CN111781127B - Visual detection system and method for detecting wear particle body in lubricating oil - Google Patents
Visual detection system and method for detecting wear particle body in lubricating oil Download PDFInfo
- Publication number
- CN111781127B CN111781127B CN202010575724.4A CN202010575724A CN111781127B CN 111781127 B CN111781127 B CN 111781127B CN 202010575724 A CN202010575724 A CN 202010575724A CN 111781127 B CN111781127 B CN 111781127B
- Authority
- CN
- China
- Prior art keywords
- container
- detection
- sample
- pump body
- detecting
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 132
- 239000002245 particle Substances 0.000 title claims abstract description 79
- 239000010687 lubricating oil Substances 0.000 title claims abstract description 32
- 230000000007 visual effect Effects 0.000 title claims abstract description 20
- 238000000034 method Methods 0.000 title claims description 21
- 238000003860 storage Methods 0.000 claims abstract description 34
- 238000004140 cleaning Methods 0.000 claims abstract description 16
- 238000012545 processing Methods 0.000 claims abstract description 16
- 230000003993 interaction Effects 0.000 claims abstract description 12
- 238000005086 pumping Methods 0.000 claims abstract description 7
- 230000009471 action Effects 0.000 claims abstract description 3
- 239000000523 sample Substances 0.000 claims description 49
- 238000011179 visual inspection Methods 0.000 claims description 15
- 238000013135 deep learning Methods 0.000 claims description 12
- 239000003599 detergent Substances 0.000 claims description 12
- 238000005299 abrasion Methods 0.000 claims description 10
- 238000012360 testing method Methods 0.000 claims description 8
- 230000015572 biosynthetic process Effects 0.000 claims description 6
- 239000007788 liquid Substances 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 5
- 239000012459 cleaning agent Substances 0.000 claims description 3
- 238000003825 pressing Methods 0.000 claims description 3
- 239000012488 sample solution Substances 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 description 6
- 238000001228 spectrum Methods 0.000 description 4
- 238000002347 injection Methods 0.000 description 3
- 239000007924 injection Substances 0.000 description 3
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 2
- 238000010521 absorption reaction Methods 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 229910052742 iron Inorganic materials 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000003921 oil Substances 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 239000000243 solution Substances 0.000 description 1
- 239000000758 substrate Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N15/00—Investigating characteristics of particles; Investigating permeability, pore-volume, or surface-area of porous materials
- G01N15/10—Investigating individual particles
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B08—CLEANING
- B08B—CLEANING IN GENERAL; PREVENTION OF FOULING IN GENERAL
- B08B9/00—Cleaning hollow articles by methods or apparatus specially adapted thereto
- B08B9/08—Cleaning containers, e.g. tanks
- B08B9/093—Cleaning containers, e.g. tanks by the force of jets or sprays
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G01N2015/1022—
-
- G01N2015/1024—
-
- G01N2015/103—
Abstract
The invention provides a visual detection system for detecting wear particle shapes in lubricating oil, which comprises a container module, a visual module, an image acquisition and processing module and a control execution module, wherein the container module comprises a detection container, a storage container and a pumping device, and is used for loading a detection sample and cleaning the detection container; the visual module comprises a CCD camera and a stroboscopic light source and is used for acquiring images of the detection container and the lubricating oil in the detection container; the image acquisition processing module comprises a human-computer interaction program and an image processing program, and is used for image processing, analysis and human-computer interaction; the control execution module is used for controlling the work of the CCD camera, the pump and the stroboscopic light source and controlling the rotation action of the detection container. The invention provides a visual detection system for wear particle shapes in lubricating oil, which can be applied to visual detection of the wear particle shapes in the lubricating oil and count the number of wear particles in different shapes.
Description
Technical Field
The invention relates to the technical field of particle detection, in particular to a visual detection system and a detection method for detecting wear particle shapes in lubricating oil.
Background
The existing lubricating oil wear particle detection technology mainly comprises an iron spectrum technology and a spectrum technology. The ferrograph technology needs to apply a strong magnetic field to separate wear particles from lubricating oil, and then detects information such as appearance, quantity and the like of the wear particles by an instrument, but the separated wear particles need to be manually placed on a substrate for counting without overlapping when in use, so that errors are easy to generate. An atomic absorption spectrometer in the spectrum technology has high analysis sensitivity and reliable analysis, but has the defects of more frequent oil sample pretreatment, multiple elements to be detected needing multiple element lamps and the like, so that the operation is troublesome and the analysis speed is slow; the atomic emission spectrometer in the spectrum technology is simple to operate and high in analysis speed, but the atomic emission spectrometer needs to be placed in a constant-temperature cabinet with stable temperature for operation.
At present, a detection system with a simple structure is needed in the aspect of detecting the wear particle shape of lubricating oil, and for this purpose, a visual detection system for the wear particle shape with higher precision is provided.
Disclosure of Invention
The invention aims to solve the problem that the existing method for detecting the wear particle shape in the lubricating oil cannot meet the market demand due to the factors of complex operation, specific environment requirement and the like, and provides a visual detection system and a detection method for detecting the wear particle shape in the lubricating oil.
In particular, the invention provides a visual detection system for detecting wear particle shapes in lubricating oil, which comprises a container module, a visual module, an image acquisition processing module and a control execution module,
the container module comprises a detection container, a storage container and a pumping device, and is used for loading a detection sample and cleaning the detection container;
the visual module comprises a CCD camera and a stroboscopic light source and is used for acquiring images of the detection container and the lubricating oil in the detection container;
the image acquisition processing module comprises a human-computer interaction program and an image processing program, and is used for image processing, analysis and human-computer interaction;
the control execution module is used for controlling the work of the CCD camera, the pump and the stroboscopic light source and controlling the rotation action of the detection container;
the container module comprises a detection container and a storage container, the detection container comprises a circular transparent container and a rotating accessory, the thickness of the circular transparent container is smaller than or equal to 2mm, the outer surface of the circular transparent container is provided with a circular area with a fixed size, the volume corresponding to the area of the circular area is a fixed value, the circular area is used for measuring abrasion particles in a certain volume, the rotating accessory is used for fixing the detection container and is matched with a steering engine to realize the motion control of the container, so that the 180-degree reciprocating rotation operation is realized,
four hole-shaped structures for fixing are uniformly arranged on the outer surface of the periphery of the container, namely a first fixing hole, a second fixing hole, a third fixing hole and a fourth fixing hole;
the rotary accessory is uniformly provided with four columnar structures which are respectively a first fixed column, a second fixed column, a third fixed column and a fourth fixed column, and the first fixed column, the second fixed column, the third fixed column and the fourth fixed column are respectively matched with a first fixed hole, a second fixed hole, a third fixed hole and a fourth fixed hole on the periphery of the detection container, so that the detection container is fixed;
the center of the rotating accessory is provided with a center hole, and the center hole is used for installing a steering engine so as to realize the motion control of the control execution module on the container, realize the 180-degree reciprocating rotation and enable the inlet of the detection container and the outlet of the detection container to face the designated positions;
the pumping device is connected between the detection container and the storage container and is used for pumping the detection sample and the cleaning agent in the storage container into the detection container by using a pump, so that the detection sample is automatically injected and the detection container is automatically cleaned.
Preferably, a sample container and a detergent container are arranged in the storage container, the storage container can be connected with an external pipeline by switching the two containers, a first pump body and a second pump body are respectively arranged at an inlet and an outlet of the storage container, the first pump body is used for drawing liquid in the detection container back into the storage container, and the second pump body is used for pressing the liquid in the storage container into the detection container.
Preferably, the image acquisition processing module is specifically configured to acquire data for performing deep learning before detection, obtain network models for detecting wear particles in different forms, perform wear particle detection on a sample to be detected by using the obtained network models for the wear particles in different forms, count the number of the wear particles in different forms, give data of the wear particles after aggregation, and output the data to a human-computer interaction page.
Preferably, the control execution module is matched with the container module so as to obtain target image data under a certain condition, and the steering engine is used for controlling the detection container to perform 180-degree reciprocating rotation operation.
The invention also provides a method for carrying out visual detection based on the visual detection system, which comprises the following steps:
s1, collecting related data before detection to perform deep learning to obtain network models for detecting wear particles in different forms;
and S2, detecting the image acquired by the CCD camera by using the model obtained in the step S1, counting the number of the wear particles in different forms, summarizing the number of the wear particles, giving out data of the wear particles, and outputting the data to a man-machine interaction page.
