CN116109634B - Round appearance data detection system and method based on big data processing - Google Patents
Round appearance data detection system and method based on big data processing Download PDFInfo
- Publication number
- CN116109634B CN116109634B CN202310382261.3A CN202310382261A CN116109634B CN 116109634 B CN116109634 B CN 116109634B CN 202310382261 A CN202310382261 A CN 202310382261A CN 116109634 B CN116109634 B CN 116109634B
- Authority
- CN
- China
- Prior art keywords
- circles
- round
- unqualified
- abnormal
- information
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- 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
- G06T7/40—Analysis of texture
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
- G06T2207/10012—Stereo images
-
- 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
- G06T2207/30168—Image quality inspection
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Geometry (AREA)
- Image Processing (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention belongs to the technical field of big data detection, and discloses a round appearance data detection method based on big data processing, which comprises the following steps: acquiring a circle gray level image, identifying circle parameter information from the circle gray level image, wherein the circle parameter information comprises circle outer surface data information and outer surface texture gray level information corresponding to the circle outer surface data information, analyzing a quality evaluation prediction coefficient corresponding to the circle according to a contour error coefficient and an outer surface texture defect coefficient, and marking the corresponding circle as a qualified circle or a disqualified circle; analyzing the number of the unqualified circles, and generating a fault early warning instruction when the unqualified circles are marked as long-term abnormality; and obtaining the abnormal ratio and standard deviation of the unqualified circles with abnormal time length according to the fault early-warning instruction, and selecting a history solution from prestored history information as a reference solution of the fault early-warning instruction.
Description
Technical Field
The invention relates to the technical field of big data detection, in particular to a round appearance data detection system and method based on big data processing.
Background
The goal of big data appearance detection is to present patterns and trends of data through data visualization techniques to help users better understand the data and make meaningful decisions. The method involves the use of various techniques and tools, and is widely used in practical applications in a variety of fields such as product design, manufacturing, and the like.
Along with the development of manufacturing industry, in order to ensure that the appearance of the plastic package meets the generation requirement, the detection requirement of the corresponding plastic package is also improved; the plastic wrap inspection can now be performed on a wide variety of shapes, where the appearance of a round shape can be more difficult to inspect, as products with a round appearance often have no obvious corners or straight lines, resulting in defects or deformations on the surface that are difficult to find. In addition, the surface of a round article is generally smoother, and it is difficult to detect minute defects such as minute scratches, depressions, or foreign substances by the naked eye or by general observation means;
the existing analysis of the appearance detection data of the round objects has the following defects:
1. in the prior art, the detection analysis is carried out on the appearance standard of the round objects by mostly relying on the manual visual detection and automatic detection technology, the intelligent level is low, the cost is high, the detection result is greatly influenced by the subjective factors of people, the manual detection efficiency and the accuracy are low, and the appearance of the round objects is difficult to find due to the detection blind area of the sight of human eyes, so that the disqualification rate of products produced by plastic bags is increased;
2. In the aspect of manufacturing errors, the existing analysis of the appearance detection data of the round objects is mainly specific comparison analysis of parameter information of the round objects, but the attention degree of the appearance size errors of the round objects is not high, so that the inaccuracy of error analysis results is caused, the phenomenon that the error of the appearance detection data of the round objects is too large or too small is possibly caused, and the analysis results are not referenced, so that the judgment accuracy of the round objects is affected.
3. The existing round-shaped object fault analysis needs to carry out statistical analysis on the number of unqualified round-shaped objects on site, if maintenance personnel cannot reach the site, the cost of personnel for enterprises is increased, the efficiency is low, the error probability is high, and inconvenience is brought to the subsequent maintenance and retest of the on-site round-shaped object appearance detection equipment.
In view of this, the present invention provides a system and method for detecting the appearance of a circle based on big data processing.
Disclosure of Invention
In order to overcome the above-described drawbacks of the prior art, embodiments of the present invention provide a system and method for detecting rounded appearance data based on big data processing.
In order to achieve the above purpose, the present invention provides the following technical solutions: the method for detecting the appearance data of the round objects based on big data processing comprises the following steps:
Collecting gray level images of the circles;
constructing a three-dimensional standard model of the round object through a big data three-dimensional modeling system according to the design data of the round object, and extracting design parameter information corresponding to the round object; the design parameter information comprises design area information and outer surface texture gray scale design information;
constructing a three-dimensional actual model of the round object through a big data three-dimensional modeling system according to the acquired gray level image of the round object; identifying the parameter information of the round object according to the three-dimensional actual model of the round object, wherein the parameter information of the round object comprises the outer surface data information of the round object and the outer surface texture gray information corresponding to the outer surface data information of the round object;
extracting the data information of the outer surface of the round object from the three-dimensional solid actual model of the round object, and analyzing the data information of the outer surface of the round object to obtain the corresponding profile error coefficient of the round object:
acquiring outer surface texture gray information from a three-dimensional solid actual model of the round object, and analyzing according to the outer surface texture gray information to obtain an outer surface texture defect coefficient corresponding to the round object;
analyzing a quality evaluation prediction coefficient corresponding to the circle according to the contour error coefficient and the external surface texture defect coefficient, comparing and analyzing the quality evaluation prediction coefficient with a preset quality evaluation coefficient threshold value, and marking the corresponding circle as a qualified circle or a disqualified circle;
Analyzing the number of the unqualified circles, calculating to obtain an abnormal ratio and a standard deviation of the unqualified circles, comparing the abnormal ratio with a preset abnormal threshold value, comparing the standard deviation with a preset standard deviation threshold value, judging the abnormal type of the unqualified circles, marking the unqualified circles as sporadic abnormality or long-term abnormality, and generating a fault early warning instruction if the unqualified circles are marked as long-term abnormality;
obtaining the abnormal ratio and standard deviation of the unqualified round with the abnormal time length according to the fault early warning instruction, carrying out fault analysis on the unqualified round with the abnormal time length, and selecting a historical solution as a reference solution of the fault early warning instruction according to the abnormal ratio and standard deviation of the unqualified round with the abnormal time length and pre-stored historical information.
In a preferred embodiment, the analysis logic for generating the profile tolerance error coefficient is as follows:
the outer surface of the round object is divided into a plurality of round surface areas, n represents the serial number of the round surface areas,wherein k is the total number of the circular areas; obtaining the area of a corresponding round area in the data information of the outer surface of the round object according to the three-dimensional solid actual model of the round object >And design area information corresponding to the circular area +.>;
Area of the round areaAnd the corresponding design area information +.>Comparing, calculating the area error coefficient corresponding to the circular area by a formula>The calculation formula is as follows:
the area error coefficients corresponding to all the circular surface areas on the data information of the outer surface of the round objectConstitute the area error coefficient set->Wherein->Obtaining an average value +.f of the area error coefficient of the data information of the outer surface of the round object for the area error coefficient set>And calculating to obtain the profile tolerance error coefficient by a formulaThe specific calculation formula is as follows:
In a preferred embodiment, the analysis logic for generating the external surface texture defect coefficients is as follows:
comparing the outer surface texture gray level information corresponding to each circular area with the outer surface texture gray level design information, marking the area with the difference between the outer surface texture gray level information and the outer surface texture gray level design information as an outer surface texture defect area, dividing the outer surface texture defect area to obtain m dry defect units,wherein j is the total number of the outer surface texture defect areas in the circular area; determining the outer surface texture defect type of each defect unit according to the gray value of the outer surface texture gray information, wherein the size of the corresponding gray value area of the outer surface texture defect type in the current defect unit is the outer surface texture defect area; extracting the outer surface texture defect type and the outer surface texture defect area of each defect unit;
Calculating the defect coefficient of the texture of the outer surface by a calculation formulaThe calculation formula is as follows:
wherein:an outer surface texture defect area for the mth defective cell,/->Is an evaluation influence factor of the defect type of the texture of the outer surface in the unit area in the mth defect unit.
