CN117554374A - Automatic defect detection classification system for ball cage and data tracking analysis method - Google Patents
Automatic defect detection classification system for ball cage and data tracking analysis method Download PDFInfo
- Publication number
- CN117554374A CN117554374A CN202311502922.8A CN202311502922A CN117554374A CN 117554374 A CN117554374 A CN 117554374A CN 202311502922 A CN202311502922 A CN 202311502922A CN 117554374 A CN117554374 A CN 117554374A
- Authority
- CN
- China
- Prior art keywords
- contour
- ball cage
- analysis result
- detection
- strategy
- 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 96
- 238000004458 analytical method Methods 0.000 title claims abstract description 71
- 230000007547 defect Effects 0.000 title claims abstract description 22
- 238000005286 illumination Methods 0.000 claims abstract description 49
- 238000000034 method Methods 0.000 claims abstract description 35
- 230000001502 supplementing effect Effects 0.000 claims abstract description 5
- 230000002159 abnormal effect Effects 0.000 claims description 21
- 238000004140 cleaning Methods 0.000 claims description 18
- 238000009826 distribution Methods 0.000 claims description 12
- 238000007689 inspection Methods 0.000 claims description 10
- 238000007664 blowing Methods 0.000 claims description 3
- 230000003247 decreasing effect Effects 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000007306 turnover Effects 0.000 description 3
- 229910000831 Steel Inorganic materials 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 239000012535 impurity Substances 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 239000010959 steel Substances 0.000 description 2
- 238000011179 visual inspection Methods 0.000 description 2
- 208000035874 Excoriation Diseases 0.000 description 1
- 238000005299 abrasion Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- JEIPFZHSYJVQDO-UHFFFAOYSA-N iron(III) oxide Inorganic materials O=[Fe]O[Fe]=O JEIPFZHSYJVQDO-UHFFFAOYSA-N 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/34—Sorting according to other particular properties
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/34—Sorting according to other particular properties
- B07C5/344—Sorting according to other particular properties according to electric or electromagnetic properties
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/36—Sorting apparatus characterised by the means used for distribution
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/01—Arrangements or apparatus for facilitating the optical investigation
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N27/00—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means
- G01N27/72—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating magnetic variables
- G01N27/82—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating magnetic variables for investigating the presence of flaws
- G01N27/90—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating magnetic variables for investigating the presence of flaws using eddy currents
- G01N27/9093—Arrangements for supporting the sensor; Combinations of eddy-current sensors and auxiliary arrangements for marking or for rejecting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- 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
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8854—Grading and classifying of flaws
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques
-
- 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
- G06T2207/30164—Workpiece; Machine component
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Chemical & Material Sciences (AREA)
- Biochemistry (AREA)
- Health & Medical Sciences (AREA)
- Analytical Chemistry (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Theoretical Computer Science (AREA)
- Signal Processing (AREA)
- Electrochemistry (AREA)
- Quality & Reliability (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Geometry (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention relates to the technical field of ball cage detection, and particularly discloses a ball cage automatic defect detection classification system and a data tracking analysis method, wherein the method comprises the following steps: the appearance detection station is provided with a plurality of groups of industrial cameras and light supplementing light sources and is used for acquiring images of all parts of the ball cage under a standard illumination strategy and a modulated illumination strategy for acquisition; the eddy current flaw detection station is provided with two groups and is used for carrying out eddy current flaw detection on the two end surfaces of the ball cage and the inner ring and the outer ring; the overturning station is used for overturning the position of the ball cage by 180 degrees; the analysis center is used for determining the analysis results of the ball cage according to the images of all parts of the ball cage under the standard illumination strategy and the modulated illumination strategy, wherein the analysis results comprise passing, failing and risk, sending out a workpiece kicking command when the analysis result is not passing, judging according to the eddy current flaw detection result at the corresponding position when the analysis result is at risk, and determining whether to send out the workpiece kicking command.
Description
Technical Field
The invention relates to the technical field of ball cage detection, in particular to a ball cage automatic defect detection classification system and a data tracking analysis method.
