CN111412834A - Tobacco bale paper indentation data detection system and detection method thereof - Google Patents
Tobacco bale paper indentation data detection system and detection method thereof Download PDFInfo
- Publication number
- CN111412834A CN111412834A CN202010277741.XA CN202010277741A CN111412834A CN 111412834 A CN111412834 A CN 111412834A CN 202010277741 A CN202010277741 A CN 202010277741A CN 111412834 A CN111412834 A CN 111412834A
- Authority
- CN
- China
- Prior art keywords
- indentation
- data
- detection
- paper
- detected
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/02—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/02—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness
- G01B11/06—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness for measuring thickness ; e.g. of sheet material
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/14—Measuring arrangements characterised by the use of optical techniques for measuring distance or clearance between spaced objects or spaced apertures
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/24—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/26—Measuring arrangements characterised by the use of optical techniques for measuring angles or tapers; for testing the alignment of axes
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to the technical field of paper detection, in particular to a detection system and a detection method for cigarette packet paper indentation data. The detection system for cigarette packet paper indentation data comprises an acquisition module for acquiring images of indentations, a processing module for processing data, an execution module for performing transmission control on the data and a visualization module for displaying detection results; the execution module comprises an industrial control servo unit and a data storage unit, the acquisition module is used for transmitting laser to the paper to be detected in the detection area and receiving laser signals reflected by the paper to obtain a plurality of data of the detected points on the surface to be detected, the processing module is used for processing and analyzing the point cloud data file acquired by the acquisition module to obtain the outline coordinates and the indentation parameter information of the paper, and automatically comparing the obtained indentation parameters with standard parameters in a database to generate and display a detection result. The invention can quickly detect each item of indentation, compare the indentation with a standard interval and quickly output a test report.
Description
Technical Field
The invention relates to the technical field of paper detection, in particular to a detection system and a detection method for cigarette packet paper indentation data.
Background
The tobacco bale indentation line is a trace which is easy to fold and cut and is processed by utilizing the physical pressure, and the processing mode can better finish the later forming process of the tobacco bale. However, in the production process, the crease line parameter value is too high, and the cigarette packet package is not easy to form; if the value of the crease line parameter is too low, the tobacco bale is easy to explode, and the phenomena of tobacco extrusion and oblique wrapping are easily caused on the ultrahigh-speed packaging machine, so that the appearance quality of the tobacco bale is poor, and even waste products are caused. In order to meet the production requirements of an ultra-high-speed packaging machine, the stability of the values of crease line parameters of cigarette packets is guaranteed. Therefore, before the cigarette packet box is produced, cigarette packet paper indentation parameters (such as indentation lines, indentation space and contour lines) need to be detected so as to determine the optimal indentation technological parameters for production. At present, the common paper indentation detection adopts manual detection, so that the workload is large, the period is long, the detection precision is inaccurate, the efficiency is low, and the requirement of a cigarette factory cannot be met.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides the detection method of the indentation data of the cigarette paper with reliable performance, which can quickly detect each item of indentation, compare the indentation with a standard interval and quickly output an inspection report.
The technical scheme of the invention is realized as follows: the device comprises an acquisition module for acquiring images of the indentations, a processing module for processing data, an execution module for controlling data transmission and a visualization module for displaying detection results; the execution module comprises an industrial control servo unit and a data storage unit, the acquisition module is used for transmitting laser to the paper to be detected in the detection area and receiving laser signals reflected by the paper to obtain a plurality of data of the detected points on the surface to be detected, the processing module is used for processing and analyzing the point cloud data file acquired by the acquisition module to obtain indentation parameter information and contour coordinates of the paper, and automatically comparing the obtained indentation parameters with standard parameters in the database to generate and display a detection result.
The data processed and obtained by the data processing module comprises information of the length, the width, the height and the saturation of the indentation of the cigarette wrapping paper, the distance between the indentation and the indentation, and the length, the angle and the distance between the edge and the edge of the outer contour of the paper.
The visualization module comprises a display screen and a printer, and is electrically connected with the execution module.
The acquisition module is a laser profile scanning device, and the execution module is electrically connected with the acquisition module through a communication port.