In a preferred embodiment, the analysis logic for generating the quality assessment prediction coefficients is as follows:
analyzing to obtain a quality evaluation prediction coefficient, wherein the specific calculation formula is as follows:
wherein:the round shape quality evaluation coefficients are expressed as round shape quality evaluation coefficients, and a1 and a2 are respectively expressed as rationality evaluation weight coefficients corresponding to preset round shape appearance and outer surface texture gray level information;
setting a quality evaluation coefficient threshold YZ1, and evaluating a quality prediction coefficientSubstituting the quality evaluation coefficient threshold value for comparison analysis; when->Less than or equal to YZ1, marking the corresponding circle as a qualified circle; when->And if the number is larger than YZ1, marking the corresponding round as a failed round at the time t.
In a preferred embodiment, the analysis logic for the number of reject circles is as follows:
acquiring the number of the unqualified circles in the t+1 time period, wherein the t+1 time period is the next time period of the t moment, detecting i circles in the time from t to t+1, wherein i is an integer greater than 1, and counting the number of the unqualified circles of the i circles; and comparing the counted number of the unqualified circles with a preset threshold value of the number of the unqualified circles for analysis to generate a normal period and an abnormal period,
Marking a period corresponding to the number of the unqualified circles being smaller than the threshold value of the number of the unqualified circles as a normal period; marking a period corresponding to the number of the unqualified circles being greater than or equal to the threshold value of the number of the unqualified circles as an abnormal period;
marking the ratio of the number of marks for generating abnormal time periods to the number of unqualified circles in t+1 time periods as abnormal ratio, establishing a data set of all the number of unqualified circles in t+1 time periods,
and calculating the standard deviation of the data set, and if the abnormal ratio is smaller than or equal to an abnormal threshold value and the standard deviation is smaller than or equal to a standard deviation threshold value, marking the unqualified circles as sporadic anomalies, otherwise marking the unqualified circles as long-term anomalies, and generating a fault early warning instruction.
In a preferred embodiment, the generation logic for selecting the history solution as the reference solution of the current fault early warning instruction according to the abnormality ratio and standard deviation of the current time-duration abnormal disqualified circles and the pre-stored history information is as follows:
marking the abnormal ratio and standard deviation of the abnormal unqualified circles with the time length as the current abnormal ratio and the current standard deviation, calculating the difference value of the abnormal ratio of the abnormal unqualified circles with the time length in the current abnormal ratio and the history time length in the pre-stored history information as a first difference value, marking the pre-stored history information corresponding to the abnormal ratio of the abnormal unqualified circles with the time length smaller than the history time length corresponding to the first threshold value as a first sample,
In the first sample, calculating a difference value of the current standard deviation and the standard deviation of the historical long-duration abnormal unqualified round object in the pre-stored historical information, marking the difference value as a second difference value, and taking a historical solution corresponding to the standard deviation of the historical long-duration abnormal unqualified round object, which is smaller than a preset second threshold value, as a reference solution of the fault early warning instruction.
A big data processing based circle appearance data detection system comprising:
the data acquisition module (1) acquires the gray level image of the round object and sends the gray level image of the round object to the round object model construction module (3);
round design parameter information extraction module (2): constructing a three-dimensional standard model of the round object through a big data three-dimensional modeling system according to the design data of the round object, and extracting design parameter information corresponding to the round object; the design parameter information comprises design area information and outer surface texture gray scale design information, and the design parameter information is sent to a data analysis module (4);
round model building module (3): constructing a three-dimensional actual model of the round object through a big data three-dimensional modeling system according to the acquired gray level image of the round object; according to the three-dimensional actual model of the round, identifying round parameter information from the three-dimensional actual model of the round, wherein the round parameter information comprises round outer surface data information and outer surface texture gray scale information corresponding to the round outer surface data information, and transmitting the round parameter information to a data analysis module (4);
A data analysis module (4) for receiving the data information of the outer surface of the round object and the gray information of the texture of the outer surface corresponding to the data information of the outer surface of the round object,
analyzing according to the data information of the outer surface of the round object to obtain a contour error coefficient corresponding to the round object:
analyzing according to the outer surface texture gray information to obtain an outer surface texture defect coefficient corresponding to the round object;
analyzing a quality evaluation prediction coefficient corresponding to the circles according to the contour error coefficient and the external surface texture defect coefficient, comparing the quality evaluation prediction coefficient with a preset quality evaluation coefficient threshold value, marking the corresponding circles as qualified circles or unqualified circles, and sending the number of the unqualified circles to a depth analysis module (5);
the depth analysis module (5) receives the number of the unqualified circles, analyzes the number of the unqualified circles, calculates to obtain an abnormal ratio and a standard deviation of the unqualified circles, compares the abnormal ratio with an abnormal threshold value, compares the standard deviation with a standard deviation threshold value, judges that the unqualified circles are marked as sporadic abnormality or long-term abnormality, generates a fault early warning instruction when the unqualified circles are marked as long-term abnormality, and sends the fault early warning instruction to the fault analysis module (7);
The history abnormal storage module (6) is used for storing history fault early warning instructions of the history unqualified circles and solving a history solving method corresponding to the fault early warning instructions;
the fault analysis module (7) is used for carrying out fault analysis on unqualified circles with long-term abnormality, matching a history fault early-warning instruction closest to the current fault early-warning instruction from the history abnormality storage module (6), and taking a history solution corresponding to the history fault early-warning instruction as a reference solution of the current fault early-warning instruction.
In a preferred embodiment, the analysis logic for generating the profile tolerance error coefficient is as follows:
the outer surface of the round object is divided into a plurality of round surface areas, n represents the serial number of the round surface areas,wherein k is the total number of the circular areas; obtaining the area of a corresponding round area in the data information of the outer surface of the round object according to the three-dimensional solid actual model of the round object>And design area information corresponding to the circular area +.>;
Area of the round areaAnd the corresponding design area information +.>Comparing, calculating the area error coefficient corresponding to the circular area by a formula >The calculation formula is as follows:
the area error coefficients corresponding to all the circular surface areas on the data information of the outer surface of the round objectConstitute the area error coefficient set->Wherein->Obtaining an average value +.f of the area error coefficient of the data information of the outer surface of the round object for the area error coefficient set>And calculating to obtain the profile tolerance error coefficient by a formulaThe specific calculation formula is as follows:
The analysis logic for generating the exterior surface texture defect coefficients is as follows:
comparing the outer surface texture gray level information corresponding to each circular area with the outer surface texture gray level design information, further extracting the corresponding outer surface texture defect areas in the circular areas, and dividing the outer surface texture defect areas to obtain m dry defect units, wherein,wherein j is the total number of the outer surface texture defect areas in the circular area; extracting the outer surface texture defect type and the outer surface texture defect area of each defect unit;
matching the external surface texture defect type of the corresponding defect unit with a preset evaluation influence factor of the unit area to which the external surface texture defect type belongs to obtain the evaluation influence factor of the external surface texture defect type in each defect unit under the unit area, and further calculating the external surface texture defect coefficient The calculation formula is as follows:
wherein:an outer surface texture defect area for the mth defective cell,/->An evaluation influence factor for the type of the outer surface texture defect per unit area in the mth defect unit,
the analysis logic to generate the quality assessment prediction coefficients is as follows:
analyzing to obtain a quality evaluation prediction coefficient, wherein the specific calculation formula is as follows:
wherein:the round shape quality evaluation coefficients are expressed as round shape quality evaluation coefficients, and a1 and a2 are respectively expressed as rationality evaluation weight coefficients corresponding to preset round shape appearance and outer surface texture gray level information;
setting a quality evaluation coefficient threshold YZ1, and evaluating a quality prediction coefficientSubstituting the quality evaluation coefficient threshold value for comparison analysis; when->Less than or equal to YZ1, marking the corresponding circle as a qualified circle; when->And if the number is larger than YZ1, marking the corresponding round as a failed round at the time t.