Background
The ball cage type universal joint is a common transmission component in the automotive field, and is formed by arranging steel balls in a ball cage (a retainer) and matching with structures such as a bell housing, an inner race and the like, wherein ball cage window holes are in direct contact and control all the steel balls (generally 6 or 8) to be always kept on the same plane, and ball cage spherical surfaces are precisely matched with the bell housing and the inner race, so that the problems of rotation, swing angle clamping stagnation, abnormal starting or rotating moment, early abrasion abnormal sound and the like of the ball cage type universal joint during normal operation are avoided, and the problems of cracks, defects, waste materials, rust, burr and the like of the ball cage are avoided in the manufacturing stage.
The existing ball cage detection mode gradually replaces the traditional manual visual inspection detection mode through an intelligent identification mode, and mainly realizes the judgment of the appearance quality of the ball cage through visual inspection and eddy current flaw detection, so that when obvious quality defects appear, the ball cage is kicked out to a special storage box by the most NG part, and further, the product on the production line is ensured to meet the requirements.
In the existing ball cage appearance detection procedure, although the problem can be accurately judged even if the impurity and the foreign matter exist on the ball cage, the subsequent processing mode is the same as the defective workpiece processing mode, which is not beneficial to the subsequent processing process.
Disclosure of Invention
The invention aims to provide a ball cage automatic defect detection classification system and a data tracking analysis method, which solve the following technical problems:
how to improve the accuracy of the detection of the appearance quality of the ball cage and the convenience of NG piece processing.
The aim of the invention can be achieved by the following technical scheme:
a ball cage automated defect detection classification system, the system comprising:
the appearance detection station is provided with a plurality of groups of industrial cameras and light supplementing light sources and is used for acquiring images of all parts of the ball cage under a standard illumination strategy and a modulated illumination strategy for acquisition;
the eddy current flaw detection station is provided with two groups and is used for carrying out eddy current flaw detection on the two end surfaces of the ball cage and the inner ring and the outer ring;
the overturning station is used for overturning the position of the ball cage by 180 degrees;
the analysis center is used for determining the analysis result of the ball cage according to the images of each part of the ball cage under the standard illumination strategy and the modulated illumination strategy, wherein the analysis result comprises passing, failing and risks, sending out a workpiece kicking command when the analysis result is not passing, judging according to the eddy current flaw detection result at the corresponding position when the analysis result is at risk, and determining whether to send out the workpiece kicking command.
Further, the process of collecting the image by the appearance detection station comprises the following steps: obtaining a first image of each position point by rotating the ball cage under a standard illumination strategy, and obtaining a second image of each position point by rotating the ball cage under a modulated illumination strategy;
the standard light strategy is to acquire an image under white sunlight and illumination intensity with a preset size; the illumination strategy is to acquire an image under the illumination with the wavelength of 450-500nm and the illumination intensity after modulation; the process for acquiring the illumination intensity after modulation comprises the following steps:
identifying a ball cage region based on AI in the first image, and carrying out gray processing on the ball cage region image;
by the formula lux=l0-f b (br) calculating to obtain a modulated illumination intensity Lux;
wherein br is the difference between the brightness of the image of the ball cage area after the graying treatment and the standard brightness, L0 is the preset illumination intensity, f b F is a light filling adjustment function b As a decreasing function, and f b (0)=0。
Further, the system also comprises an air blowing device for cleaning the ball cage on the appearance detection station;
the process for determining the analysis result comprises the following steps:
s1, carrying out contour recognition on a first image and a second image corresponding to each position point to obtain a first contour set and a second contour set, and judging whether abnormal contour lines exist in the first contour set and the second contour set or not based on comparison of standard contours;
if yes, step S2 is carried out;
if not, performing a step S3;
s2, performing coincidence comparison on the first contour and the second contour corresponding to the same position points in the first contour set and the second contour set:
if the overlap ratio is more than A ", judging that the analysis result is 'not passing', and sending a workpiece kicking command;
if the overlap ratio is less than or equal to A%, a cleaning procedure detection instruction is sent out, and step S3 is carried out;
wherein A% ∈ [0.91,1];
s3, carrying out anomaly analysis on the first contour and the second contour corresponding to the same-position points in the first contour set and the second contour set and the standard contour, and judging an analysis result according to the anomaly analysis result.