A detection method for cigarette packet paper indentation data is characterized by comprising the following steps: the method comprises the following steps:
s1, carrying out longitudinal and transverse point-by-point laser emission on the cigarette packet paper in the product detection area by the acquisition module, and receiving the laser after the laser is reflected by the cigarette packet paper to be detected to obtain a series of detected point data on the surface of the cigarette packet paper to be detected;
s2, the execution module transmits the point data to the processing module through the communication port, the processing module denoises and simplifies the point cloud data, and then performs point cloud splicing storage to generate a depth map;
s3, carrying out indentation detection on the point cloud data by a processing module, setting indentation mask pictures in the transverse direction and the vertical direction, detecting a mask area by a Z-type detection method, effectively distinguishing a paper background part and an indentation part according to the smoothness degree of the point cloud data, finding out an indentation position, and measuring the length, the width, the height and the distance measurement between indentations;
s4, the processing module carries out edge detection on the point cloud data, carries out fuzzy denoising and binarization processing on the depth map, separates the background from the object to be detected, and carries out linear classification and fitting to obtain edge length, angle and distance;
s5, the processing module acquires contour coordinates and indentation parameters and stores the contour coordinates and the indentation parameters in the data storage unit, wherein the contour coordinates comprise the contour of each indentation line, and the indentation parameters comprise the length, the height and the width of each indentation line and the spacing between the indentation lines; the angle of the indentation line and the die cut line and the angle of the key position;
and S6, analyzing the data, automatically comparing the obtained parameter data with standard parameters of a database in the storage unit by the execution module, generating a detection result, and outputting a detection report by the visualization module.
Preferably, in step S2, the denoising includes removing a maximum value of the sample to be detected, removing a portion with a steep change in a background portion, and performing a low-pass filtering method to convolve the image with a 50X50 target template to obtain a blurred image and smooth the target template.
In the step S3, the indentation positions are found once every several data points, the width and height of each found indentation are measured, the width and the distance between indentations are calculated by using a triangular area formula, the width of an indentation can be calculated by taking three points of two sections of indentation positions, the distance between indentations can be obtained by taking three points of two indentations, after the indentation positions of each section are obtained, the two ends of the indentation are traversed respectively, template matching is performed by using a circle, and the end points of the two ends of the indentation are found, so that the length of the indentation is obtained.
Preferably, in step S4, performing connected region detection on the obtained binary image, and removing smaller connected regions; performing edge extraction by using a Canny edge detection algorithm, performing color space conversion on the extracted edge, converting the extracted edge into an RGB color space, after the edge extraction is carried out, straight line detection is carried out based on Hough transform, smaller straight line segments are removed, classifying the rest straight line segments by using a decision tree, dividing the straight line into a horizontal line, a vertical line and a diagonal line, and the three straight lines are classified by a decision tree respectively, the straight lines are subdivided into each edge class of the products to be detected, performing RANSAC line fitting on the lines in the same class, expressing the line fitting by using a mathematical expression, combining the intersecting line expressions to form an equation set, solving to obtain an intersection point, calculating the Euclidean distance of each line in a three-dimensional space, regarding the angle of the bevel edge, a vertical line on the left side of the article to be detected is taken as a reference line, and an included angle between an oblique line and the reference line is taken as the angle of the bevel edge.
In the step S6, a database confidence interval of each indentation parameter is obtained through a plurality of actual tests to generate a qualified product parameter interval, and when the measured data is in a non-standard interval, the system is marked with different colors.
And step S3, performing three-dimensional attitude correction on the point cloud data, performing moving and rotating operation on the integral point cloud data to enable the plane of the to-be-detected object to be parallel to the XY plane, translating one vertex of the minimum rectangular bounding box to the coordinate origin position, performing triangularization processing to obtain three-dimensional visual data, and performing three-dimensional display through a visual tool kit.
The invention solves the defects in the background technology and has the following beneficial effects:
the invention can quickly detect each item of indentation, compare the indentation with standard parameters and quickly issue a test report. The method has the advantages that through massive data acquisition of the paper to be detected, denoising, simplification and three-dimensional reconstruction, accurate indentation parameter values are obtained, and compared with manual detection, the accuracy is higher; through comparing indentation parameter value and the standard value of built-in database, generate the measuring report fast, response speed is fast, and the result is accurate, has satisfied cigarette enterprise's production inspection demand. In addition, the invention also provides a three-dimensional reconstruction model of the detection product for the user, so that the user can conveniently observe the conditions of other undetected parts and manually measure the length and angle parameters of the interested part.