Preferably, step S1 specifically includes the following sub-steps:
s11, setting the number of picture samples needed by training, and detecting whether the number of the picture samples reaches the set number:
s12, if the number of the samples of the pictures reaches the set number, the step is directly proceeded to S14, and if the number of the samples of the pictures does not reach the set number, the step is proceeded to S13;
s13, replacing the sample, rotating the sample 180 degrees back and forth to make the wear particles in the sample move, starting the CCD camera and the stroboscopic light source, obtaining the sequence image, processing the obtained image by using the frame difference method, finding out the wear particles in the container by using the frame difference image, combining the result obtained by the frame difference method, marking the wear particles with different forms in the shot original picture, ending the marking process until the number of the samples of the picture reaches the set number, and going into the step S14;
and S14, learning the marked sample picture by using a deep learning network to obtain and store a network model for detecting the form of the wear particles.
Preferably, step S2 specifically includes the following sub-steps:
s21, detecting whether a sample to be detected exists or not, continuously waiting until the sample to be detected is detected if no sample to be detected exists, and entering the step S22;
s22, when a sample to be detected is detected, injecting a detection sample and controlling the detection container to perform 180-degree reciprocating rotation operation by using a steering engine so as to enable abrasion particles in the container to move;
s23, starting the CCD camera and the stroboscopic light source to photograph the container to obtain a picture to be detected, and detecting a target image by using a detection model obtained by deep learning;
s24, counting different detected targets after detection is finished, and outputting and displaying a detection result;
and S25, obtaining the abrasion condition of different parts of the machine using the lubricating oil according to the detection result, and cleaning the detection container after the detection is finished.
Preferably, step S22 specifically includes the following steps:
s221, controlling the rotating accessory to rotate by using the steering engine to enable the inlet of the detection container to face downwards;
s222, switching the container in the storage container to a sample container, starting the first pump body and the second pump body simultaneously, injecting the sample solution and circulating the injected sample.
Preferably, step S25 is specifically: the rotary accessory is controlled by the steering engine to rotate, so that the outlet of the detection container faces downwards, the second pump body is started, a test sample in the detection container is discharged, the second pump body is stopped after the discharge, the container in the storage container is switched to a detergent container, the first pump body and the second pump body are started simultaneously, detergent is injected for cleaning, the first pump body is stopped after the cleaning, the second pump body is started, the detergent is discharged, and the second pump body is stopped after the discharge.
Compared with the prior art, the invention has the following beneficial effects:
(1) the invention provides a visual detection system for detecting wear particle shapes in lubricating oil, which can be applied to visual detection of wear particle shapes in lubricating oil, count the number of wear particles in different forms, give out data of the wear particles after summarizing, and output the data to a man-machine interaction page. And the target image can be detected by using a detection model obtained by deep learning; counting different detected targets after detection is finished; outputting and displaying the detection result; the abrasion conditions of different parts of the machine using the lubricating oil are given according to the detection result, and the detection container is cleaned after the detection is finished.
(2) According to the invention, a stroboscopic light source and a digital controller are used for enabling a CCD camera to acquire images, so that the phenomenon of particle trailing caused by mechanical shaking is avoided; and the digital controller is used for controlling the operation of the two pumps in the container module and the switching of the two small containers in the storage container so as to realize the injection of the detection sample and the cleaning of the detection container.
Drawings
FIG. 1 is a system block diagram of a visual inspection system of the present invention for detecting wear particle formation in lubricating oil;
FIG. 2 is a schematic structural view of a container module used in the visual inspection system for detecting wear particle forms in lubricating oil of the present invention;
FIG. 3 is a schematic view of a rotary attachment used with a test receptacle in a visual inspection system for detecting wear particle species in lubricating oil according to the present invention;
FIG. 4 is a flow chart of model training of the present invention;
FIG. 5 is a flow chart of the overall control of the visual inspection system of the present invention for detecting wear particle formation in lubricating oil;
FIG. 6 is a flow chart of the injection testing of samples according to the present invention; and
FIG. 7 is a flow chart of the present invention for cleaning a test receptacle.
Reference numbers in the figures: 1-detecting the container; 2-a circular area; 3-a first fixing hole; 4-a second fixing hole; 5-a third fixing hole; 6-a fourth fixing hole; 7-detecting the outlet of the container; 8-detection of the container inlet; 9-a first pump body; 10-a sample container; 11-a storage container; 12-a detergent container; 13-a second pump body; 14-first fixed column; 15-a second fixed column; 16-a third fixed column; 17-a fourth fixed column; 18-central hole.
Detailed Description
Hereinafter, embodiments of the present invention will be described with reference to the drawings.
The visual detection system for detecting the wear particle shape in the lubricating oil is mainly divided into a container module, a visual module, an image acquisition and processing module and a control execution module as shown in figure 1.
The container module comprises a detection container, a storage container, a pump and a pipeline, the detection container comprises a circular transparent container and a rotary accessory, the used container module is shown in figure 2, the detection container 1 is a circular transparent container, the container is only 2mm thick, a circular area 2 with a fixed size is arranged on the surface of the container, the volume corresponding to the area of the circular area 2 is a fixed value, and the abrasion particles of lubricating oil in a certain volume can be conveniently measured. And this circular area will be the specified operation area for all image processing, and target marking and detection of wear particles will not be performed for all targets outside the area.
Four hole-shaped structures for fixing are uniformly arranged on the outer surface of the periphery of the container, namely a first fixing hole 3, a second fixing hole 4, a third fixing hole 5 and a fourth fixing hole 6.
The rotating accessory matched with the detection container for use is shown in fig. 3, the rotating accessory is uniformly provided with four columnar structures, namely a first fixing column 14, a second fixing column 15, a third fixing column 16 and a fourth fixing column 17, the first fixing column 14, the second fixing column 15, the third fixing column 16 and the fourth fixing column 17 are matched with first fixing holes 3, second fixing holes 4, third fixing holes 5 and fourth fixing holes 6 on the periphery of the detection container, and therefore the detection container is fixed.
In fig. 3, a central hole 18 is formed in the center of the rotary accessory, and the central hole 18 is used for installing a steering engine, so that the motion control of the control execution module on the container is realized, 180-degree reciprocating rotation is realized, and the detection container inlet 8 and the detection container outlet 7 face the designated positions.
Two small containers are arranged in the storage container 11, the two small containers are respectively a sample container 10 and a detergent container 12, the storage container 11 can be connected with an external pipeline in a switching mode, two pumps are arranged at an inlet and an outlet of the storage container and respectively comprise a first pump body 9 and a second pump body 13, the first pump body 9 is used for drawing liquid in the detection container back into the storage container, and the second pump body 13 is used for pressing the liquid in the storage container into the detection container 1.
The image acquisition processing module is used for completing two different stages of work:
the first working phase is to collect data for deep learning before detection: the collected data is used for model training, and the model training process is as shown in fig. 4, and the number of picture samples required for training is first set, and whether the number of the picture samples reaches the set number is detected.
If the number of the samples does not reach the set number, replacing the samples, rotating the samples back and forth by 180 degrees to enable the internal wear particles to move, starting a CCD camera and a stroboscopic light source, and acquiring sequence images; the acquired images are processed using a frame difference method, and the frame difference map is used to find out the wear particles in the container frame, and an important criterion for determining the wear particles is that the wear particles move with the container when the container rotates. The frame difference method can effectively filter out the interference caused by the container or other reasons, and avoid marking wrong targets when marking the worn particle targets, wherein the interference mainly refers to stains and scratches on the surface of the container. And combining the results obtained by the frame difference method, and marking the wear particles with different forms in the shot original picture.
And if the set number is reached, ending the marking process, and learning the marked sample pictures by using a deep learning network to obtain and store a network model for detecting the form of the wear particles.
The second working stage is to carry out abrasion particle detection on the sample to be detected: and detecting the image acquired by the CCD camera by using the wear particle detection model after obtaining the wear particle detection model. And counting the number of the wear particles in different forms, summarizing the number of the wear particles, giving out data of the wear particles, and outputting the data to a man-machine interaction page.
The control execution module is mainly used for controlling the work of the CCD camera and the stroboscopic light source. The module is used for matching with a container module to obtain target image data under certain conditions, and the detection container is controlled by a steering engine to rotate back and forth for 180 degrees.
In the using process of the invention, in order to avoid the phenomenon of particle trailing caused by mechanical shaking, a stroboscopic light source and a digital controller are used for enabling a CCD camera to acquire images; the module is also used for controlling the operation of two pumps in the container module and the switching of two small containers in the storage container, so as to realize the injection of the detection sample and the cleaning of the detection container.
FIG. 5 is a flow chart showing the overall control of visual inspection for detecting wear particle formation in lubricating oil, as follows: whether a sample to be detected is detected or not is detected, no sample to be detected continues to wait, and when the sample to be detected is detected, a detection sample is injected firstly, and the steering engine is used for controlling the detection container to rotate back and forth for 180 degrees, so that the wear particles in the container move.
Then, starting a CCD camera and a strobe light source to photograph the container to obtain a picture to be detected; and detecting the target image by using a detection model obtained by deep learning.
And counting different detected targets after detection is finished, outputting and displaying a detection result, giving the abrasion conditions of different parts of the machine using the lubricating oil according to the detection result, and cleaning the detection container after the detection is finished.
As shown in fig. 6, a flow chart of injecting a test sample is shown, in which a steering engine is used to control a rotary accessory to rotate, so that an inlet of a test container faces downwards, a small container in a storage container is switched to a sample container, a first pump body and a second pump body are simultaneously started, a sample solution is injected, and the injected sample is circulated.
As shown in fig. 7, a flow chart for cleaning the detection container is shown, a steering engine is used to control the rotation of the rotary accessory, so that the outlet of the detection container faces downward, the second pump body is started, the test sample in the detection container is discharged, the second pump body is stopped after the discharge, the small container in the storage container is switched to a detergent container, the first pump body and the second pump body are started simultaneously, detergent is injected for cleaning, the first pump body is stopped after the cleaning, the second pump body is started, the detergent is discharged, and the second pump body is stopped after the discharge.
The above-mentioned embodiments are merely illustrative of the preferred embodiments of the present invention, and do not limit the scope of the present invention, and various modifications and improvements made to the technical solution of the present invention by those skilled in the art without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.
Claims (9)
1. A visual inspection system for detecting wear particle features in lubricating oil, comprising: which comprises a container module, a vision module, an image acquisition processing module and a control execution module,
the container module comprises a detection container, a storage container and a pumping device, and is used for loading a detection sample and cleaning the detection container;
the visual module comprises a CCD camera and a stroboscopic light source and is used for acquiring images of the detection container and the lubricating oil in the detection container;
the image acquisition processing module comprises a human-computer interaction program and an image processing program, and is used for image processing, analysis and human-computer interaction;
the control execution module is used for controlling the work of the CCD camera, the pump and the stroboscopic light source and controlling the rotation action of the detection container;
the container module comprises a detection container and a storage container, the detection container comprises a circular transparent container and a rotary accessory, the outer surface of the circular transparent container is provided with a circular area with a fixed size, the volume corresponding to the area of the circular area is a fixed value, the circular area is used for measuring abrasion particles in a certain volume, the rotary accessory is used for fixing the detection container and is matched with a steering engine to realize the motion control of the container, so that the 180-degree reciprocating rotation operation is realized;
four hole-shaped structures for fixing are uniformly arranged on the outer surface of the periphery of the container, namely a first fixing hole, a second fixing hole, a third fixing hole and a fourth fixing hole;
the rotary accessory is uniformly provided with four columnar structures which are respectively a first fixed column, a second fixed column, a third fixed column and a fourth fixed column, and the first fixed column, the second fixed column, the third fixed column and the fourth fixed column are respectively matched with a first fixed hole, a second fixed hole, a third fixed hole and a fourth fixed hole on the periphery of the detection container, so that the detection container is fixed;
the center of the rotating accessory is provided with a center hole, and the center hole is used for installing a steering engine so as to realize the motion control of the control execution module on the container, realize the 180-degree reciprocating rotation and enable the inlet of the detection container and the outlet of the detection container to face the designated positions;
the pumping device is connected between the detection container and the storage container and is used for pumping the detection sample and the cleaning agent in the storage container into the detection container by using a pump, so that the detection sample is automatically injected and the detection container is automatically cleaned.
2. A visual inspection system for detecting wear particle formation in lubricating oil as claimed in claim 1, wherein: the storage container is internally provided with a sample container and a cleaning agent container, the storage container can be switched between two containers and connected with an external pipeline, an inlet and an outlet of the storage container are respectively provided with a first pump body and a second pump body, the first pump body is used for drawing liquid in the detection container back into the storage container, and the second pump body is used for pressing the liquid in the storage container into the detection container.
3. A visual inspection system for detecting wear particle formation in lubricating oil as claimed in claim 1, wherein: the image acquisition processing module is specifically used for acquiring data for deep learning before detection, obtaining network models for detecting different forms of wear particles, detecting the wear particles of a sample to be detected by using the obtained network models for detecting the wear particles of different forms, counting the number of the wear particles of different forms, summarizing, then giving out data of the wear particles, and outputting the data to a human-computer interaction page.
4. A visual inspection system for detecting wear particle formation in lubricating oil as claimed in claim 2, wherein: the control execution module is matched with the container module so as to obtain target image data under certain conditions, and the steering engine is used for controlling the detection container to rotate back and forth for 180 degrees.
5. A method of visual inspection by a visual inspection system according to any one of claims 1 to 4, wherein: which comprises the following steps:
s1, collecting related data before detection to perform deep learning to obtain network models for detecting wear particles in different forms;
and S2, detecting the image acquired by the CCD camera by using the model obtained in the step S1, counting the number of the wear particles in different forms, summarizing the number of the wear particles, giving out data of the wear particles, and outputting the data to a man-machine interaction page.
6. The method of visual inspection of claim 5, wherein: step S1 specifically includes the following substeps:
s11, setting the number of picture samples needed by training, and detecting whether the number of the picture samples reaches the set number:
s12, if the number of the samples of the pictures reaches the set number, the step is directly proceeded to S14, and if the number of the samples of the pictures does not reach the set number, the step is proceeded to S13;
s13, replacing the sample, rotating the sample 180 degrees back and forth to make the wear particles in the sample move, starting the CCD camera and the stroboscopic light source, obtaining the sequence image, processing the obtained image by using the frame difference method, finding out the wear particles in the container by using the frame difference image, combining the result obtained by the frame difference method, marking the wear particles with different forms in the shot original picture, ending the marking process until the number of the samples of the picture reaches the set number, and going into the step S14;
and S14, learning the marked sample picture by using a deep learning network to obtain and store a network model for detecting the form of the wear particles.
7. The method of visual inspection of claim 5, wherein: step S2 specifically includes the following substeps:
s21, detecting whether a sample to be detected exists or not, continuously waiting until the sample to be detected is detected if no sample to be detected exists, and entering the step S22;
s22, when a sample to be detected is detected, injecting a detection sample and controlling the detection container to perform 180-degree reciprocating rotation operation by using a steering engine so as to enable abrasion particles in the container to move;
s23, starting the CCD camera and the stroboscopic light source to photograph the container to obtain a picture to be detected, and detecting a target image by using a detection model obtained by deep learning;
s24, counting different detected targets after detection is finished, and outputting and displaying a detection result;
and S25, obtaining the abrasion condition of different parts of the machine using the lubricating oil according to the detection result, and cleaning the detection container after the detection is finished.
8. The method of visual inspection of claim 7, wherein: step S22 specifically includes the following steps:
s221, controlling the rotating accessory to rotate by using the steering engine to enable the inlet of the detection container to face downwards;
s222, switching the container in the storage container to a sample container, starting the first pump body and the second pump body simultaneously, injecting the sample solution and circulating the injected sample.