Further, the anomaly analysis process includes:
by the formula:
calculating to obtain the ith contour outlier U i U is set up i And a preset fixed threshold interval [ U1, U2 ]]And (3) performing comparison:
when U is i If the value is more than U2, judging that the analysis result is 'not passed';
when U is i If the analysis result is less than U1, judging that the analysis result is 'pass';
when U is i∈ [U1,U2]Judging the analysis result as 'risk';
wherein Su i The profile area anomaly coefficient is y, and the parameter adjusting coefficient is y; lu (Lu) i Is the abnormal coefficient of the outline shape, S 0i Standard area for ith contour, SA i For the actual area of the ith contour corresponding to the first contour, S Bi For the actual area of the ith contour corresponding to the second contour, x 1 、x 2 For presetting an adjustment coefficient and meeting x 1 <x 2 ;L Ai For the normal line of the standard contour to fall into the maximum value of the overlapping area of the first contour and the standard contour corresponding to the ith contour, L Bi The normal line of the standard contour falls into the maximum value of the overlapping area of the second contour and the standard contour corresponding to the ith contour.
Further, the process of judging according to the eddy current flaw detection result at the corresponding position comprises the following steps:
judging the distribution and the size of vortex occurrence in the process of vortex flaw detection:
if the eddy current distribution is abnormal and the eddy current is larger than the error allowable value, a workpiece kicking command is sent out;
if the eddy current distribution is abnormal, but the eddy current size is smaller than the error allowable value, determining according to the analysis result:
if the analysis result is 'risk', a workpiece kick command is sent out;
if the analysis result is 'pass', a workpiece kick command is not sent out;
if the vortex distribution is normal, a workpiece kick command is not sent out.
A data tracking analysis method of a ball cage automated defect detection classification system, the method comprising:
counting the occurrence times of kicking commands and the sending times of detection instructions of a cleaning procedure;
and step two, respectively executing a first early warning strategy and a second early warning strategy according to the analysis result, the eddy current flaw detection result and the detection instruction sending times of the cleaning procedure, and judging whether to stop checking according to the results of the first early warning strategy and the second early warning strategy.
Further, the first early warning strategy includes:
when the continuous emission times of the kicking commands are larger than a preset value or the number of times of the kicking commands in the continuous M groups is larger than one or two of the preset proportion, an outage checking instruction is emitted;
when the continuous sending times of the cleaning procedure detection instructions are larger than a preset value or the sending times of the cleaning procedure detection instructions in the continuous M groups are larger than one or two of the preset proportion, a shutdown check instruction is sent;
wherein M is more than or equal to 5.
Further, the second early warning strategy includes:
by the formula:
calculating to obtain a workpiece continuity coefficient S u ;
When w occurs u If not less than w1, sending out stop checkAn instruction;
wherein N is the detection number of the contour of the ball cage, i epsilon [1, N],α i An influence coefficient for the ith contour; g is a set value of continuously selected workpiece number, j is E [1, G];U ij The ith contour abnormal value of the jth workpiece;for the mean value of the ith contour outlier of all workpieces, w1 is a continuity reference threshold.
The invention has the beneficial effects that:
(1) According to the method, the accuracy of the detection result can be further improved by the double detection strategy mode, and on the other hand, the modulation illumination strategy can be adaptively adjusted according to the result of the standard illumination strategy, so that the suitability of the environmental condition in the secondary detection process is further ensured; in addition, the double detection strategies can carry out preliminary judgment on the types of the NG pieces on the ball cage, so that the subsequent processing of the NG pieces is facilitated; meanwhile, the analysis result obtained by the process detection is comprehensively judged according to the eddy current flaw detection result at the corresponding position, and whether a workpiece kicking command is sent or not is determined, so that the accuracy of the detection result can be further improved.
Drawings
The invention is further described below with reference to the accompanying drawings.