Drawings
FIG. 1 is a schematic block diagram of a detection system of the present invention;
FIG. 2 is a flow chart of the detection method of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The embodiment of the detection system for cigarette packet paper indentation data shown in fig. 1 comprises a laser profile scanning device for collecting images of indentations, a processing module for processing data, an execution module for performing transmission control on data, a display screen for displaying detection results, and a printer. The execution module comprises an industrial control servo unit and a data storage unit, the laser contour scanning device is used for emitting laser to the paper to be detected in the detection area and receiving laser signals reflected by the paper to obtain a plurality of data of the detected points on the surface to be detected, the processing module can be used for processing and analyzing the point cloud data file collected by the collection module to obtain contour coordinates and indentation parameter information of the paper, automatically comparing the obtained indentation parameters with standard parameters in a database, and generating and displaying detection results on a display screen and a printer.
As shown in fig. 2, the method for detecting cigarette packet paper indentation data comprises the following steps:
s1, the acquisition module emits laser to the cigarette packet paper in the product detection area in a longitudinal and transverse point-by-point mode, and the laser is received after being reflected by the cigarette packet paper to be detected, so that a series of detected point data on the surface of the cigarette packet paper to be detected are obtained. The datum distance of a sensing head of the data acquisition module is 20mm, the sensing head triggering mode is triggered by an encoder, a single pen scans a 260mmX8mm area region, 10000X800 data points can be acquired, the minimum outline data interval is 0.01mm, and the error of the acquisition height of the equipment is less than 0.02 mm. The scanning path is planned by the control system, the scanning process is carried out according to an S-shaped route, every two data are overlapped and scanned for 1mm, and the operation is repeated.
And S2, the execution module transmits the point data to the processing module through the communication port, the processing module denoises and simplifies the point cloud data, and then the point cloud data is spliced and stored to generate the depth map. After data are collected, the problem of data splicing is solved firstly, the system starts from a hardware system, namely, the hardware installation precision is improved as much as possible, the influence of hardware is reduced, and on the basis, the data are further processed by utilizing a data segmentation algorithm. The data obtained by scanning the sensor head needs to be rejected as invalid data caused by light and the like, the rejection method is to calibrate the maximum value of the object to be detected according to the specific scanning condition, and sampling noise is necessary as long as the maximum value is higher than the maximum value. For the processing of the background part, the part with the steepest data change can be judged to be the edge part of the detection product as long as the part is found, and then the background data is set to be zero. Because the sensing head is scanned in a turning-back mode, the point cloud data obtained by even scanning needs to be turned over, adjacent data cannot be simply and directly spliced according to correct-sequence data, the optimal splicing position between two adjacent data needs to be found, so that the data needs to be moved integrally, in order to find the optimal splicing position, a greedy method is adopted to perform a moving splicing attempt, the point cloud data is continuously subjected to a moving attempt in an interval (-200,500), and the optimal splicing point is found, so that the coincidence degree is highest. Missing portions of data due to moving data are filled with zeros, while excess portions are discarded entirely. After all the point cloud data are spliced, the integral data of the to-be-detected object are stored in a container, and a depth map is generated for subsequent measurement.
The method includes the steps of S3, carrying out indentation detection on point cloud data, setting indentation mask pictures in the horizontal direction and the vertical direction, detecting a mask area by a Z-type detection method, effectively distinguishing a paper background part and an indentation part according to the smoothness degree of the point cloud data, finding out an indentation position, measuring the length, the width and the distance between indentations of a paper background point and a point cloud of a point, obtaining a coordinate value of a point cloud of a point cloud of a point, calculating a point cloud of a point cloud of a point cloud of a point cloud of a point cloud of a point of.