In a preferred embodiment, the analysis logic for the number of reject circles is as follows:
acquiring the number of the unqualified circles in the t+1 time period, wherein the t+1 time period is the next time period of the t moment, detecting i circles in the time from t to t+1, wherein i is an integer greater than 1, and counting the number of the unqualified circles of the i circles; and comparing the counted number of the unqualified circles with a preset threshold value of the number of the unqualified circles for analysis to generate a normal period and an abnormal period,
Marking a period corresponding to the number of the unqualified circles being smaller than the threshold value of the number of the unqualified circles as a normal period; marking a period corresponding to the number of the unqualified circles being greater than or equal to the threshold value of the number of the unqualified circles as an abnormal period;
marking the ratio of the number of marks for generating abnormal time periods to the number of unqualified circles in t+1 time periods as abnormal ratio, establishing a data set of all the number of unqualified circles in t+1 time periods,
calculating standard deviation of the data set, comparing the abnormal ratio with an abnormal threshold value, and comparing the standard deviation with a standard deviation threshold value;
if the abnormal ratio is smaller than or equal to the abnormal threshold value and the standard deviation is smaller than or equal to the standard deviation threshold value, marking the unqualified circles as sporadic anomalies, otherwise marking the unqualified circles as long-term anomalies, and generating fault signals.
In a preferred embodiment, the generation logic for selecting the history solution as the reference solution of the current fault early warning instruction according to the abnormality ratio and standard deviation of the current time-duration abnormal disqualified circles and the pre-stored history information is as follows:
marking the abnormal ratio and standard deviation of the abnormal unqualified circles with the time length as the current abnormal ratio and the current standard deviation, calculating the difference value of the abnormal ratio of the abnormal unqualified circles with the time length in the current abnormal ratio and the history time length in the pre-stored history information as a first difference value, marking the pre-stored history information corresponding to the abnormal ratio of the abnormal unqualified circles with the time length smaller than the history time length corresponding to the first threshold value as a first sample,
In the first sample, calculating a difference value of the current standard deviation and the standard deviation of the historical long-duration abnormal unqualified round object in the pre-stored historical information, marking the difference value as a second difference value, and taking a historical solution corresponding to the standard deviation of the historical long-duration abnormal unqualified round object, which is smaller than a preset second threshold value, as a reference solution of the fault early warning instruction.
The round appearance data detection system and method based on big data processing have the technical effects and advantages that:
1. the appearance detection of the round objects adopted by the invention is an important quality control step in the production process, so that the product can be ensured to meet the requirements, the production efficiency is improved, and the cost is reduced; firstly, determining the size of the round object to be consistent with the designed round object, and detecting surface defects of the round object, such as bulges, pits, scratches, stains and the like, so as to avoid influencing the quality and appearance of the product; the appearance of the round shape was evaluated to ensure product quality and aesthetics.
2. According to the appearance detection of the round object, a three-dimensional standard model of the round object and a three-dimensional actual model of the round object are built through a big data three-dimensional modeling system, corresponding data information of the outer surface of the round object and outer surface texture gray scale information are extracted according to the three-dimensional standard model of the round object and the three-dimensional actual model of the round object, quality evaluation prediction coefficients are obtained through calculation according to the data information of the outer surface of the round object and the outer surface texture gray scale information, and the round object corresponding to the round object is marked as a qualified round object or a disqualified round object; and analyzing according to the number of the unqualified circles, marking the unqualified circles as sporadic abnormality or long-term abnormality, generating a fault early warning instruction when the unqualified circles are marked as long-term abnormality, matching a history fault early warning instruction closest to the current fault early warning instruction from a history abnormality storage module, taking a history solution corresponding to the history fault early warning instruction as a reference solution of the current fault early warning instruction, improving the analysis accuracy of the fault early warning instruction by a tester, thereby improving the subsequent maintenance and retesting efficiency of the circle appearance detection device, improving the product quality and production efficiency, and reducing the production cost and loss.
Drawings
FIG. 1 is a schematic diagram of a circle appearance data detection system according to the present invention;
FIG. 2 is a flow chart of a method for detecting appearance data of a dome according to the present invention;
in the figure: 1. a data acquisition module; 2. a round design parameter information extraction module; 3. a round model building module; 4. a data analysis module; 5. a depth analysis module; 6. a history exception storage module; 7. and a fault analysis module.
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.
Example 1
Referring to fig. 1, the system for detecting appearance data of a round object based on big data processing according to the present embodiment includes: the device comprises a data acquisition module 1, a round object design parameter information extraction module 2, a round object model construction module 3, a data analysis module 4, a depth analysis module 5, a history abnormal storage module 6 and a fault analysis module 7;
The data acquisition module 1 acquires a round gray image; the image acquisition of the circular object image can be carried out through the omnibearing camera, and the gray level pretreatment is carried out on the circular object image, so that a circular object gray level image is generated; and sends the circle gray scale image into the circle model construction module 3.
Round design parameter information extraction module 2: constructing a three-dimensional standard model of the round object by using a big data three-dimensional modeling system according to a large amount of design data (known design data) of the round object, presenting the design data of the round object through a visualization technology, and extracting design parameter information corresponding to the round object; the design parameter information includes design area information and outer surface texture gray scale design information, and the design parameter information is sent to the data analysis module 4.
Round model building module 3: acquiring a large number of gray images of the round objects, constructing a three-dimensional actual model of the round objects by using a big data three-dimensional modeling system, and presenting the production data of the round objects by a visualization technology; according to the three-dimensional actual model of the round, the round parameter information is identified, the round parameter information comprises round outer surface data information and outer surface texture gray information corresponding to the round outer surface data information, and the round parameter information is sent to the data analysis module 4, and the big data three-dimensional modeling system such as rim Sketchup and Rhino 3D and the like is widely used in the fields of product design, industrial design and the like.