FIG. 1 is a schematic diagram of the layout of the automated defect detection classification system of the present invention;
FIG. 2 is a flow chart of the process of analysis result determination of the present invention;
FIG. 3 is a flow chart of steps of a data tracking analysis method of the automated ball cage defect detection classification system of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, in one embodiment, a system for detecting and classifying defects of an automated ball cage is provided, the system includes an appearance detection station, an eddy current inspection station, a turnover station and an analysis center, wherein the appearance detection station is provided with a plurality of groups, and the groups of appearance detection stations, the eddy current inspection station and the turnover station are respectively: the window hole plane appearance detection station is an inner sphere appearance detection station, an outer sphere appearance detection station is an eddy current flaw detection station, namely an outer sphere and one side end face, a turnover station is an window hole plane appearance detection station, an inner sphere appearance detection station is an eddy current flaw detection station, namely an inner sphere and the other side end face; the eddy current flaw detection station is used for carrying out eddy current flaw detection on the two end surfaces of the ball cage and the inner ring and the outer ring; the overturning station is used for overturning the position of the ball cage by 180 degrees; in addition, all appearance detection stations in the work comprise industrial cameras and light supplementing light sources, the industrial cameras and the light supplementing light sources are used for acquiring images of all parts of the ball cage under a standard illumination strategy and a modulated illumination strategy, and an analysis center is used for determining analysis results of the ball cage according to the images of all parts of the ball cage under the standard illumination strategy and the modulated illumination strategy; in addition, the double detection strategies can carry out preliminary judgment on the types of the NG pieces on the ball cage, so that the subsequent processing of the NG pieces is facilitated; meanwhile, the analysis results obtained by the process detection comprise 'pass', 'fail' and 'risk', when the analysis result is 'fail', a workpiece kicking command is sent out, and when the analysis result is 'risk', judgment is carried out according to the eddy current flaw detection result at the corresponding position, whether the workpiece kicking command is sent out is determined, and the mode is combined with the data of the eddy current flaw detection to carry out comprehensive judgment, so that the accuracy of the detection result can be further improved.
As one embodiment of the present invention, the process of capturing an image at the appearance inspection station includes:obtaining a first image of each position point by rotating the ball cage under a standard illumination strategy, and obtaining a second image of each position point by rotating the ball cage under a modulated illumination strategy; the standard light strategy is to acquire an image under white sunlight and illumination intensity with preset size, the illumination intensity with preset size is selected and set according to test data acquired from the environment where the device is located, the modulated light strategy is to acquire the image under illumination with wavelength of 450-500nm and illumination intensity after modulation, wherein the wavelength of 450-500nm is blue light, light reflection on the metal surface can be reduced, and the process of acquiring the illumination intensity after modulation comprises the following steps: identifying a ball cage region based on AI in the first image, and carrying out gray processing on the ball cage region image; by the formula lux=l0-f b (br) calculating to obtain a modulated illumination intensity Lux; wherein br is the difference between the brightness of the image of the ball cage area after the graying treatment and the standard brightness, L0 is the preset illumination intensity, f b F is a light filling adjustment function b As a decreasing function, and f b (0) The method comprises the steps of (1) carrying out matching on illumination test images according to different illumination intensities, and then obtaining the illumination test images by fitting, so that the adjusted illumination is more suitable for the actual environment state, the image acquisition definition is improved, and the accuracy of a detection result is further improved; in addition, the system also comprises an air blowing device for cleaning the ball cage on the appearance detection station; as described with reference to fig. 2, the process of determining the analysis result includes: s1, carrying out contour recognition on a first image and a second image corresponding to each position point to obtain a first contour set and a second contour set, and judging whether abnormal contour lines exist in the first contour set and the second contour set or not based on comparison of standard contours; if yes, step S2 is carried out; if not, performing a step S3; the contour recognition is based on an image processing technology and an AI technology in the prior art, wherein the image processing technology can be realized by adopting graying processing and combining an image edge recognition algorithm, and further details are not provided herein; s2, performing coincidence comparison on the first contour and the second contour corresponding to the same position points in the first contour set and the second contour set: if the overlap ratio is more than A%, the overlap ratio of the first contour and the second contour is higher, namely the surface of the ball cage has fixed defects, so that the problems of cracks, flash and the like are solvedTherefore, judging that the analysis result is 'not passed', and sending out a workpiece kicking command; if the overlap ratio is less than or equal to A, indicating that impurities and the like exist on the surface of the ball cage to influence the detection result, further sending a detection instruction of a cleaning procedure, and carrying out step S3; wherein A% E [0.91,1]]Which selects settings based on empirical data according to the type of the ball cage product; s3, carrying out anomaly analysis on the first contour and the second contour corresponding to the same-position points in the first contour set and the second contour set and the standard contour, judging an analysis result according to the anomaly analysis result, wherein the anomaly analysis process comprises the following steps: by the formula:
