A (en) | Image processing method for quickly and automatically detecting soldered dot unsatisfied products of electric connectors | |
| CN109060842A (en) | Citrus detection method of surface flaw, computer program based on surface fitting correction | |
| CN111047655A (en) | High-definition camera cloth defect detection method based on convolutional neural network | |
| CN112730251B (en) | Device and method for detecting screen color defects | |
| CN106228541A (en) | The method and device of screen location in vision-based detection | |
| CN111062941B (en) | Point light source fault detection device and method | |
| Aziz et al. | Fabric defect detection algorithm using morphological processing and DCT | |
| CN104749801B (en) | High Precision Automatic optical detecting method and system | |
| KR20190023374A (en) | Method for testing display pannel | |
| KR101677070B1 (en) | System and Method for Automatically Detecting a Mura Defect using Morphological Image Processing and Labeling | |
| CN114596329A (en) | Gas image enhancement and gas leakage detection method and system | |
| CN108107611B (en) | Self-adaptive defect detection method and device and electronic equipment | |
| CN103761521A (en) | LBP-based microscopic image definition measuring method | |
| CN115330702A (en) | Beverage bottle filling defect identification method based on deep vision | |
| CN107818583A (en) | Cross searching detection method and device | |
| CN112634179B (en) | Camera shake prevention power transformation equipment image change detection method and system | |
| CN111008960B (en) | Aluminum electrolytic capacitor bottom appearance detection method and device based on machine vision | |
| JP2005165387A (en) | Method and device for detecting stripe defective of picture and display device | |
| CN116012371A (en) | Machine learning-based mechanical equipment oil leakage detection method and system | |
| CN111489338A (en) | Nondestructive testing method for internal defects of hydraulic pipeline of shield tunneling machine | |
| CN113008470B (en) | Gas leakage detection device and gas leakage detection method | |
| JP2022125593A (en) | Abnormality detecting method and abnormality detecting device |
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: 20230425 |