A data analysis module 4 for receiving the data information of the outer surface of the round object and the gray information of the texture of the outer surface corresponding to the data information of the outer surface of the round object,
analyzing according to the data information of the outer surface of the round object to obtain a contour error coefficient corresponding to the round object:
what needs to be explained here is: common circles comprise a cylinder, a cone and a truncated cone, and are similar to the cylinder, the cone and the truncated cone; these all have the same feature that the outer surfaces have a circle or circle-like shape, i.e. the above mentioned circle outer surface data information;
the outer surface of the round object is divided into a plurality of round surface areas, n represents the serial number of the round surface areas,wherein k is the total number of the circular areas; obtaining the area of a corresponding round area in the data information of the outer surface of the round object according to the three-dimensional solid actual model of the round object>And design area information corresponding to the circular area +.>;
Area of the round areaAnd the corresponding design area information +.>Comparing, calculating the area error coefficient corresponding to the circular area by a formula>The calculation formula is as follows:
data of the outer surface of the round objectArea error coefficients corresponding to all the circular areas in the information Constitute the area error coefficient set->Wherein->Obtaining an average value +.f of the area error coefficient of the data information of the outer surface of the round object for the area error coefficient set>And obtaining the profile tolerance error coefficient by formula calculation>The specific calculation formula is as follows:
What needs to be explained here is: when analyzing the contour degree of the data information of the outer surface of the round object, firstly dividing the data information of the outer surface of the round object into areas according to actual conditions; if the circle outer surface data information has only one circle, n=1 here, for example, balls, which is a circle when acquiring the circle gray level image; therefore, the round surface area must be a complete round surface area, for example, in the plastic packaging industry, a cylindrical plastic bottle with very high use frequency is needed to divide the round surface when analyzing the round object at the bottle mouth, wherein the data information of the outer surface of the round object contains 2 round surface areas,the first round surface area is an inner round area of the bottle mouth, and the second round surface area is an outer round area of the bottle mouth; the annular area formed between the first circular area and the second circular area is the thickness of the bottle mouth of the plastic bottle; the area of the first circular area is compared with the corresponding design area of the first circular area, so that the area error coefficient of the first circular area can be obtained ,
The area of the second circular area is compared with the corresponding design area of the second circular area, so that the area error coefficient of the second circular area can be obtained;
Then integrating the area error coefficients of the two circular areas to generate a contour error coefficient, and passing through the contour error coefficientIt can be preliminarily judged whether the outline of the data information on the outer surface of the round is complete or not.
There is also a problem in that if the data information of the outer surface of the round object cannot form a finished round surface area, an abnormality caused by all appearance factors such as an abnormality of a camera or a computer is eliminated, an image is deviated, and a finished round surface cannot be formed; the detected round object is an obvious abnormal product, the significance of the round object for detecting the appearance of the round object is low, and the artificial subjective judgment is easy to realize, so the round object is not used as an object for detecting the system;
therefore, in the analysis of the data information of the outer surface of the round object, a round area of 1 or more can be obtained, then the detection of the round object is divided into the detection of the round area, and whether the current round object is a qualified round object or not is judged by detecting whether each round area meets the preset design parameter information or not, so that the contour error coefficient is obtained Smaller, illustrate the current circle andthe smaller the difference of the preset design area information is, the more similar the design parameter information corresponding to the round is, and the more the design parameter information meets the production requirement. Conversely, the contour error coefficient ++>The larger the current round object is, the larger the difference between the current round object and the preset design area information is, and the production requirement is not met.
Analyzing according to the outer surface texture gray information to obtain an outer surface texture defect coefficient corresponding to the round object;
what needs to be explained here is: the detection of the circular area is mainly carried out by analyzing texture gray information in the circular area, for example, the surface texture gray information of a circular area which is not finished can be analyzed, and the circular area is defaulted as a part of the current circular area; comparing the surface texture gray level information corresponding to the circular area with preset outer surface texture gray level design information, selecting the outer surface texture defect type and the outer surface texture defect area corresponding to the current circular area, and then obtaining the surface texture defect coefficient.
The analysis logic for generating the exterior surface texture defect coefficients is as follows:
comparing the outer surface texture gray level information corresponding to each circular area with the outer surface texture gray level design information, marking the area with the difference between the outer surface texture gray level information and the outer surface texture gray level design information as an outer surface texture defect area, dividing the outer surface texture defect area to obtain m dry defect units, Wherein j is the total number of the outer surface texture defect areas in the circular area; determining the outer surface texture defect type of each defect unit according to the gray value of the outer surface texture gray information, wherein the size of the corresponding gray value area of the outer surface texture defect type in the current defect unit is the outer surface texture defect area; extracting the outer surface texture defect type and the outer surface texture defect area of each defect unit;
calculating the defect coefficient of the texture of the outer surface by a calculation formulaThe calculation formula is as follows:
wherein:an outer surface texture defect area for the mth defective cell,/->Is an evaluation influence factor of the defect type of the texture of the outer surface in the unit area in the mth defect unit.
What needs to be explained here is: because the outer surface texture gray scale information corresponding to each circular area is inconsistent, distinguishing abnormal texture gray scale information according to the difference of the outer surface texture gray scale information, so as to obtain the outer surface texture defect type and the outer surface texture defect area corresponding to the outer surface texture defect type; thereby obtaining the defect coefficient of the outer surface texture;
wherein the external surface texture defect coefficientThe smaller the difference between the outer surface texture gray scale information in the current round surface area and the preset outer surface texture gray scale design information is, the more similar the design parameter information corresponding to the round object is, and the more the design parameter information meets the production requirement. On the contrary, the external surface texture defect coefficient +. >The larger the current round object is, the larger the difference between the current round object and preset outer surface texture gray design information is, and the production requirement is not met.
Analyzing a quality evaluation prediction coefficient corresponding to the circle according to the contour error coefficient and the external surface texture defect coefficient, comparing and analyzing the quality evaluation prediction coefficient with a preset quality evaluation coefficient threshold value, and marking the corresponding circle as a qualified circle or a disqualified circle;
the analysis logic to generate the quality assessment prediction coefficients is as follows:
analyzing to obtain a quality evaluation prediction coefficient, wherein the specific calculation formula is as follows:
wherein:the round shape quality evaluation coefficients are expressed as round shape quality evaluation coefficients, and a1 and a2 are respectively expressed as rationality evaluation weight coefficients corresponding to preset round shape appearance and outer surface texture gray level information;
setting a quality evaluation coefficient threshold YZ1, and evaluating a quality prediction coefficientSubstituting the quality evaluation coefficient threshold value for comparison analysis; when->Less than or equal to YZ1, marking the corresponding circle as a qualified circle; when->And if the number is larger than YZ1, marking the corresponding round as a failed round at the time t.
What needs to be explained here is: quality assessment prediction coefficientThe contour error coefficient and the external surface texture defect coefficient are integrated and obtained, so that whether the whole round object meets the production requirement is evaluated; wherein the quality assessment prediction coefficient- >The smaller the design parameter information corresponding to the current round shape is, the smaller the difference between the current round shape parameter information and the design parameter information corresponding to the round shape is, and the more production requirements are met. Conversely, the quality assessment prediction coefficient +.>The larger the current round-shaped object parameter information is, the larger the difference between the current round-shaped object parameter information and the design parameter information corresponding to the round-shaped object is, and the more the current round-shaped object parameter information does not meet the production requirement.
And a quality evaluation coefficient threshold YZ1 is set as a quality evaluation prediction coefficientLess than or equal to YZ1, marking the corresponding circle as a qualified circle; when the quality evaluates the prediction coefficient->Greater than YZ1, the corresponding circle is marked as a reject circle, and then the number of reject circles is sent to the depth analysis module 5.
The depth analysis module 5 receives the number of the unqualified circles, analyzes the number of the unqualified circles, calculates and obtains an abnormal ratio and a standard deviation of the unqualified circles, compares the abnormal ratio with an abnormal threshold value, compares the standard deviation with a standard deviation threshold value, judges that the unqualified circles are marked as sporadic abnormality or long-term abnormality, generates a fault early warning instruction when the unqualified circles are marked as long-term abnormality, and sends the fault early warning instruction to the fault analysis module 7;
the analysis logic for the number of reject circles is as follows:
Acquiring the number of the unqualified circles in the t+1 time period, wherein the t+1 time period is the next time period of the t moment, detecting i circles in the time from t to t+1, wherein i is an integer greater than 1, and counting the number of the unqualified circles of the i circles; and comparing the counted number of the unqualified circles with a preset threshold value of the number of the unqualified circles for analysis to generate a normal period and an abnormal period,
marking a period corresponding to the number of the unqualified circles being smaller than the threshold value of the number of the unqualified circles as a normal period; marking a period corresponding to the number of the unqualified circles being greater than or equal to the threshold value of the number of the unqualified circles as an abnormal period;
marking the ratio of the number of marks for generating abnormal time periods to the number of unqualified circles in t+1 time periods as abnormal ratio, establishing a data set of all the number of unqualified circles in t+1 time periods,
calculating standard deviation of the data set, comparing the abnormal ratio with an abnormal threshold value, and comparing the standard deviation with a standard deviation threshold value;
if the abnormal ratio is smaller than or equal to the abnormal threshold value and the standard deviation is smaller than or equal to the standard deviation threshold value, marking the unqualified circles as sporadic anomalies, otherwise marking the unqualified circles as long-term anomalies, and generating a fault early warning instruction.