S4, the processing module carries out edge detection on the data points, carries out fuzzy denoising and binarization processing on the depth map, separates the background from the to-be-detected object, and carries out linear classification and fitting to obtain edge length, angle and distance; in the indentation detection section, the length of each of the horizontal and vertical sides, the distance between each of the horizontal and vertical sides, and the angle of the oblique side need to be detected. After splicing data, a depth map is obtained, and for the obtained depth map, preprocessing is carried out by using a traditional algorithm, and the method comprises the following specific steps: firstly, carrying out fuzzy denoising and binarization processing on the binary image for separating a background from a to-be-detected product, and then carrying out connected region detection on the obtained binary image to remove a smaller connected region; and then, performing edge extraction by using a Canny edge detection algorithm, and performing color space conversion on the extracted edge to convert the extracted edge into an RGB color space. After edge extraction is carried out, straight line detection is carried out based on Hough transformation, smaller straight line segments are removed, the remaining straight line segments are classified by using a decision tree, the straight lines are divided into three types, namely a horizontal line, a vertical line and a diagonal line, because the edge of a product to be detected is obviously changed, and the straight lines can be further classified more easily after primary classification, so that the decision tree classification is carried out on the three types of straight lines respectively, the straight lines are subdivided into each edge type of the product to be detected, RANSAC straight line fitting is carried out on the straight lines in the same type, all the straight lines are fitted into one straight line, and the straight lines are represented by a mathematical expression. For the mathematical expression of each type of straight line, simultaneous equations of the intersected straight line expressions are solved to obtain intersection points, and the Euclidean distance of each straight line on the three-dimensional space can be calculated by solving all the intersection points. Regarding the angle of the bevel edge, the left vertical line of the article to be detected is taken as a reference line, and the included angle between the oblique line and the reference line is worked out to be used as the angle of the bevel edge.
S5, the processing module acquires contour coordinates and indentation parameters and stores the contour coordinates and the indentation parameters in the data storage unit, wherein the contour coordinates comprise the contour of each indentation line, and the indentation parameters comprise the length, the height and the width of each indentation line and the spacing between the indentation lines; the angle of the indentation line and the die cut line and the angle of the key position;
and S6, analyzing the data, automatically comparing the obtained parameter data with the standard parameters of the standard database in the storage unit by the execution module, generating a detection result, and outputting a detection report by the visualization module. For detection and evaluation of the length and saturation of the indentation, after a plurality of standard packages are measured, a confidence interval of each section of indentation parameter is obtained, when the system obtains the measurement data of the to-be-detected product, the to-be-detected product is displayed according to the specific numerical value, if the measurement data is in the standard interval, the numerical value is marked by blue, otherwise, the numerical value is marked by red, and finally, the system gives an evaluation grade to the to-be-detected product according to the overall condition. In the system, for the indentation of each section of detection position, three parameters are given, namely absolute height, height and width value, wherein the absolute height refers to the actual equipment detection value of the position, the height refers to the height of the highest point relative to the paper surface background, and the width is the indentation width calculated by the detection position. After multiple measurement statistical calculations, the distance interval between each indentation is obtained. The system can give three parameters for detecting the distance between every two indentations, namely the distance between two end positions and the distance between middle positions. And then obtaining a qualified product parameter interval according to a large number of actual tests, and identifying the value of the non-parameter interval by the system in different colors.
After multiple measurement statistical calculations, the distance interval between each indentation is obtained. The system can give three parameters for detecting the distance between every two indentations, namely the distance between two end positions and the distance between middle positions. The interval between each indentation is narrower, and the longest interval is 0.1939mm, the shortest interval is 0.0517mm and the average interval length is 0.1365 mm. The large range of the longest interval is mainly because the indentation on the position is located near the paper edge and the printed matter shearing position, which causes the insufficient adsorption capacity of the vacuum adsorption platform to the paper, and the whole paper has a certain inclination angle, which can not be adsorbed on the platform well, thus causing the large detection parameter interval. For the outer contour of the object to be detected, the system solves the length and the distance of the horizontal straight line part and the vertical straight line part, and only solves the angle of the inclined straight line part. Similarly, through multiple tests, the maximum interval length of the contour length is 1.2447mm, the shortest interval length is 0.0772mm, and the average interval length is 0.4872 mm; the longest interval of the profile distance is 0.8591mm, the shortest interval is 0.1383mm, and the average interval is 0.3457 mm; the maximum angle interval is 1.5668mm, the minimum angle interval is 0.1761mm, and the average angle interval is 0.6457 mm. The main reason that the longest section length of the outline has larger difference than the average section length is the same as the indentation distance detection part, the adsorption force of the adsorption platform on the edges of the paper and near the cutting area is insufficient, the deviation of the detection position is larger, the deviation of the measurement result is increased finally, the reason of the angle deviation angle is not only the reason that the adsorption platform adsorbs unevenness, the length of part of the inclined edge is shorter, the number of noise points is more, the deviation of the result detected by a straight line is larger, the deviation between the final fitting result and an actual line segment is caused, and finally, a certain error exists between the angle calculation value and the actual value. For the detection error caused by the reason of the adsorption platform, a method for increasing the density of the adsorption holes of the adsorption platform can be considered for solving the problem.