calculating to obtain the ith contour outlier U i Wherein S is ui The profile area anomaly coefficient is defined by y, which is a parameter adjusting coefficient according to empirical data; lu (Lu) i Is the abnormal coefficient of the outline shape, S 0i Is the standard area of the ith contour, which is obtained according to the product test data, S Ai For the actual area of the ith contour corresponding to the first contour, S Bi For the actual area of the ith contour corresponding to the second contour, x 1 、x 2 For presetting an adjustment coefficient and meeting x 1 <x 2 Setting the result accuracy according to different detection strategies in the experience data; l (L) Ai For the normal line of the standard contour to fall into the maximum value of the overlapping area of the first contour and the standard contour corresponding to the ith contour, L Bi Since the normal line of the standard contour falls into the maximum value of the region where the i-th contour corresponds to the second contour and the standard contour overlap, the abnormal coefficient of the contour area obtained by the formula (2) can determine the detection area relative to the standard areaThe deviation state of the contour shape is calculated by the formula (3), the critical state of the contour edge deviating from the standard contour can be judged, the abnormal state of the contour edge of the ball cage can be reflected according to the abnormal value of the contour calculated by the formula (1), and the abnormal state of the contour edge of the ball cage is reflected by U i And a preset fixed threshold interval [ U1, U2 ]]Alignment is performed, [ U1, U2 ]]Fitting settings according to test data, thus when U i If the value is more than U2, judging that the analysis result is 'not passed'; when U is i If the analysis result is less than U1, judging that the analysis result is 'pass'; when U is i ∈[U1,U2]And judging the analysis result as 'risk'.
As one embodiment of the present invention, the process of determining according to the eddy current inspection result at the corresponding position includes: judging the distribution and the size of vortex occurrence in the process of vortex flaw detection:
if the eddy current distribution is abnormal and the eddy current is larger than the error allowable value, a workpiece kicking command is sent out;
if the eddy current distribution is abnormal, but the eddy current size is smaller than the error allowable value, determining according to the analysis result:
if the analysis result is 'risk', a workpiece kick command is sent out;
if the analysis result is 'pass', a workpiece kick command is not sent out;
if the vortex distribution is normal, a workpiece kick command is not sent out.
Through the process, comprehensive judgment can be performed by integrating the vortex detection and the appearance detection station results, accurate judgment can be given to the workpiece which is positioned in the range of the critical value of the detection result, and further, the accuracy of the detection result can be improved while misjudgment can be reduced.
It should be noted that, the specific process of the eddy current inspection is implemented according to the prior art, and will not be further described in detail in this embodiment.
Referring to fig. 3, the embodiment provides a data tracking analysis method of a ball cage automatic defect detection classification system, which includes: counting the occurrence times of kicking commands and the sending times of detection instructions of a cleaning procedure; step two, respectively executing a first early warning strategy and a second early warning strategy according to analysis results, eddy current flaw detection results and the sending times of detection instructions of a cleaning procedure, and judging whether to stop checking according to the results of the first early warning strategy and the second early warning strategy; the first early warning strategy comprises the following steps: when the continuous emission times of the kicking commands are larger than a preset value or the number of times of the kicking commands in the continuous M groups is larger than one or two of the preset proportion, an outage checking instruction is emitted; when the continuous sending times of the cleaning procedure detection instructions are larger than a preset value or the sending times of the cleaning procedure detection instructions in the continuous M groups are larger than one or two of the preset proportion, a shutdown check instruction is sent; wherein M is more than or equal to 5.
The second early warning strategy comprises the following steps: by the formula:
calculating to obtain a workpiece continuity coefficient w u The method comprises the steps of carrying out a first treatment on the surface of the Wherein N is the detection number of the contour of the ball cage, i epsilon [1, N],α i An influence coefficient for the ith contour, which is set according to the selection of the ith contour; g is a set value of continuously selected workpiece number, j is E [1, G];U ij The ith contour abnormal value of the jth workpiece;for the ith profile anomaly value mean of all workpieces, w1 is a continuity reference threshold value, which is set based on the test data, thus setting the workpiece continuity factor w u Comparing with w1, and then at w u And when the first early warning is not less than w1, judging that the problem risk of the workpiece is larger, and then sending out a shutdown checking instruction, so that the timely early warning process of the ball cage preparation process is realized on the basis of judging the obvious fault problem by the first early warning.