What needs to be explained here is: sporadic anomalies indicate that the probability of occurrence of a circle currently marked as failed is below a threshold, possibly randomly occurring anomalies, such as machine voltage instability resulting in a small number of circles not meeting production requirements; after a short abnormality, normal use is resumed again, and the observation can be continued without giving fault alarm; and for the long-term abnormality, the occurrence probability and the occurrence frequency of the current circle marked as the disqualified circle in a period of time are both beyond the threshold range of normal control, and the failure analysis is needed for the reason of the failure.
A history exception storage module 6, configured to store history information of the history failed circles, where the history information records an exception ratio, a standard deviation, and a corresponding solution method of the history long-term abnormal failed circles, where the solution method is the history solution method, and the history solution method is provided and stored by a maintainer and recorded in the history information.
The fault analysis module 7 is used for carrying out fault analysis on the unqualified circles with the long-time abnormal, and selecting a historical solution according to the abnormal ratio and standard deviation of the unqualified circles with the long-time abnormal and pre-stored historical information to be used as a reference solution of the fault early warning instruction; the history information closest to the current fault early warning instruction is matched from the history abnormal storage module 6 through the method, and the history solution corresponding to the history information is used as the reference solution of the current fault early warning instruction.
According to the abnormal ratio, standard deviation and pre-stored history information of the abnormal disqualified circles with the time length, a history solution is selected as a reference solution of the fault early warning instruction, and a specific fault analysis process of the fault analysis module comprises the following steps: the method comprises the steps of calling the number of unqualified circles corresponding to a fault early-warning instruction, wherein the number of the unqualified circles is larger than a threshold value of the number of the unqualified circles for the first time, screening the number of x historical unqualified circles with the smallest absolute value of the difference value between the number of the unqualified circles and the number of the unqualified circles from a historical abnormal storage module, taking the historical fault early-warning instruction corresponding to the number of the unqualified circles as a first matching sample, and x is an integer larger than 1;
screening y historical abnormal ratios with the smallest absolute value of the difference value of the abnormal ratios from the first matching sample by using the abnormal ratio corresponding to the fault early-warning instruction, and taking the historical fault early-warning instruction corresponding to the historical abnormal ratio as a second matching sample, wherein y is an integer larger than 1;
screening z historical standard deviations with the smallest absolute value of the historical standard deviation difference value from a second matching sample, wherein z historical solution methods corresponding to the z historical fault early warning instructions are used as reference solution methods of the fault early warning instructions, and z is an integer larger than 1, wherein: 。
Obtaining abnormal ratio and standard deviation of unqualified circles with abnormal time length according to a fault early warning instruction, marking the abnormal ratio and standard deviation of unqualified circles with abnormal time length as current abnormal ratio and current standard deviation, calculating difference value of the abnormal ratio of unqualified circles with abnormal time length in current abnormal ratio and pre-stored history information as first difference value, marking pre-stored history information corresponding to the abnormal ratio of unqualified circles with abnormal time length, which is smaller than a preset first threshold value, as first sample,
in the first sample, calculating a difference value of the current standard deviation and the standard deviation of the historical long-duration abnormal unqualified round object in the pre-stored historical information, marking the difference value as a second difference value, and taking a historical solution corresponding to the standard deviation of the historical long-duration abnormal unqualified round object, which is smaller than a preset second threshold value, as a reference solution of the fault early warning instruction.
Example two
Referring to fig. 2, the embodiment is not described in detail in the first description of the embodiment, and the embodiment provides a method for detecting appearance data of a round object based on big data processing, which includes the following steps:
Collecting gray level images of the circles;
constructing a three-dimensional standard model of the round object through a big data three-dimensional modeling system according to the design data of the round object, and extracting design parameter information corresponding to the round object; the design parameter information comprises design area information and outer surface texture gray scale design information;
constructing a three-dimensional actual model of the round object through a big data three-dimensional modeling system according to the acquired gray level image of the round object; identifying the parameter information of the round object according to the three-dimensional actual model of the round object, wherein the parameter information of the round object comprises the outer surface data information of the round object and the outer surface texture gray information corresponding to the outer surface data information of the round object;
extracting the data information of the outer surface of the round object from the three-dimensional solid actual model of the round object, and analyzing the data information of the outer surface of the round object to obtain the corresponding profile error coefficient of the round object:
acquiring outer surface texture gray information from a three-dimensional solid actual model of the round object, and analyzing according to the outer surface texture gray information to obtain an outer surface texture defect coefficient corresponding to the round object;
analyzing a quality evaluation prediction coefficient corresponding to the circle according to the contour error coefficient and the external surface texture defect coefficient, comparing and analyzing the quality evaluation prediction coefficient with a preset quality evaluation coefficient threshold value, and marking the corresponding circle as a qualified circle or a disqualified circle;
Analyzing the number of the unqualified circles, calculating to obtain an abnormal ratio and a standard deviation of the unqualified circles, comparing the abnormal ratio with a preset abnormal threshold value, comparing the standard deviation with a preset standard deviation threshold value, judging the abnormal type of the unqualified circles, marking the unqualified circles as sporadic abnormality or long-term abnormality, and generating a fault early warning instruction if the unqualified circles are marked as long-term abnormality;
obtaining the abnormal ratio and standard deviation of the unqualified round with the abnormal time length according to the fault early warning instruction, carrying out fault analysis on the unqualified round with the abnormal time length, and selecting a historical solution as a reference solution of the fault early warning instruction according to the abnormal ratio and standard deviation of the unqualified round with the abnormal time length and pre-stored historical information.
The analysis logic for generating the profile error coefficients is as follows:
the outer surface of the round object is divided into a plurality of round surface areas, n represents the serial number of the round surface areas,wherein k is the total number of the circular areas; obtaining the area of a corresponding round area in the data information of the outer surface of the round object according to the three-dimensional solid actual model of the round object >And design area information corresponding to the circular area +.>;
Area of the round areaAnd the corresponding design area information +.>Comparing, calculating the area error coefficient corresponding to the circular area by a formula>The calculation formula is as follows:
the area error coefficients corresponding to all the circular surface areas on the data information of the outer surface of the round objectConstitute the area error coefficient set->Wherein->Obtaining an average value +.f of the area error coefficient of the data information of the outer surface of the round object for the area error coefficient set>And calculating to obtain the profile tolerance error coefficient by a formulaThe specific calculation formula is as follows:
The analysis logic for generating the exterior surface texture defect coefficients is as follows:
comparing the outer surface texture gray level information corresponding to each circular area with the outer surface texture gray level design information, marking the area with the difference between the outer surface texture gray level information and the outer surface texture gray level design information as an outer surface texture defect area, dividing the outer surface texture defect area to obtain m dry defect units,wherein j is the total number of the outer surface texture defect areas in the circular area; determining the outer surface texture defect type of each defect unit according to the gray value of the outer surface texture gray information, wherein the size of the corresponding gray value area of the outer surface texture defect type in the current defect unit is the outer surface texture defect area; extracting the outer surface texture defect type and the outer surface texture defect area of each defect unit;
Calculating the defect coefficient of the texture of the outer surface by a calculation formulaThe calculation formula is as follows:
wherein:an outer surface texture defect area for the mth defective cell,/->Is an evaluation influence factor of the defect type of the texture of the outer surface in the unit area in the mth defect unit.