The invention preprocesses the data, namely, eliminates invalid data and processes background data, and splices and stores a plurality of point cloud data, then generates a depth map of the to-be-detected product, the data measurement is fast and the precision is high, after the measured data is obtained, the data and the parameter interval of the standard product are contrasted and analyzed, thereby obtaining an inspection report which is used for judging whether the detected product is beneficial to the use of a packaging folding machine under the current factory environment, and then the printing work of a printing factory is guided and adjusted according to the report result. The standard data is from multiple detections of a standard product, an optimal parameter interval is obtained by an interval estimation method according to multiple detection results, the confidence coefficient selected by the system is 99%, and the system can be modified according to the requirements of users. The system also provides a three-dimensional reconstruction model for the detection product of the user, so that the user can conveniently observe the conditions of other undetected parts and manually measure the length and angle parameters of the interested part.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (10)
1. A detection system for cigarette packet paper indentation data is characterized by comprising an acquisition module for carrying out image acquisition on indentations, a processing module for processing data, an execution module for carrying out transmission control on the data and a visualization module for displaying detection results; the execution module comprises an industrial control servo unit and a data storage unit, the acquisition module is used for transmitting laser to the paper to be detected in the detection area and receiving laser signals reflected by the paper to obtain a plurality of data of the detected points on the surface to be detected, the processing module is used for processing and analyzing the point cloud data file acquired by the acquisition module to obtain indentation parameter information and contour coordinates of the paper, and automatically comparing the obtained indentation parameters with standard parameters in the database to generate and display a detection result.
2. The cigarette packet paper indentation data detection system according to claim 1, wherein the data processed and obtained by the data processing module includes indentation length, width, height, saturation information of cigarette packet paper, spacing between indentations, length, angle of outer contour of paper, and distance information between edges.
3. The cigarette packet paper indentation data detection system according to claim 1, wherein the visualization module comprises a display screen and a printer, and the visualization module is electrically connected with the execution module.
4. The cigarette packet paper indentation data detection system according to claim 1, wherein the acquisition module is a laser profile scanning device, and the execution module is electrically connected with the acquisition module through a communication port.
5. A detection method for cigarette packet paper indentation data is characterized by comprising the following steps: the method comprises the following steps:
s1, carrying out longitudinal and transverse point-by-point laser emission on the cigarette packet paper in the product detection area by the acquisition module, and receiving the laser after the laser is reflected by the cigarette packet paper to be detected to obtain a series of detected point data on the surface of the cigarette packet paper to be detected;
s2, the execution module transmits the point data to the processing module through the communication port, the processing module denoises and simplifies the point cloud data, and then performs point cloud splicing storage to generate a depth map;
s3, carrying out indentation detection on the point cloud data by a processing module, setting indentation mask pictures in the transverse direction and the vertical direction, detecting a mask area by a Z-type detection method, effectively distinguishing a paper background part and an indentation part according to the smoothness degree of the point cloud data, finding out an indentation position, and measuring the length, the width, the height and the distance measurement between indentations;
s4, the processing module carries out edge detection on the point cloud data, carries out fuzzy denoising and binarization processing on the depth map, separates the background from the object to be detected, and carries out linear classification and fitting to obtain edge length, angle and distance;
s5, the processing module acquires contour coordinates and indentation parameters and stores the contour coordinates and the indentation parameters in the data storage unit, wherein the contour coordinates comprise the contour of each indentation line, and the indentation parameters comprise the length, the height and the width of each indentation line and the spacing between the indentation lines; the angle of the indentation line and the die cut line and the angle of the key position;
and S6, analyzing the data, automatically comparing the obtained parameter data with standard parameters of a database in the storage unit by the execution module, generating a detection result, and outputting a detection report by the visualization module.