9. The method of visual inspection of claim 7, wherein: step S25 specifically includes: the rotary accessory is controlled by the steering engine to rotate, so that the outlet of the detection container faces downwards, the second pump body is started, a test sample in the detection container is discharged, the second pump body is stopped after the discharge, the container in the storage container is switched to a detergent container, the first pump body and the second pump body are started simultaneously, detergent is injected for cleaning, the first pump body is stopped after the cleaning, the second pump body is started, the detergent is discharged, and the second pump body is stopped after the discharge.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010575724.4A CN111781127B (en) | 2020-06-22 | 2020-06-22 | Visual detection system and method for detecting wear particle body in lubricating oil |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010575724.4A CN111781127B (en) | 2020-06-22 | 2020-06-22 | Visual detection system and method for detecting wear particle body in lubricating oil |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111781127A CN111781127A (en) | 2020-10-16 |
CN111781127B true CN111781127B (en) | 2021-06-11 |
Family
ID=72756084
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010575724.4A Active CN111781127B (en) | 2020-06-22 | 2020-06-22 | Visual detection system and method for detecting wear particle body in lubricating oil |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111781127B (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114460084A (en) * | 2022-01-24 | 2022-05-10 | 上海道均技术有限公司 | Wear detection method/apparatus, wear detection control method, wear detection medium, and electronic device |
CN116908062B (en) * | 2023-09-07 | 2023-12-19 | 儒亚科技(北京)有限公司 | Online detection analyzer for particles in lubricating oil |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TWI648532B (en) * | 2011-08-29 | 2019-01-21 | 美商安美基公司 | Method and apparatus for non-destructive detection - undissolved particles in a fluid |
CN103293165A (en) * | 2012-02-27 | 2013-09-11 | 刘云龙 | Solid granular foreign matter detection device and method for automatic light examining machine |
CN105352858B (en) * | 2015-11-16 | 2018-09-28 | 中国矿业大学 | A kind of lubricating oil abrasive grain on-Line Monitor Device and working method based on Image Acquisition |
CN110320379B (en) * | 2018-03-30 | 2024-01-09 | 成都深迈瑞医疗电子技术研究院有限公司 | Sample analyzer and cleaning method thereof |
CN210803200U (en) * | 2019-05-28 | 2020-06-19 | 南京卡威数字科技有限公司 | Analytical ferrograph |
-
2020
- 2020-06-22 CN CN202010575724.4A patent/CN111781127B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN111781127A (en) | 2020-10-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111781127B (en) | Visual detection system and method for detecting wear particle body in lubricating oil | |
US6718053B1 (en) | Method and apparatus for automated image analysis of biological specimens | |
US7433027B2 (en) | Apparatus and method for detecting lens thickness | |
CN100501312C (en) | Gem tri-dimensional cut detection device based on machine vision | |
CN101068309A (en) | Method and device for automatic rotating mobile terminal camera head shooting results | |
EP0588969A1 (en) | Optical imaging for agglutination detection | |
CN203695443U (en) | Screening and measuring device for precise hardware | |
FR2814565A1 (en) | Procedure to improve incorporation of metrology in computer-assisted manufacture, uses decomposition of component data held in large CAM file into smaller data files and analysis files that guide analysis of surface geometry of part | |
JP2002541494A (en) | Histological reconstruction and automatic image analysis | |
CN1808056A (en) | Apparatus for measuring thickness profile and refractive index distribution of multiple layers of thin films by means of two-dimensional refrectometry and method of measuring the same | |
CN102621116A (en) | Detection device and method for quality evaluation of American ginseng | |
CN218412570U (en) | Full-automatic chemiluminescence determinator | |
CN105115797B (en) | Blood pushing piece machine, blood film preparation method, the device and method for determining push jack parameter | |
CN109557110A (en) | The full surface blemish detection device of bearing ring and method based on machine vision | |
CN201795784U (en) | Device for measuring optical axis angle of polarization characteristic component | |
CN207439958U (en) | Surface defects of products detection device | |
KR101644815B1 (en) | testing system for compact precision mechanical component and its method | |
CN208782988U (en) | A kind of veiling glare detection board | |
CN106770913A (en) | A kind of fatty acid value automatic tester | |
JP3676878B2 (en) | Pollen automatic collection and analysis system | |
CN109855802A (en) | A kind of dynamic balancing algorithm | |
CN109829897B (en) | Gear burr detection method and gear high-precision vision measurement system | |
Gadelmawla et al. | Application of computer vision for the prediction of cutting conditions in milling operations | |
CN208537027U (en) | Automatic erasing is divided brightness measurement system | |
CN113156146A (en) | Full-automatic high-throughput cell analysis equipment and detection method thereof |
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 |