The analysis logic to generate the quality assessment prediction coefficients is as follows:
analyzing to obtain a quality evaluation prediction coefficient, wherein the specific calculation formula is as follows:
wherein:the round shape quality evaluation coefficients are expressed as round shape quality evaluation coefficients, and a1 and a2 are respectively expressed as rationality evaluation weight coefficients corresponding to preset round shape appearance and outer surface texture gray level information;
setting a quality evaluation coefficient threshold YZ1, and evaluating a quality prediction coefficientSubstituting the quality evaluation coefficient threshold value for comparison analysis; when->Less than or equal to YZ1, marking the corresponding circle as a qualified circle; when->And if the number is larger than YZ1, marking the corresponding round as a failed round at the time t.
The analysis logic for the number of reject circles is as follows:
acquiring the number of the unqualified circles in the t+1 time period, wherein the t+1 time period is the next time period of the t moment, detecting i circles in the time from t to t+1, wherein i is an integer greater than 1, and counting the number of the unqualified circles of the i circles; and comparing the counted number of the unqualified circles with a preset threshold value of the number of the unqualified circles for analysis to generate a normal period and an abnormal period,
Marking a period corresponding to the number of the unqualified circles being smaller than the threshold value of the number of the unqualified circles as a normal period; marking a period corresponding to the number of the unqualified circles being greater than or equal to the threshold value of the number of the unqualified circles as an abnormal period;
marking the ratio of the number of marks for generating abnormal time periods to the number of unqualified circles in t+1 time periods as abnormal ratio, establishing a data set of all the number of unqualified circles in t+1 time periods,
and calculating the standard deviation of the data set, and if the abnormal ratio is smaller than or equal to an abnormal threshold value and the standard deviation is smaller than or equal to a standard deviation threshold value, marking the unqualified circles as sporadic anomalies, otherwise marking the unqualified circles as long-term anomalies, and generating a fault early warning instruction.
According to the abnormal ratio, standard deviation and pre-stored history information of the current time length abnormal disqualified round, a history solution is selected as a reference solution of the current fault early warning instruction, and the generation logic is as follows:
marking the abnormal ratio and standard deviation of the abnormal unqualified circles with the time length as the current abnormal ratio and the current standard deviation, calculating the difference value of the abnormal ratio of the abnormal unqualified circles with the time length in the current abnormal ratio and the history time length in the pre-stored history information as a first difference value, marking the pre-stored history information corresponding to the abnormal ratio of the abnormal unqualified circles with the time length smaller than the history time length corresponding to the first threshold value as a first sample,
In the first sample, calculating a difference value of the current standard deviation and the standard deviation of the historical long-duration abnormal unqualified round object in the pre-stored historical information, marking the difference value as a second difference value, and taking a historical solution corresponding to the standard deviation of the historical long-duration abnormal unqualified round object, which is smaller than a preset second threshold value, as a reference solution of the fault early warning instruction.
The above formulas are all formulas with dimensionality removed and numerical calculation, the formulas are formulas with the latest real situation obtained by software simulation through collecting a large amount of data, and preset parameters and threshold selection in the formulas are set by those skilled in the art according to the actual situation.
The above embodiments may be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented in software, the above-described embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product comprises one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the processes or functions described in accordance with the embodiments of the present application are all or partially produced. The computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from one website site, computer, server, or data center to another website site, computer, server, or data center by wired (e.g., infrared, wireless, microwave, etc.). The computer readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains one or more sets of available media. The usable medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium may be a solid state disk.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, and are not repeated herein.
In the several embodiments provided in the present application, it should be understood that the disclosed systems, apparatuses, and methods may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of the units is merely one, and there may be additional divisions in actual implementation, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other forms.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed over a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in each embodiment of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a read-only memory (ROM), a random access memory (random access memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The foregoing is merely specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily think about changes or substitutions within the technical scope of the present application, and the changes and substitutions are intended to be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Finally: the foregoing description of the preferred embodiments of the invention is not intended to limit the invention to the precise form disclosed, and any such modifications, equivalents, and alternatives falling within the spirit and principles of the invention are intended to be included within the scope of the invention.
Claims (7)
1. The method for detecting the appearance data of the round objects based on big data processing is characterized by comprising the following steps:
collecting gray level images of the circles;
constructing a three-dimensional standard model of the round object through a big data three-dimensional modeling system according to the design data of the round object, and extracting design parameter information corresponding to the round object; the design parameter information comprises design area information and outer surface texture gray scale design information;
constructing a three-dimensional actual model of the round object through a big data three-dimensional modeling system according to the acquired gray level image of the round object; identifying the parameter information of the round object according to the three-dimensional actual model of the round object, wherein the parameter information of the round object comprises the outer surface data information of the round object and the outer surface texture gray information corresponding to the outer surface data information of the round object;
Extracting the data information of the outer surface of the round object from the three-dimensional solid actual model of the round object, and analyzing the data information of the outer surface of the round object to obtain the corresponding profile error coefficient of the round object:
the analysis logic for generating the profile error coefficients is as follows:
the outer surface of the round object is divided into a plurality of round surface areas, n represents the serial number of the round surface areas,wherein k is the total number of the circular areas; obtaining the area of a corresponding round area in the data information of the outer surface of the round object according to the three-dimensional solid actual model of the round object>And design area information corresponding to the circular area +.>;
Area of the round areaAnd the corresponding design area information +.>Comparing, calculating the circular area by formulationArea error coefficient corresponding to field->The calculation formula is as follows:
the area error coefficients corresponding to all the circular surface areas on the data information of the outer surface of the round objectConstitute the area error coefficient set->Wherein->Obtaining an average value +.f of the area error coefficient of the data information of the outer surface of the round object for the area error coefficient set>And obtaining the profile tolerance error coefficient by formula calculation>The specific calculation formula is as follows:
Acquiring outer surface texture gray information from a three-dimensional solid actual model of the round object, and analyzing according to the outer surface texture gray information to obtain an outer surface texture defect coefficient corresponding to the round object;
the analysis logic for generating the exterior surface texture defect coefficients is as follows:
comparing the outer surface texture gray level information corresponding to each circular area with the outer surface texture gray level design information, marking the area with the difference between the outer surface texture gray level information and the outer surface texture gray level design information as an outer surface texture defect area, dividing the outer surface texture defect area to obtain m dry defect units,wherein j is the total number of the outer surface texture defect areas in the circular area; determining the outer surface texture defect type of each defect unit according to the gray value of the outer surface texture gray information, wherein the size of the corresponding gray value area of the outer surface texture defect type in the current defect unit is the outer surface texture defect area; extracting the outer surface texture defect type and the outer surface texture defect area of each defect unit;
calculating the defect coefficient of the texture of the outer surface by a calculation formula The calculation formula is as follows:
wherein:an outer surface texture defect area for the mth defective cell,/->The method comprises the steps of evaluating influence factors for the defect types of the outer surface texture in the unit area in an mth defect unit;
analyzing a quality evaluation prediction coefficient corresponding to the circle according to the contour error coefficient and the external surface texture defect coefficient, comparing and analyzing the quality evaluation prediction coefficient with a preset quality evaluation coefficient threshold value, and marking the corresponding circle as a qualified circle or a disqualified circle;
analyzing the number of the unqualified circles, calculating to obtain an abnormal ratio and a standard deviation of the unqualified circles, comparing the abnormal ratio with a preset abnormal threshold value, comparing the standard deviation with a preset standard deviation threshold value, judging the abnormal type of the unqualified circles, marking the unqualified circles as sporadic abnormality or long-term abnormality, and generating a fault early warning instruction if the unqualified circles are marked as long-term abnormality;
obtaining the abnormal ratio and standard deviation of the unqualified round with the abnormal time length according to the fault early warning instruction, carrying out fault analysis on the unqualified round with the abnormal time length, and selecting a historical solution as a reference solution of the fault early warning instruction according to the abnormal ratio and standard deviation of the unqualified round with the abnormal time length and pre-stored historical information.
2. The method for detecting appearance data of a round object based on big data processing according to claim 1, wherein: the analysis logic to generate the quality assessment prediction coefficients is as follows:
analyzing to obtain a quality evaluation prediction coefficient, wherein the specific calculation formula is as follows:
wherein:the round shape quality evaluation coefficients are expressed as round shape quality evaluation coefficients, and a1 and a2 are respectively expressed as rationality evaluation weight coefficients corresponding to preset round shape appearance and outer surface texture gray level information;
setting a quality evaluation coefficient threshold YZ1, and evaluating a quality prediction coefficientSubstitution quality evaluation systemComparing and analyzing the number threshold value; when->Less than or equal to YZ1, marking the corresponding circle as a qualified circle; when->And if the number is larger than YZ1, marking the corresponding round as a failed round at the time t.
3. The method for detecting appearance data of a round object based on big data processing according to claim 2, wherein: the analysis logic for the number of reject circles is as follows:
acquiring the number of the unqualified circles in the t+1 time period, wherein the t+1 time period is the next time period of the t moment, detecting i circles in the time from t to t+1, wherein i is an integer greater than 1, and counting the number of the unqualified circles of the i circles; and comparing the counted number of the unqualified circles with a preset threshold value of the number of the unqualified circles for analysis to generate a normal period and an abnormal period,
Marking a period corresponding to the number of the unqualified circles being smaller than the threshold value of the number of the unqualified circles as a normal period; marking a period corresponding to the number of the unqualified circles being greater than or equal to the threshold value of the number of the unqualified circles as an abnormal period;
marking the ratio of the number of marks for generating abnormal time periods to the number of unqualified circles in t+1 time periods as abnormal ratio, establishing a data set of all the number of unqualified circles in t+1 time periods,
and calculating the standard deviation of the data set, and if the abnormal ratio is smaller than or equal to an abnormal threshold value and the standard deviation is smaller than or equal to a standard deviation threshold value, marking the unqualified circles as sporadic anomalies, otherwise marking the unqualified circles as long-term anomalies, and generating a fault early warning instruction.
4. A method for detecting appearance data of a round object based on big data processing according to claim 3, wherein: according to the abnormal ratio, standard deviation and pre-stored history information of the current time length abnormal disqualified round, a history solution is selected as a reference solution of the current fault early warning instruction, and the generation logic is as follows:
marking the abnormal ratio and standard deviation of the abnormal unqualified circles with the time length as the current abnormal ratio and the current standard deviation, calculating the difference value of the abnormal ratio of the abnormal unqualified circles with the time length in the current abnormal ratio and the history time length in the pre-stored history information as a first difference value, marking the pre-stored history information corresponding to the abnormal ratio of the abnormal unqualified circles with the time length smaller than the history time length corresponding to the first threshold value as a first sample,
In the first sample, calculating a difference value of the current standard deviation and the standard deviation of the historical long-duration abnormal unqualified round object in the pre-stored historical information, marking the difference value as a second difference value, and taking a historical solution corresponding to the standard deviation of the historical long-duration abnormal unqualified round object, which is smaller than a preset second threshold value, as a reference solution of the fault early warning instruction.
5. A big data processing based circle appearance data detection system, comprising:
the data acquisition module (1) acquires the gray level image of the round object and sends the gray level image of the round object to the round object model construction module (3);
the round object design parameter information extraction module (2) is used for constructing a three-dimensional standard model of the round object through a big data three-dimensional modeling system according to the design data of the round object, and extracting design parameter information corresponding to the round object; the design parameter information comprises design area information and outer surface texture gray scale design information, and the design parameter information is sent to a data analysis module (4);
the round object model building module (3) builds a round object three-dimensional actual model through a big data three-dimensional modeling system according to the collected round object gray level image; according to the three-dimensional actual model of the round, identifying round parameter information from the three-dimensional actual model of the round, wherein the round parameter information comprises round outer surface data information and outer surface texture gray scale information corresponding to the round outer surface data information, and transmitting the round parameter information to a data analysis module (4);
A data analysis module (4) for receiving the data information of the outer surface of the round object and the gray information of the texture of the outer surface corresponding to the data information of the outer surface of the round object,
analyzing according to the data information of the outer surface of the round object to obtain a contour error coefficient corresponding to the round object:
the analysis logic for generating the profile error coefficients is as follows:
the outer surface of the round object is divided into a plurality of round surface areas, n represents the serial number of the round surface areas,wherein k is the total number of the circular areas; obtaining the area of a corresponding round area in the data information of the outer surface of the round object according to the three-dimensional solid actual model of the round object>And design area information corresponding to the circular area +.>;
Area of the round areaAnd the corresponding design area information +.>Comparing, calculating the area error coefficient corresponding to the circular area by a formula>The calculation formula is as follows:
the round object is put intoArea error coefficients corresponding to all the circular areas on the outer surface data informationConstitute the area error coefficient set->Wherein->Obtaining an average value +.f of the area error coefficient of the data information of the outer surface of the round object for the area error coefficient set>And obtaining the profile tolerance error coefficient by formula calculation >The specific calculation formula is as follows:
The analysis logic for generating the exterior surface texture defect coefficients is as follows:
comparing the outer surface texture gray level information corresponding to each circular area with the outer surface texture gray level design information, further extracting the corresponding outer surface texture defect areas in the circular areas, and dividing the outer surface texture defect areas to obtain m dry defect units, wherein,which is provided withMiddle j is the total number of the outer surface texture defect areas in the circular area; extracting the outer surface texture defect type and the outer surface texture defect area of each defect unit;
matching the external surface texture defect type of the corresponding defect unit with a preset evaluation influence factor of the unit area to which the external surface texture defect type belongs to obtain the evaluation influence factor of the external surface texture defect type in each defect unit under the unit area, and further calculating the external surface texture defect coefficientThe calculation formula is as follows:
wherein:an outer surface texture defect area for the mth defective cell,/->An evaluation influence factor for the type of the outer surface texture defect per unit area in the mth defect unit,
analyzing according to the outer surface texture gray information to obtain an outer surface texture defect coefficient corresponding to the round object;
The analysis logic to generate the quality assessment prediction coefficients is as follows:
analyzing to obtain a quality evaluation prediction coefficient, wherein the specific calculation formula is as follows:
wherein:expressed as a circle quality evaluation coefficient, a1 and a2 are respectively expressed as a preset circle appearance and a rationality evaluation right corresponding to the outer surface texture gray level informationA weight coefficient;
setting a quality evaluation coefficient threshold YZ1, and evaluating a quality prediction coefficientSubstituting the quality evaluation coefficient threshold value for comparison analysis; when->Less than or equal to YZ1, marking the corresponding circle as a qualified circle; when->If the number is larger than YZ1, marking the corresponding round as a disqualified round at the time t;
analyzing a quality evaluation prediction coefficient corresponding to the circles according to the contour error coefficient and the external surface texture defect coefficient, comparing the quality evaluation prediction coefficient with a preset quality evaluation coefficient threshold value, marking the corresponding circles as qualified circles or unqualified circles, and sending the number of the unqualified circles to a depth analysis module (5);
the depth analysis module (5) is used for receiving the number of the unqualified circles, analyzing the number of the unqualified circles, calculating to obtain an abnormal ratio and a standard deviation of the unqualified circles, comparing the abnormal ratio with an abnormal threshold value, comparing the standard deviation with a standard deviation threshold value, judging that the unqualified circles are marked as sporadic abnormality or long-term abnormality, generating a fault early warning instruction when the unqualified circles are marked as long-term abnormality, and sending the fault early warning instruction to the fault analysis module (7);
The history abnormal storage module (6) is used for storing history fault early warning instructions of the history unqualified circles and solving a history solving method corresponding to the fault early warning instructions;
the fault analysis module (7) is used for carrying out fault analysis on unqualified circles with long-term abnormality, matching a history fault early-warning instruction closest to the current fault early-warning instruction from the history abnormality storage module (6), and taking a history solution corresponding to the history fault early-warning instruction as a reference solution of the current fault early-warning instruction.