The foregoing describes one embodiment of the present invention in detail, but the description is only a preferred embodiment of the present invention and should not be construed as limiting the scope of the invention. All equivalent changes and modifications within the scope of the present invention are intended to be covered by the present invention.
Claims (8)
1. A ball cage automated defect detection classification system, the system comprising:
the appearance detection station is provided with a plurality of groups of industrial cameras and light supplementing light sources and is used for acquiring images of all parts of the ball cage under a standard illumination strategy and a modulated illumination strategy for acquisition;
the eddy current flaw detection station is provided with two groups and is used for carrying out eddy current flaw detection on the two end surfaces of the ball cage and the inner ring and the outer ring;
the overturning station is used for overturning the position of the ball cage by 180 degrees;
the analysis center is used for determining the analysis result of the ball cage according to the images of each part of the ball cage under the standard illumination strategy and the modulated illumination strategy, wherein the analysis result comprises passing, failing and risks, sending out a workpiece kicking command when the analysis result is not passing, judging according to the eddy current flaw detection result at the corresponding position when the analysis result is at risk, and determining whether to send out the workpiece kicking command.
2. The automated ball cage defect detection classification system of claim 1, wherein the process of capturing images at the appearance inspection station comprises: obtaining a first image of each position point by rotating the ball cage under a standard illumination strategy, and obtaining a second image of each position point by rotating the ball cage under a modulated illumination strategy;
the standard light strategy is to acquire an image under white sunlight and illumination intensity with a preset size; the illumination strategy is to acquire an image under the illumination with the wavelength of 450-500nm and the illumination intensity after modulation; the process for acquiring the illumination intensity after modulation comprises the following steps:
identifying a ball cage region based on AI in the first image, and carrying out gray processing on the ball cage region image;
by the formula lux=l0-f b (br) calculating to obtain a modulated illumination intensity Lux;
wherein br is grayingThe difference value between the brightness of the processed ball cage region image and the standard brightness is L0, wherein L0 is preset illumination intensity, f b F is a light filling adjustment function b As a decreasing function, and f b (0)=0。
3. The automated ball cage defect inspection and classification system of claim 2, further comprising an air blowing device for cleaning the ball cage at the appearance inspection station;
the process for determining the analysis result comprises the following steps:
s1, carrying out contour recognition on a first image and a second image corresponding to each position point to obtain a first contour set and a second contour set, and judging whether abnormal contour lines exist in the first contour set and the second contour set or not based on comparison of standard contours;
if yes, step S2 is carried out;
if not, performing a step S3;
s2, performing coincidence comparison on the first contour and the second contour corresponding to the same position points in the first contour set and the second contour set:
if the overlap ratio is more than A ", judging that the analysis result is 'not passing', and sending a workpiece kicking command;
if the overlap ratio is less than or equal to A%, a cleaning procedure detection instruction is sent out, and step S3 is carried out;
wherein A% ∈ [0.91,1];
s3, carrying out anomaly analysis on the first contour and the second contour corresponding to the same-position points in the first contour set and the second contour set and the standard contour, and judging an analysis result according to the anomaly analysis result.
4. A system for automated ball cage defect detection classification as claimed in claim 3, wherein said anomaly analysis process comprises:
by the formula:
calculating to obtain the ith contour outlier U i U is set up i And a preset fixed threshold interval [ U1, U2 ]]And (3) performing comparison:
when U is i If the value is more than U2, judging that the analysis result is 'not passed';
when U is i If the analysis result is less than U1, judging that the analysis result is 'pass';
when U is i ∈[U1,U2]Judging the analysis result as 'risk';
wherein Su i The profile area anomaly coefficient is y, and the parameter adjusting coefficient is y; lu (Lu) i Is the abnormal coefficient of the outline shape, S 0i Standard area of ith contour, S Ai For the actual area of the ith contour corresponding to the first contour, S Bi For the actual area of the ith contour corresponding to the second contour, x 1 、x 2 For presetting an adjustment coefficient and meeting x 1 <x 2 ;L Ai For the normal line of the standard contour to fall into the maximum value of the overlapping area of the first contour and the standard contour corresponding to the ith contour, L Bi The normal line of the standard contour falls into the maximum value of the overlapping area of the second contour and the standard contour corresponding to the ith contour.
5. The automated ball cage defect inspection classification system of claim 4, wherein the determining based on the eddy current inspection results at the corresponding locations comprises:
judging the distribution and the size of vortex occurrence in the process of vortex flaw detection:
if the eddy current distribution is abnormal and the eddy current is larger than the error allowable value, a workpiece kicking command is sent out;
if the eddy current distribution is abnormal, but the eddy current size is smaller than the error allowable value, determining according to the analysis result:
if the analysis result is 'risk', a workpiece kick command is sent out;
if the analysis result is 'pass', a workpiece kick command is not sent out;
if the vortex distribution is normal, a workpiece kick command is not sent out.
6. A method of data tracking analysis of a ball cage automated defect detection classification system in accordance with claim 5, said method comprising:
counting the occurrence times of kicking commands and the sending times of detection instructions of a cleaning procedure;
and step two, respectively executing a first early warning strategy and a second early warning strategy according to the analysis result, the eddy current flaw detection result and the detection instruction sending times of the cleaning procedure, and judging whether to stop checking according to the results of the first early warning strategy and the second early warning strategy.
7. The method for data tracking analysis of a ball cage automated defect detection classification system of claim 6, wherein the first pre-warning strategy comprises:
when the continuous emission times of the kicking commands are larger than a preset value or the number of times of the kicking commands in the continuous M groups is larger than one or two of the preset proportion, an outage checking instruction is emitted;
when the continuous sending times of the cleaning procedure detection instructions are larger than a preset value or the sending times of the cleaning procedure detection instructions in the continuous M groups are larger than one or two of the preset proportion, a shutdown check instruction is sent;
wherein M is more than or equal to 5.
8. The method for data tracking analysis of a ball cage automated defect detection classification system of claim 6, wherein the second pre-warning strategy comprises:
by the formula:
calculating to obtain a workpiece continuity coefficient S u ;
When w occurs u If not less than w1, sending out a stop checking instruction;
wherein N is the detection number of the contour of the ball cage, i epsilon [1, N],α i An influence coefficient for the ith contour; g is a set value of continuously selected workpiece number, j is E [1, G];U ij The ith contour abnormal value of the jth workpiece;for the mean value of the ith contour outlier of all workpieces, w1 is a continuity reference threshold.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311502922.8A CN117554374A (en) | 2023-11-13 | 2023-11-13 | Automatic defect detection classification system for ball cage and data tracking analysis method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311502922.8A CN117554374A (en) | 2023-11-13 | 2023-11-13 | Automatic defect detection classification system for ball cage and data tracking analysis method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN117554374A true CN117554374A (en) | 2024-02-13 |
Family
ID=89819689
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311502922.8A Pending CN117554374A (en) | 2023-11-13 | 2023-11-13 | Automatic defect detection classification system for ball cage and data tracking analysis method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117554374A (en) |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR100748754B1 (en) * | 2006-07-24 | 2007-08-13 | 현대자동차주식회사 | A method for recycling cage of a constant velocity joint |
JP2011002352A (en) * | 2009-06-19 | 2011-01-06 | Jtekt Corp | Rotary device stress measuring method |
CN202974182U (en) * | 2013-01-10 | 2013-06-05 | 浙江欧迪恩传动科技股份有限公司 | Double-station eddy-current transducer detecting device of constant velocity joint |
US20170234836A1 (en) * | 2014-08-12 | 2017-08-17 | Ntn Corporation | Device for inspecting junction-type outer joint member of constant velocity universal joint |
CN215599026U (en) * | 2021-08-19 | 2022-01-21 | 江西世海龙翔科技有限公司 | Detection apparatus suitable for car ball cage production |
CN114994057A (en) * | 2022-01-07 | 2022-09-02 | 深圳思谋信息科技有限公司 | Product appearance defect detection device |