6. The method for detecting cigarette packet paper indentation data as claimed in claim 5, characterized in that in step S2, the denoising includes a process of eliminating a maximum value of a sample scan to be detected, eliminating a portion where a background portion changes steeply, and convolving an image by a low-pass filtering method using a 50X50 target template to obtain a blurred image, and smoothing the target template.
7. The method for detecting cigarette packet paper indentation data as claimed in claim 5, characterized in that in step S3, indentation positions are found once every several data points, the width and height of each indentation are measured, the width and the distance between indentations are calculated by using a triangular area formula, the width of an indentation can be calculated by taking three points of two sections of indentation positions, the distance between indentations can be obtained by taking three points of two indentations, after the indentation positions of each section are obtained, the two ends of the indentation are traversed respectively, template matching is performed by using a circle, and the end points of the two ends of the indentation are found, so as to obtain the length of the indentation.
8. The method for detecting cigarette packet paper indentation data as claimed in claim 5, characterized in that in step S4, connected region detection is performed on the obtained binary image, and smaller connected regions are removed; performing edge extraction by using a Canny edge detection algorithm, performing color space conversion on the extracted edge, converting the extracted edge into an RGB color space, after the edge extraction is carried out, straight line detection is carried out based on Hough transform, smaller straight line segments are removed, classifying the rest straight line segments by using a decision tree, dividing the straight line into a horizontal line, a vertical line and a diagonal line, and the three straight lines are classified by a decision tree respectively, the straight lines are subdivided into each edge class of the products to be detected, performing RANSAC line fitting on the lines in the same class, expressing the line fitting by using a mathematical expression, combining the intersecting line expressions to form an equation set, solving to obtain an intersection point, calculating the Euclidean distance of each line in a three-dimensional space, regarding the angle of the bevel edge, a vertical line on the left side of the article to be detected is taken as a reference line, and an included angle between an oblique line and the reference line is taken as the angle of the bevel edge.
9. The method for detecting cigarette packet paper indentation data as claimed in claim 5, characterized in that in step S6, a database confidence interval of each indentation parameter is obtained through a plurality of actual tests to create a database confidence interval of each indentation parameter, a qualified product parameter interval is generated, and when the measured data is in a non-standard interval, the system is marked with different colors.
10. The method for detecting cigarette packet paper indentation data as claimed in claim 5, characterized in that in step S3, the method further comprises performing three-dimensional attitude correction on the point cloud data, performing a moving and rotating operation on the whole point cloud data to make the plane of the article to be detected parallel to the XY plane, translating one vertex of the smallest rectangular bounding box of the point cloud data to the origin of coordinates, triangularizing the point cloud data to obtain three-dimensional visualized data, and performing three-dimensional display through a visualization tool kit.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010277741.XA CN111412834B (en) | 2020-04-08 | 2020-04-08 | Method for detecting cigarette packet paper indentation data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010277741.XA CN111412834B (en) | 2020-04-08 | 2020-04-08 | Method for detecting cigarette packet paper indentation data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111412834A true CN111412834A (en) | 2020-07-14 |
| CN111412834B CN111412834B (en) | 2022-02-08 |
Family
ID=71489791
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010277741.XA Active CN111412834B (en) | 2020-04-08 | 2020-04-08 | Method for detecting cigarette packet paper indentation data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111412834B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114965093A (en) * | 2022-06-17 | 2022-08-30 | 山东德瑞***仪器有限公司 | Crease stiffness testing device for non-metal sheet |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2124154C (en) * | 1993-11-22 | 2002-07-23 | Christopher L.B. Lavelle | Dental modeling simulator |