6. The big data processing based circle appearance data detection system of claim 5, wherein: the analysis logic for the number of reject circles is as follows:
acquiring the number of the unqualified circles in the t+1 time period, wherein the t+1 time period is the next time period of the t moment, detecting i circles in the time from t to t+1, wherein i is an integer greater than 1, and counting the number of the unqualified circles of the i circles; and comparing the counted number of the unqualified circles with a preset threshold value of the number of the unqualified circles for analysis to generate a normal period and an abnormal period,
marking a period corresponding to the number of the unqualified circles being smaller than the threshold value of the number of the unqualified circles as a normal period; marking a period corresponding to the number of the unqualified circles being greater than or equal to the threshold value of the number of the unqualified circles as an abnormal period;
Marking the ratio of the number of marks for generating abnormal time periods to the number of unqualified circles in t+1 time periods as abnormal ratio, establishing a data set of all the number of unqualified circles in t+1 time periods,
calculating standard deviation of the data set, comparing the abnormal ratio with an abnormal threshold value, and comparing the standard deviation with a standard deviation threshold value;
if the abnormal ratio is smaller than or equal to the abnormal threshold value and the standard deviation is smaller than or equal to the standard deviation threshold value, marking the unqualified circles as sporadic anomalies, otherwise marking the unqualified circles as long-term anomalies, and generating fault signals.
7. The big data processing based circle appearance data detection system of claim 6, wherein: according to the abnormal ratio, standard deviation and pre-stored history information of the current time length abnormal disqualified round, a history solution is selected as a reference solution of the current fault early warning instruction, and the generation logic is as follows:
marking the abnormal ratio and standard deviation of the abnormal unqualified circles with the time length as the current abnormal ratio and the current standard deviation, calculating the difference value of the abnormal ratio of the abnormal unqualified circles with the time length in the current abnormal ratio and the history time length in the pre-stored history information as a first difference value, marking the pre-stored history information corresponding to the abnormal ratio of the abnormal unqualified circles with the time length smaller than the history time length corresponding to the first threshold value as a first sample,
In the first sample, calculating a difference value of the current standard deviation and the standard deviation of the historical long-duration abnormal unqualified round object in the pre-stored historical information, marking the difference value as a second difference value, and taking a historical solution corresponding to the standard deviation of the historical long-duration abnormal unqualified round object, which is smaller than a preset second threshold value, as a reference solution of the fault early warning instruction.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310382261.3A CN116109634B (en) | 2023-04-12 | 2023-04-12 | Round appearance data detection system and method based on big data processing |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310382261.3A CN116109634B (en) | 2023-04-12 | 2023-04-12 | Round appearance data detection system and method based on big data processing |
Publications (2)
Publication Number | Publication Date |
---|---|
CN116109634A CN116109634A (en) | 2023-05-12 |
CN116109634B true CN116109634B (en) | 2023-06-16 |
Family
ID=86258266
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310382261.3A Active CN116109634B (en) | 2023-04-12 | 2023-04-12 | Round appearance data detection system and method based on big data processing |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116109634B (en) |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114998352A (en) * | 2022-08-05 | 2022-09-02 | 武汉飞恩微电子有限公司 | Production equipment fault detection method based on image processing |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5808735A (en) * | 1993-06-17 | 1998-09-15 | Ultrapointe Corporation | Method for characterizing defects on semiconductor wafers |
CN107345915A (en) * | 2017-07-08 | 2017-11-14 | 北京林业大学 | A kind of research method of the sawn timber surface defect based on structure laser scanner technique |
CN111896556B (en) * | 2020-08-04 | 2021-05-28 | 湖南大学 | Glass bottle bottom defect detection method and system based on machine vision |
-
2023
- 2023-04-12 CN CN202310382261.3A patent/CN116109634B/en active Active
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114998352A (en) * | 2022-08-05 | 2022-09-02 | 武汉飞恩微电子有限公司 | Production equipment fault detection method based on image processing |
Also Published As
Publication number | Publication date |
---|---|
CN116109634A (en) | 2023-05-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109524139B (en) | Real-time equipment performance monitoring method based on equipment working condition change | |
CN114280076B (en) | Product surface powder spraying quality monitoring analysis management system based on machine vision | |
Hamilton et al. | Development of a novel wear detection system for wind turbine gearboxes | |
CN110334957A (en) | Production technology quality information management method, system, equipment and storage medium | |
CN113757093B (en) | Flash steam compressor unit fault diagnosis method | |
CN117270514B (en) | Production process whole-flow fault detection method based on industrial Internet of things | |
CN112580858A (en) | Equipment parameter prediction analysis method and system | |
CN115410342B (en) | Landslide hazard intelligent early warning method based on real-time monitoring of crack meter | |
CN116189407B (en) | Intelligent early warning system based on data monitoring | |
CN110969185A (en) | Equipment abnormal state detection method based on data reconstruction | |
CN116109634B (en) | Round appearance data detection system and method based on big data processing | |
CN117057644A (en) | Equipment production quality detection method and system based on characteristic matching | |
CN113962320A (en) | Underground water monitoring data processing method and device | |
CN117372966B (en) | Turntable state monitoring method based on video monitoring | |
CN114326593B (en) | Cutter life prediction system and method | |
CN116609354B (en) | Quality inspection early warning system for impregnated paper production | |
CN117312879A (en) | Injection molding machine production data supervision and early warning method, system and medium | |
CN113671911A (en) | Production condition monitoring system | |
CN109614758B (en) | Method for monitoring circular shape error with spatial correlation | |
CN117173177A (en) | Image-based metal shell optical size detection method and system | |
CN115112167A (en) | Temperature and humidity sensor early warning system and method applied to production area of cigarette factory | |
CN114596296A (en) | High-sensitivity hot-rolled steel coil end surface defect identification system and method | |
CN112184636A (en) | Detection method and detection device for fixing part in circuit board | |
CN117689269B (en) | Fault tracing method for intelligent watch processing stage | |
US11573153B2 (en) | Prediction of machine failure based on vibration trend information |
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 |