CN115184368A (en) * | 2022-09-07 | 2022-10-14 | 枣庄市胜达精密铸造有限公司 | Casting defect detection control system |
CN115356261A (en) * | 2022-07-29 | 2022-11-18 | 燕山大学 | Defect detection system and method for automobile ball cage dust cover |
CN115456955A (en) * | 2022-08-19 | 2022-12-09 | 燕山大学 | Method for detecting internal burr defect of ball cage dust cover |
CN116718616A (en) * | 2023-08-04 | 2023-09-08 | 苏州视谷视觉技术有限公司 | Machine vision detection system and detection method for flaw detection |
-
2023
- 2023-11-13 CN CN202311502922.8A patent/CN117554374A/en active Pending
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR100748754B1 (en) * | 2006-07-24 | 2007-08-13 | 현대자동차주식회사 | A method for recycling cage of a constant velocity joint |
JP2011002352A (en) * | 2009-06-19 | 2011-01-06 | Jtekt Corp | Rotary device stress measuring method |
CN202974182U (en) * | 2013-01-10 | 2013-06-05 | 浙江欧迪恩传动科技股份有限公司 | Double-station eddy-current transducer detecting device of constant velocity joint |
US20170234836A1 (en) * | 2014-08-12 | 2017-08-17 | Ntn Corporation | Device for inspecting junction-type outer joint member of constant velocity universal joint |
CN215599026U (en) * | 2021-08-19 | 2022-01-21 | 江西世海龙翔科技有限公司 | Detection apparatus suitable for car ball cage production |
CN114994057A (en) * | 2022-01-07 | 2022-09-02 | 深圳思谋信息科技有限公司 | Product appearance defect detection device |
CN115356261A (en) * | 2022-07-29 | 2022-11-18 | 燕山大学 | Defect detection system and method for automobile ball cage dust cover |
CN115456955A (en) * | 2022-08-19 | 2022-12-09 | 燕山大学 | Method for detecting internal burr defect of ball cage dust cover |
CN115184368A (en) * | 2022-09-07 | 2022-10-14 | 枣庄市胜达精密铸造有限公司 | Casting defect detection control system |
CN116718616A (en) * | 2023-08-04 | 2023-09-08 | 苏州视谷视觉技术有限公司 | Machine vision detection system and detection method for flaw detection |
Non-Patent Citations (3)
Title |
---|
方琳;戴曙光;穆平安;: "球笼内表面裂纹自动识别系统图像预处理研究", 信息技术, no. 04, 25 April 2015 (2015-04-25), pages 126 - 129 * |
蔡艳等: "基于机器视觉的球笼保持架窗口位置检测方法", 《上海汽车》, 31 October 2012 (2012-10-31), pages 38 - 46 * |
邵贵平等: "基于计算机视觉的间距检测", 《机械制造与自动化》, 6 February 2004 (2004-02-06), pages 47 - 48 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10041888B2 (en) | Surface defect inspecting device and method for hot-dip coated steel sheets | |
Satorres Martínez et al. | A machine vision system for defect characterization on transparent parts with non-plane surfaces | |
CN111127571A (en) | Small sample defect classification method and device | |
CN115184368A (en) | Casting defect detection control system | |
CN110728657A (en) | Annular bearing outer surface defect detection method based on deep learning | |
Kong et al. | Online detection of external thread surface defects based on an improved template matching algorithm | |
CN115493843B (en) | Quality monitoring method and equipment based on bearing retainer | |
JP4758823B2 (en) | Female thread molded part quality inspection device and female thread molding quality inspection method | |
CN111968100A (en) | Machine vision detection method and system | |
CN113814331B (en) | Intelligent detection method for abnormal sheet metal parts of automobile production line | |
CN117554374A (en) | Automatic defect detection classification system for ball cage and data tracking analysis method | |
CN114022441A (en) | Defect detection method for irregular hardware | |
CN111833350B (en) | Machine vision detection method and system | |
CN113780830A (en) | Steel wire part quality inspection system, method and equipment based on projection | |
Armingol et al. | Statistical pattern modeling in vision-based quality control systems | |
CN113139943A (en) | Method and system for detecting appearance defects of open circular ring workpiece and computer storage medium | |
US11327028B2 (en) | Ceramic ball automatic sorting system and method | |
CN111815575A (en) | Bearing steel ball part detection method based on machine vision | |
Meng et al. | Rubber hose surface defect detection system based on machine vision | |
Tout et al. | Automatic vision system for wheel surface inspection and monitoring | |
JP2019190911A (en) | Inspection device | |
Ye et al. | Automatic optical apparatus for inspecting bearing assembly defects | |
Zhifeng | Measuring diameter of non-threaded hex bolts based on hough transform | |
WO2017217121A1 (en) | Appearance inspection device, surface processing system, appearance inspection method, program, and projection material replacement determination method | |
Charles et al. | Survey Paper on Visual Inspection of a Mechanical Part using Machine Learning |
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 |