| CN104197838A (en) * | 2014-09-19 | 2014-12-10 | 安徽中烟工业有限责任公司 | Computer vision based cigarette carton and box packing paper dimension measurement method |
| CN106932271A (en) * | 2017-03-10 | 2017-07-07 | 厦门大学 | A kind of ball indentation test impression dimension measurement method based on reverse-engineering |
| CN208704669U (en) * | 2018-06-25 | 2019-04-05 | 万能材料测试研究公司 | A kind of paper for daily use crepe structure measurement analysis device |
| CN110345874A (en) * | 2019-08-16 | 2019-10-18 | 上海创和亿电子科技发展有限公司 | A kind of new method based on mechanical vision inspection technology measurement pipe tobacco width |
-
2020
- 2020-04-08 CN CN202010277741.XA patent/CN111412834B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2124154C (en) * | 1993-11-22 | 2002-07-23 | Christopher L.B. Lavelle | Dental modeling simulator |
| CN104197838A (en) * | 2014-09-19 | 2014-12-10 | 安徽中烟工业有限责任公司 | Computer vision based cigarette carton and box packing paper dimension measurement method |
| CN106932271A (en) * | 2017-03-10 | 2017-07-07 | 厦门大学 | A kind of ball indentation test impression dimension measurement method based on reverse-engineering |
| CN208704669U (en) * | 2018-06-25 | 2019-04-05 | 万能材料测试研究公司 | A kind of paper for daily use crepe structure measurement analysis device |
| CN110345874A (en) * | 2019-08-16 | 2019-10-18 | 上海创和亿电子科技发展有限公司 | A kind of new method based on mechanical vision inspection technology measurement pipe tobacco width |
Non-Patent Citations (1)
| Title |
|---|
| 吴梅君: "基于散乱电压的烟盒结构光视觉检测***设计与研究", 《中国优秀硕士论文全文集信息科技辑》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114965093A (en) * | 2022-06-17 | 2022-08-30 | 山东德瑞***仪器有限公司 | Crease stiffness testing device for non-metal sheet |
| CN114965093B (en) * | 2022-06-17 | 2022-12-06 | 山东德瑞克仪器股份有限公司 | Crease stiffness testing device for non-metal sheet |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111412834B (en) | 2022-02-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107064170B (en) | Method for detecting profile defect of mobile phone shell | |
| CN111340797B (en) | Laser radar and binocular camera data fusion detection method and system | |
| CN112669318B (en) | Surface defect detection method, device, equipment and storage medium | |
| CN106951905A (en) | Apple identification and localization method on a kind of tree based on TOF camera | |
| CN110910350B (en) | Nut loosening detection method for wind power tower cylinder | |
| CN111412842B (en) | Method, device and system for measuring cross-sectional dimension of wall surface | |
| CN107292309B (en) | A kind of no color differnece marks character identifying method | |
| KR910017328A (en) | Position Recognition Device | |
| CN115035092A (en) | Image-based bottle detection method, device, equipment and storage medium | |
| CN111461133A (en) | Express delivery surface single item name identification method, device, equipment and storage medium | |
| CN111412834B (en) | Method for detecting cigarette packet paper indentation data | |
| CN105160754A (en) | Coin surface quality detection apparatus and coin surface quality detection method based on height measurement | |
| CN114280075A (en) | Online visual inspection system and method for surface defects of pipe parts | |
| CN113610933A (en) | Log stacking dynamic scale detecting system and method based on binocular region parallax | |
| CN117649404A (en) | Medicine packaging box quality detection method and system based on image data analysis | |
| CN116665126A (en) | Robot inspection part defect detection method and application thereof | |
| CN109622404B (en) | Automatic sorting system and method for micro-workpieces based on machine vision | |
| CN116245882A (en) | Circuit board electronic element detection method and device and computer equipment | |
| CN117147699B (en) | Medical non-woven fabric detection method and system | |
| Ilchev et al. | A Stereo Line Sensor System to High Speed Capturing of Surfaces in Color and 3D Shape. | |
| CN113793322A (en) | Method for automatically detecting magnetic material, electronic equipment and storage medium | |
| CN109785261A (en) | A kind of airborne LIDAR three-dimensional filtering method based on gray scale volume element model | |
| CN114111576B (en) | Aircraft skin gap surface difference detection method | |
| CN114951017A (en) | Online intelligent detection error reporting system for label printing | |
| CN114708243A (en) | Cigarette end face tobacco missing quantitative detection method and device based on deep learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |