CN109146841B - Visual detection method and system for filling defects of dry battery coated paper - Google Patents
Visual detection method and system for filling defects of dry battery coated paper Download PDFInfo
- Publication number
- CN109146841B CN109146841B CN201810741952.7A CN201810741952A CN109146841B CN 109146841 B CN109146841 B CN 109146841B CN 201810741952 A CN201810741952 A CN 201810741952A CN 109146841 B CN109146841 B CN 109146841B
- Authority
- CN
- China
- Prior art keywords
- coated paper
- area
- dry battery
- paper
- image
- 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
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/12—Edge-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- Quality & Reliability (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention discloses a visual detection method and a system for filling defects of dry battery coated paper, wherein the detection method comprises the following steps: initializing a detection camera; after the synchronous signal is captured, a visual light source is started, and a detection camera collects a frame of image marked as P; performing gray linear stretching and Gaussian blur on the image P; performing edge extraction on the preprocessed image P by using a canny edge detection algorithm; positioning the position of the zinc cylinder port by using a Hough algorithm, setting an ROI (region of interest) image, and filtering an irrelevant area; dividing a coated paper area by using a binarization threshold value, and obtaining the area A of the coated paper area by using a region growing algorithm; comparing the area A with the standard value, when the pulp layer paper has defects, rejecting the dry cell, and returning to next detection of the dry cell. The system mainly comprises an industrial intelligent camera and a rejecting device. The detection method and the detection system not only can detect the coated paper in real time, but also have the advantages of high detection efficiency, low cost and the like.
Description
Technical Field
The invention relates to a dry battery preparation technology, in particular to a visual detection method and system for a filling defect of dry battery coated paper.
Background
In the production process of the dry battery, the coated paper is cut into a specific length in a mechanical cutting mode, the cut coated paper is coiled by a machine and is adsorbed on a filling machine through vacuum, and the coated paper is filled into the zinc cylinder through a cam structure of the filling machine so as to be filled with electrolyte.
In order to meet the performance requirements of charging and discharging of dry batteries, the coated paper must be matched with the zinc cylinder. However, under the influence of factors such as insufficient precision of mechanical cutting, unstable vacuum adsorption force, and stability of the filling machine, the following defects are inevitably generated in the coated paper during the filling process: pulp layer paper loss, pulp layer paper overhigh, pulp layer paper overlow, pulp layer paper dislocation, pulp layer paper corner folding and the like. When the defects exist, the charging and discharging performance of the dry battery is influenced, and the problems of liquid leakage, chronic short circuit and the like are easily caused, so that the filled pulp paper needs to be detected.
At present, most battery production enterprises adopt shutdown sampling inspection and battery final inspection to detect dry batteries, wherein the sampling inspection can cause a large amount of missing inspection, and the final inspection can waste a large amount of raw materials, so a detection device capable of synchronous detection is needed, such as a board card type machine vision system, which is based on a vision system of a Personal Computer (PC), and generally consists of a light source, an optical lens, a CCD or CMOS camera, an image acquisition card, image processing software and a PC.
However, the visual system has the following disadvantages in the use process: the detection method needs to use a plurality of cameras to detect the battery at a plurality of angles, so that the volume of the whole system is increased, the system integration level is low, the structure is complex and the production cost is high; or the dry battery is driven to rotate to perform detection, which reduces the detection efficiency and thus the production efficiency.
Disclosure of Invention
The invention aims to overcome the existing problems and provides a visual detection method for the filling defect of dry battery coated paper, which not only can detect the coated paper in a zinc cylinder in real time, but also has the advantages of simple structure, high detection efficiency, low cost and the like.
Another object of the present invention is to provide a system for implementing the above-mentioned visual inspection method.
The purpose of the invention is realized by the following technical scheme:
a visual detection method for filling defects of dry battery coated paper comprises the following steps:
s1: initializing a detection camera;
s2: after the synchronous signal is captured, a visual light source is started, and a detection camera collects a frame of image marked as P;
s3: preprocessing an image P, comprising: linear stretching of gray scale and Gaussian blur;
s4: performing edge extraction on the preprocessed image P by using a canny edge detection algorithm; positioning the position of the zinc cylinder port by using a Hough circle transformation algorithm, setting a circular ROI (region of interest) image, and filtering an irrelevant area outside the zinc cylinder;
s5: dividing a coated paper area by using a binarization threshold value according to the characteristics of coated paper and the inner wall of the zinc cylinder, and obtaining the area A of the coated paper area by using a region growing algorithm;
comparing the area a of the coated paper area with the area of the coated paper area of a standard dry cell: when the area A of the coated paper area is smaller than the standard value, judging that the coated paper is too low or the coated paper is missing, rejecting the dry battery, and returning to S2; when the area A of the coated paper area is larger than the standard value, judging that the coated paper is too high, rejecting the dry battery and returning to S2; when the coated paper area a is equal to the standard value, the coated paper is judged to be normal, and the process returns to S2.
In general, the detection camera of the present invention employs a wide-angle lens, and can capture an image of a larger area of a subject within a certain distance due to a large angle of view of the wide-angle lens. Arranging a detection camera above a conveying channel of the dry battery, extending a lens to be right above the center of the zinc barrel opening, and shooting an image downwards; the distance between the lens and the zinc cylinder opening is adjusted, so that the visual angle of the lens is covered in the zinc cylinder opening, images outside the zinc cylinder cannot be collected, and the images are convenient to process.
Under a certain focal length, the distance of the image which can be shot when the lens shoots downwards is also determined, namely the distance of the shot image can be adjusted through focusing; in the detection method of the invention, the focal length of the camera is adjusted, so that the image which can be collected by the camera is a certain distance below the opening of the zinc cylinder but does not reach the bottom of the zinc cylinder, thus the collected image is in a circular ring shape, and in the collected image, the height of an object in the vertical direction is converted into the width of a circular ring in equal proportion: specifically, from outside to inside, the opening of the zinc cylinder is in a white circular shape due to being bright, the area between the inner wall of the zinc cylinder and the coated paper is in a black circular shadow due to being dark, and the coated paper is bright as a whole and is in a white circular shape; therefore, an ROI image can be set, and then the area A of the coated paper is calculated, the size of the area A is related to the height of the coated paper in the zinc cylinder, the larger the area A is, the higher the coated paper is, and after the white circular ring presented by the coated paper is fused with the white circular ring presented by the zinc cylinder opening, the coated paper is shown to reach or be higher than the position of the zinc cylinder opening; therefore, whether the position of the pulp layer paper in the zinc cylinder is qualified or not can be known by comparing the area A of the pulp layer paper with the standard value.
In a preferred embodiment of the present invention, in step S5, a set S of outer layer edge points of the coated paper area is also obtained; searching an outer layer edge point set S of the pulp layer paper area, positioning a joint area of the pulp layer paper to obtain a joint overlapping point and a joint point of the pulp layer paper, and calculating and analyzing a joint included angle theta of the pulp layer paper; the included angle θ of the bond of the coated paper was compared with that of a standard dry cell: when the joint included angle theta is larger than the standard joint included angle, judging that the pulp layer paper is dislocated or the pulp layer paper is folded, removing the dry battery from the production line, and returning to S2; when the joint angle θ is smaller than or equal to the standard joint angle, it is determined that the coated paper is normal, and the process returns to S2. In general, the joint area is an overlapping area between the head and the tail of the coated paper, normally, the head and the tail of the coated paper are flatly attached together in the overlapping area, and the distance between the joint point and the overlapping point in the radial direction is small, so that the joint included angle is small (smaller than or equal to a standard value); when the pulp layer paper is staggered or the pulp layer paper is folded, namely the head and the tail of the pulp layer paper are not orderly attached together, the distance between the joint point and the overlapping point in the radial direction is larger, so that the obtained joint included angle is larger than a standard value. Thus, it is possible to determine whether or not the coated paper has a defect.
Preferably, when the joint included angle theta is larger than 5 degrees, the pulp layer paper is judged to be dislocated or the pulp layer paper is subjected to angle folding, the dry battery is removed from the production line, and the step returns to S2. According to samples of a large number of dry batteries, the joint included angles of the qualified dry battery pulp layer paper are all smaller than 5 degrees, the joint included angle theta of the pulp layer paper is calculated by taking the joint included angles as the standard, and if the joint included angle theta is larger than 5 degrees, the dry battery is judged to have the defect of pulp layer paper dislocation or folding.
According to a preferable scheme of the invention, when searching the outer layer edge point set S in the pulp layer paper area, the outer layer edge point set S in the pulp layer paper area is searched outwards at a certain angle from the center of a circle along the winding direction of the pulp layer paper; and if more than one edge point appears in one direction, the point is considered to be in the joint area of the pulp layer paper, one point appearing for the first time is taken as a joint overlapping point, the last point in the clockwise direction of the joint area is taken as a joint point of the pulp layer paper area, and the joint included angle theta of the pulp layer paper is calculated through the two points. The calculation method of the joint included angle theta comprises the following steps: taking a straight line (e in fig. 13) connecting the overlapping point and the joint point as an included angle edge, taking a circle tangent line (f in fig. 13) of the joint point as another included angle edge, and taking an included angle between the two included angle edges as a joint included angle theta.
In a preferred embodiment of the present invention, in step S3, the gray scale linear stretching process includes: setting an interested gray level area according to the gray level range of the dry battery pulp layer paper in an image, and performing gray level linear stretching, wherein the formula is as follows:
wherein, x in the formula1，x2Is the minimum value and the maximum value, y, of the dry battery pulp layer paper in the image gray scale range1And y2The gray value is obtained after gray linear stretching is carried out on the gray range of the coated paper. In specific application, because the gray value of the dry battery pulp paper area is darker in the whole image, the slope is set to be larger than 1, namely
Improve image contrast and can highlight dry cell pasteA laminar characteristic area.
In a preferred embodiment of the present invention, in step S3, the processing of the high-speed blur is: sampling a two-dimensional Gaussian function to obtain a Gaussian smooth template, namely:
the template is used for traversing pixel points in the 3 multiplied by 3 field in the image P, and interference noise existing in the image in the acquisition process is eliminated.
In a preferred embodiment of the present invention, in step S4, a Hough circle transformation algorithm is used to perform circle detection in the image P to find out a circle with a diameter consistent with that of the zinc cylinder opening, obtain the circle center of the circle, then position the dry battery in the image, set the circle image ROI, and filter out light factors of the dry battery. Since the diameter of the zinc cylinder mouth of the dry battery is fixed, the diameter of the zinc cylinder mouth of the R03 dry battery is 10.05mm, and the diameter of the zinc cylinder mouth of the R6 dry battery is 13.45mm, the diameter of the circle presented by the zinc cylinder mouth of the dry battery is also fixed in the image P, so that the circle consistent with the diameter of the zinc cylinder mouth can be accurately and quickly found out.
In a preferred embodiment of the present invention, in step S5, the image P is divided into two binary threshold values by using an appropriate threshold value, and the following formula is used:
in the formula, T is a threshold value, after segmentation, the pixel value of the pulp layer paper area is 255, the pulp layer paper area is white, and the pixel value of the inner wall of the dry battery zinc cylinder is 0, the dry battery zinc cylinder is black; and taking one point in the pulp layer paper area as a seed point, and carrying out area growth on white pixels in the image to obtain an outer layer edge point set S of the pulp layer paper area and the pulp layer paper area A.
In a preferred embodiment of the present invention, in step S5, the standard value is divided into an upper limit value and a lower limit value, and the coated paper area a is compared with the upper limit value and the lower limit value of the coated paper area of the standard dry battery:
when the area A of the pulp layer paper area is smaller than the standard lower limit value, judging that the pulp layer paper is too low or the pulp layer paper is missing, removing the dry battery, and returning to S2; when the area A of the coated paper area is larger than the upper limit value of the standard, judging that the coated paper is too high, rejecting the dry battery, and returning to S2; when the coated paper area a is between the upper limit value and the lower limit value, it is determined that the coated paper is normal, and the process returns to S2.
The system for realizing the visual detection method comprises an industrial intelligent camera, a visual light source, a synchronous signal triggering module and a rejecting device;
the industrial intelligent camera is one and is arranged above the conveying channel of the dry battery; the synchronous signal trigger module is arranged on a conveying mechanism of the dry battery; when the intelligent camera detects that the dry battery has defects, a rejection signal is sent to the rejection device, and the rejection device rejects the dry battery with defects out of the conveying channel.
In a preferred embodiment of the present invention, the industrial intelligent camera is matched with a 2.5MM focal length low distortion lens, and when the camera is installed, the photosensitive target surface is positioned right above the conveying channel.
In a preferred embodiment of the present invention, the visual light source is a white low-angle annular light source, and the axis of the white low-angle annular light source coincides with the axis of the lens of the industrial smart camera. Therefore, the target information and the background information in the image can be optimally separated, the difficulty of image processing algorithm segmentation and identification is greatly reduced, and the performance of the whole visual system is improved.
In a preferred embodiment of the present invention, the removing device includes a nozzle disposed on one side of the transport path and connected to the blower, so that dry batteries with defects can be removed from the transport path quickly by blowing air.
Compared with the prior art, the invention has the following beneficial effects:
1. the visual detection method of the invention has high detection precision, and can accurately detect each pulp layer paper, thereby ensuring that the dry battery has higher qualification rate and quality.
2. The method can detect various different pulp layer paper defects, including pulp layer paper loss, pulp layer paper overhigh and pulp layer paper overlow, and further can detect pulp layer paper dislocation, pulp layer paper folding angle and the like in a preferred scheme; in addition to this, some non-pulp papers can be tested for their own defects, such as: the surface of the pulp layer paper is stained with foreign matters, and the like, and can be detected and removed.
3. The visual detection method has high detection speed and can match the production speed of the dry batteries on a high-speed production line, so that each dry battery in the production process is detected without missing detection.
4. The invention can realize the functions of image acquisition, image processing, image output, signal output and the like by utilizing one industrial intelligent camera, has the advantages of high integration level, stable performance, low cost and the like, and is convenient to install and use.
5. The detection system has simple structure, and only needs to slightly change the control system of the production line of the dry battery, thereby reducing the manufacturing cost of the production line.
Drawings
Figure 1 is a schematic view of a standard dry cell battery.
FIGS. 2-6 are schematic diagrams showing the presence of a coated paper defect in a dry cell, wherein a represents coated paper and b represents a zinc can; fig. 2 is a schematic diagram of the pulp layer paper being too low, fig. 3 is a schematic diagram of the pulp layer paper in the dry battery being missing, fig. 4 is a schematic diagram of the pulp layer paper in the dry battery being too high, fig. 5 is a schematic diagram of the pulp layer paper in the dry battery being folded, and fig. 6 is a schematic diagram of the pulp layer paper in the dry battery being dislocated.
Fig. 7 is a schematic perspective view of a visual inspection system according to the present invention.
FIG. 8 is a flow chart of the visual inspection method of the present invention.
FIG. 9 is a schematic representation after zinc nozzle positioning and ROI setting.
Fig. 10 is a schematic diagram of the binary segmentation.
Fig. 11 is a schematic view of searching for a joint area of a coated paper.
Fig. 12 is a schematic view of the outer layer edge points of the coated paper.
Fig. 13 is an enlarged view of a in fig. 12, i.e., a schematic view of a joint angle of the coated paper.
Detailed Description
In order to make those skilled in the art understand the technical solutions of the present invention well, the following description of the present invention is provided with reference to the embodiments and the accompanying drawings, but the embodiments of the present invention are not limited thereto.
Referring to fig. 7, the visual detection system in this embodiment includes an industrial intelligent camera 1, a visual light source 2, a synchronization signal trigger module 3, and a rejecting device 4; wherein, the industrial intelligent camera 1 is one and is arranged above the conveying channel of the dry battery; the synchronous signal trigger module 3 is arranged on a conveying mechanism of the dry battery; when the industrial intelligent camera 1 detects that the dry battery has defects, a rejection signal is sent to the rejection device 4, and the rejection device 4 rejects the dry battery with defects out of the conveying channel.
The industrial intelligent camera 1 is matched with a 2.5MM focal length low-distortion lens 1-1, and when the camera is installed, a photosensitive target surface of the industrial intelligent camera is positioned right above the conveying channel.
The visual light source 2 is a white low-angle annular light source, and the axis of the visual light source is superposed with the axis of the lens 1-1 of the industrial intelligent camera 1. Therefore, the target information and the background information in the image can be optimally separated, the difficulty of image processing algorithm segmentation and identification is greatly reduced, and the performance of the whole visual system is improved.
The removing device 4 comprises a nozzle which is arranged on one side of the conveying channel and communicated with the air blowing device, and dry batteries with defects can be removed from the conveying channel rapidly in an air blowing mode.
Referring to fig. 8, based on the above-mentioned visual inspection system, the inspection method in this embodiment includes the following steps:
step S1: initializing an industrial intelligent camera, then waiting for a synchronous trigger signal, starting an annular light source at a timing of 60ms when the industrial intelligent camera captures a rising edge of the synchronous signal, and simultaneously acquiring a frame of image, wherein the frame of image is marked as P;
step S2: preprocessing an image P, firstly setting an interested gray level area according to the gray level range of the coated paper of the dry battery in the image, and then performing gray level linear stretching, wherein the formula is as follows:
wherein, x in the formula1，x2Is the minimum value and the maximum value, y, of the dry battery pulp layer paper in the image gray scale range1And y2The gray value is obtained after gray linear stretching is carried out on the gray range of the coated paper. In specific application, because the gray value of the dry battery pulp paper area is darker in the whole image, the slope is set to be larger than 1, namely
Improve the image contrast, can highlight the dry battery pulp paper characteristic region.
Next, the gaussian blur is processed as follows: sampling a two-dimensional Gaussian function to obtain a Gaussian smooth template, namely:
the template is used for traversing pixel points in the 3 multiplied by 3 field in the image P, and interference noise existing in the image in the acquisition process is eliminated.
Step S3: performing edge extraction on the preprocessed image P by using a canny edge detection algorithm, accurately positioning the position of the zinc cylinder opening in the image by using a Hough circle transformation algorithm according to the characteristics of extremely strong light reflection effect of the zinc cylinder opening of the dry battery and the radius of the zinc cylinder opening, and setting a circular image ROI (region of interest) outside the zinc cylinder as shown in figure 9;
step S4: according to the characteristics of the coated paper and the inner wall of the zinc cylinder, the image P is subjected to binary threshold segmentation through a proper threshold value, and a coated paper region is segmented, as shown in FIG. 10, the formula is as follows:
and in the formula, T is a threshold value, the pixel value of the pulp paper area is 255 after the image is segmented, the white color is presented, and the pixel value of the inner wall of the dry battery zinc cylinder is 0, and the black color is presented. And then, taking one point in the pulp layer paper area as a seed point, and carrying out area growth on white pixels in the image to obtain an area A and a pulp layer paper area edge point set S.
Comparing the area A of the coated paper area with the upper limit value and the lower limit value of the area of the coated paper area of a standard dry battery: when the area A of the coated paper area is smaller than the lower limit value of the standard, judging that the coated paper is too low or the coated paper is missing, rejecting the dry battery as shown in figures 2 and 3, and returning to S2; when the area A of the coated paper area is larger than the upper limit value of the standard, judging that the coated paper is too high, rejecting the dry battery as shown in figure 4, and returning to S2; when the coated paper area a is between the upper limit value and the lower limit value, it is determined that the coated paper is normal, and the process returns to S2.
In general, a wide-angle lens is used as a detection camera, and since the wide-angle lens has a short focal length and a large angle of view, a larger area image of a subject can be captured within a certain distance. Arranging a detection camera above a conveying channel of the dry battery, extending a lens to be right above the center of the zinc barrel opening, and shooting an image downwards; the distance between the lens and the zinc cylinder opening is adjusted, so that the visual angle of the lens is covered in the zinc cylinder opening, images outside the zinc cylinder cannot be collected, and the images are convenient to process.
Under a certain focal length, the distance of the image which can be shot when the lens shoots downwards is also determined, namely the distance of the shot image can be adjusted through focusing; in the detection method of the invention, the focal length of the camera is adjusted, so that the image which can be collected by the camera is a certain distance below the opening of the zinc cylinder but does not reach the bottom of the zinc cylinder, thus the collected image is in a circular ring shape, and in the collected image, the height of an object in the vertical direction is converted into the width of a circular ring in equal proportion: specifically, from outside to inside, the opening of the zinc cylinder is in a white circular shape due to being bright, the area between the inner wall of the zinc cylinder and the coated paper is in a black circular shadow due to being dark, and the coated paper is bright as a whole and is in a white circular shape; therefore, an ROI image can be set, and then the area A of the coated paper is calculated, the size of the area A is related to the height of the coated paper in the zinc cylinder, the larger the area A is, the higher the coated paper is, and after the white circular ring presented by the coated paper is fused with the white circular ring presented by the zinc cylinder opening, the coated paper is shown to reach or be higher than the position of the zinc cylinder opening; therefore, whether the position of the pulp layer paper in the zinc cylinder is qualified or not can be known by comparing the area A of the pulp layer paper with the standard value.
Step S5: in the process of winding and filling, the pulp layer paper is wound along the clock direction, and according to the rule, the outer layer edge points S of the pulp layer paper are searched outwards at intervals of 3 degrees along the clockwise direction from the center of a circle, as shown in figure 11; if more than one edge point appears in one direction, the point is considered to be in the joint area of the pulp layer paper, one point is taken as the overlapping point c of the pulp layer paper, the last point in the clockwise direction of the joint area is taken as the joint point d of the pulp layer paper area, and the joint included angle theta of the pulp layer paper is calculated through the two points. The calculation method of the joint included angle theta comprises the following steps: taking a straight line (e in figure 13) connecting the overlapping point and the joint point as an included angle edge, taking a circle tangent line (f in figure 13) of the joint point as another included angle edge, and taking an included angle between the two included angle edges as a joint included angle theta; as shown in fig. 12 and 13.
Comparing the joint included angle theta of the coated paper with that of a standard dry battery (according to a large number of samples of dry batteries, the joint included angle theta of the coated paper of the standard dry battery is less than 5 degrees, and taking the joint included angle theta as the standard): when the joint included angle theta is larger than 5 degrees, judging that the pulp layer paper is dislocated or the pulp layer paper is folded, removing the dry battery from the production line as shown in figures 5 and 6, and returning to S2; when the joint angle θ is less than or equal to 5 °, the coated paper is judged to be normal, and the process returns to S2. The basis of the above determination is: in general, the joint area is an overlapping area between the head and the tail of the coated paper, normally, the head and the tail of the coated paper are flatly attached together in the overlapping area, and the distance between the joint point d and the overlapping point c in the radial direction is small, so that the joint included angle is small (smaller than or equal to a standard value); when the pulp layer paper is staggered or the pulp layer paper is folded, namely the head and the tail of the pulp layer paper are not orderly attached together, the distance between the joint point d and the overlapping point c in the radial direction is larger, so that the obtained joint included angle is larger than a standard value. Thus, it is possible to determine whether or not the coated paper has a defect.
In step S3, since the machine vision system utilizes a white annular light source and utilizes the characteristic that the zinc nozzle of the dry battery has a good light reflection effect, the zinc nozzle can present a circle with extremely high brightness in the image, and edge extraction is performed on the preprocessed image P by using a Canny edge detection algorithm; because the diameter of the zinc cylinder opening of the dry battery is a fixed value (the diameter of the zinc cylinder opening of the R03 dry battery is 10.05mm, and the diameter of the zinc cylinder opening of the R6 dry battery is 13.45mm), in an image P, the diameter of a circle presented by the zinc cylinder opening of the dry battery is fixed, the circle detection is carried out in the image P by utilizing Hough circle transformation, the circle which is consistent with the diameter of the zinc cylinder opening is found out, the center of the circle is obtained, the position of the dry battery in the image is accurately positioned, a circular image ROI (region of interest) is set, and no light factor outside the dry battery is filtered.
The present invention is not limited to the above embodiments, and any other changes, modifications, substitutions, combinations, and simplifications which do not depart from the spirit and principle of the present invention should be construed as equivalents and are included in the scope of the present invention.
Claims (9)
1. A visual detection method for filling defects of dry battery coated paper is characterized by comprising the following steps:
s1: initializing a detection camera;
s2: after the synchronous signal is captured, a visual light source is started, and a detection camera collects a frame of image marked as P;
s3: preprocessing an image P, comprising: linear stretching of gray scale and Gaussian blur;
s4: performing edge extraction on the preprocessed image P by using a canny edge detection algorithm; positioning the position of the zinc cylinder port by using a Hough circle transformation algorithm, setting a circular ROI (region of interest) image, and filtering an irrelevant area outside the zinc cylinder;
s5: dividing a coated paper area by using a binarization threshold value according to the characteristics of coated paper and the inner wall of the zinc cylinder, and obtaining the area A of the coated paper area by using a region growing algorithm;
comparing the area a of the coated paper area with the area of the coated paper area of a standard dry cell: when the area A of the coated paper area is smaller than the standard value, judging that the coated paper is too low or the coated paper is missing, rejecting the dry battery, and returning to S2; when the area A of the coated paper area is larger than the standard value, judging that the coated paper is too high, rejecting the dry battery and returning to S2; when the area A of the coated paper is equal to the standard value, judging that the coated paper is normal, and returning to S2;
in step S5, a set S of outer layer edge points of the coated paper area is also obtained; searching an outer layer edge point set S of the pulp layer paper area, positioning a joint area of the pulp layer paper to obtain a joint overlapping point and a joint point of the pulp layer paper, and calculating and analyzing a joint included angle theta of the pulp layer paper; the included angle θ of the bond of the coated paper was compared with that of a standard dry cell: when the joint included angle theta is larger than the standard joint included angle, judging that the pulp layer paper is dislocated or the pulp layer paper is folded, removing the dry battery from the production line, and returning to S2; when the joint angle θ is smaller than or equal to the standard joint angle, it is determined that the coated paper is normal, and the process returns to S2.
2. The visual inspection method of filling defects in dry battery coated paper as set forth in claim 1, wherein when the joint angle θ is greater than 5 °, it is determined that the coated paper is misaligned or the coated paper is folded, and the dry battery is removed from the production line and the process returns to S2.
3. The visual inspection method of filling defects in dry battery coated paper as claimed in claim 1, wherein in searching the set S of outer layer edge points of the coated paper area, the set S of outer layer edge points of the coated paper area is searched outward at regular intervals from the center of the circle in the direction in which the coated paper is wound; if more than one edge point appears in one direction, the point is considered to be in the joint area of the coated paper, one point appearing for the first time is taken as a joint overlapping point, the last point in the clockwise direction of the joint area is taken as a joint point of the coated paper area, and the joint included angle theta of the coated paper is calculated through the two points.
4. The visual inspection method of filling defects in dry battery coated paper as set forth in claim 1, wherein in step S3, the gray scale linear stretching process is: setting an interested gray level area according to the gray level range of the dry battery pulp layer paper in an image, and performing gray level linear stretching, wherein the formula is as follows:
wherein, x in the formula1，x2Is the minimum value and the maximum value, y, of the dry battery pulp layer paper in the image gray scale range1And y2The gray value is obtained after gray linear stretching is carried out on the gray range of the coated paper.
5. The visual inspection method of filling defects in dry battery coated paper as set forth in claim 1, wherein in step S3, the high-speed blurring process is: sampling a two-dimensional Gaussian function to obtain a Gaussian smooth template, namely:
the template is used for traversing pixel points in the 3 multiplied by 3 field in the image P, and interference noise existing in the image in the acquisition process is eliminated.
6. The visual inspection method of filling defects in dry battery coated paper as claimed in claim 1, wherein in step S4, a Hough circle transform algorithm is used to perform circle detection in the image P to find out a circle with a diameter consistent with that of the zinc cylinder opening, obtain the center of the circle, then locate the position of the dry battery in the image, set the circle ROI image, and filter out light factors outside the dry battery.
7. The visual inspection method of filling defects in dry battery coated paper as claimed in claim 1, wherein in step S5, the image P is divided into coated paper regions by binary threshold segmentation using appropriate threshold values, and the formula is as follows:
in the formula, T is a threshold value, after segmentation, the pixel value of the pulp layer paper area is 255, the pulp layer paper area is white, and the pixel value of the inner wall of the dry battery zinc cylinder is 0, the dry battery zinc cylinder is black; and taking one point in the pulp layer paper area as a seed point, and carrying out area growth on white pixels in the image to obtain an outer layer edge point set S of the pulp layer paper area and the pulp layer paper area A.
8. A method of visual inspection of dry cell coated paper filling defects as claimed in claim 1 wherein in step S5, the standard values are divided into upper and lower limits and the coated paper area a is compared with the upper and lower limits of the standard dry cell coated paper area:
when the area A of the pulp layer paper area is smaller than the standard lower limit value, judging that the pulp layer paper is too low or the pulp layer paper is missing, removing the dry battery, and returning to S2; when the area A of the coated paper area is larger than the upper limit value of the standard, judging that the coated paper is too high, rejecting the dry battery, and returning to S2; when the coated paper area a is between the upper limit value and the lower limit value, it is determined that the coated paper is normal, and the process returns to S2.
9. A system for implementing the visual detection method of filling defects of dry battery pulp paper as claimed in claim 1, which is characterized by comprising an industrial intelligent camera, a visual light source, a synchronous signal triggering module and a rejecting device;
the industrial intelligent camera is one, is matched with a 2.5MM focal length low-distortion lens and is arranged above the conveying channel of the dry battery; the visual light source is a white low-angle annular light source, and the axis of the visual light source is superposed with the axis of a lens of the industrial intelligent camera; the synchronous signal trigger module is arranged on a conveying mechanism of the dry battery;
when the intelligent camera detects that the dry battery has defects, a rejection signal is sent to the rejection device, and the rejection device rejects the dry battery with defects out of the conveying channel.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810741952.7A CN109146841B (en) | 2018-07-02 | 2018-07-02 | Visual detection method and system for filling defects of dry battery coated paper |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810741952.7A CN109146841B (en) | 2018-07-02 | 2018-07-02 | Visual detection method and system for filling defects of dry battery coated paper |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109146841A CN109146841A (en) | 2019-01-04 |
| CN109146841B true CN109146841B (en) | 2021-07-02 |
Family
ID=64799946
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810741952.7A Active CN109146841B (en) | 2018-07-02 | 2018-07-02 | Visual detection method and system for filling defects of dry battery coated paper |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109146841B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110487807A (en) * | 2019-09-20 | 2019-11-22 | 广州市讯思视控科技有限公司 | A kind of dry cell zinc cylinder open defect vision inspection apparatus and method |
| CN110687130A (en) * | 2019-09-27 | 2020-01-14 | 天津大学 | Automatic detection method for inner wall defects of micro-holes |
| CN111766245B (en) * | 2020-05-18 | 2023-05-23 | 广州市讯思视控科技有限公司 | Button cell negative electrode shell defect detection method based on machine vision |
| CN112858312B (en) * | 2021-01-20 | 2022-09-02 | 上海工程技术大学 | Pen cap silicon bead installation visual detection device and method |
| KR102555250B1 (en) * | 2021-02-03 | 2023-07-14 | (주)인스케이프 | apparatus for Secondary battery appearance inspection and method for secondary battery appearance inspection |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002273309A (en) * | 2001-03-15 | 2002-09-24 | Ishikawajima Harima Heavy Ind Co Ltd | Apparatus for detecting defect in coating liquid for curtain coater |
| CN105548186A (en) * | 2016-01-15 | 2016-05-04 | 浙江野马电池有限公司 | Visual inspection method for coated paper |
| CN108037132A (en) * | 2017-12-25 | 2018-05-15 | 华南理工大学 | A kind of visual sensor system and method for the detection of dry cell pulp layer paper winding defect |
-
2018
- 2018-07-02 CN CN201810741952.7A patent/CN109146841B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002273309A (en) * | 2001-03-15 | 2002-09-24 | Ishikawajima Harima Heavy Ind Co Ltd | Apparatus for detecting defect in coating liquid for curtain coater |
| CN105548186A (en) * | 2016-01-15 | 2016-05-04 | 浙江野马电池有限公司 | Visual inspection method for coated paper |
| CN108037132A (en) * | 2017-12-25 | 2018-05-15 | 华南理工大学 | A kind of visual sensor system and method for the detection of dry cell pulp layer paper winding defect |
Non-Patent Citations (1)
| Title |
|---|
| 基于机器视觉的干电池缺陷并行检测方法研究;陈观应;《中国优秀硕士学位论文全文数据库 工程科技II辑》;20161115;正文第2.1.3-2.2.1节，第4.1节，第4.2.3节，第4.3节，第5.2-5.3节 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109146841A (en) | 2019-01-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109146841B (en) | Visual detection method and system for filling defects of dry battery coated paper | |
| CN108918543B (en) | Dynamic detection device and method for mirror surface scratch | |
| CN109454006B (en) | Detection and classification method based on device for online detection and classification of chemical fiber spindle tripping defects | |
| CN110976338B (en) | Test paper sorting system and method based on machine vision | |
| CN105957082A (en) | Printing quality on-line monitoring method based on area-array camera | |
| CN105044114A (en) | Electrolytic capacitor appearance package defect image detection system and electrolytic capacitor appearance package defect image detection method | |
| CN105894036A (en) | Image feature template matching method being applied to detection of mobile phone screen defects | |
| CN111650205B (en) | Part surface defect detection method and system based on structured light image matching | |
| CN108537772B (en) | Visual detection method for printing defects of chip resistor positive conductor | |
| CN109406527B (en) | System and method for detecting fine appearance defects of micro camera module lens | |
| CN204789357U (en) | Electrolytic capacitor outward appearance packing defect image detection system | |
| CN111260788A (en) | Power distribution cabinet switch state identification method based on binocular vision | |
| CN104952754B (en) | Silicon chip method for separating after plated film based on machine vision | |
| CN110032946A (en) | A kind of aluminium based on machine vision/aluminium blister package tablet identification and localization method | |
| CN116523921A (en) | Detection method, device and system for tab turnover condition | |
| CN114113129B (en) | Lens micro defect recognition and grabbing system and method | |
| CN116500038A (en) | Image acquisition method for detecting defect of outer diameter cylindrical surface of micro workpiece | |
| CN107742119A (en) | A kind of contour of object extraction and coalignment and method based on figure viewed from behind imaging | |
| WO2022227103A1 (en) | Intelligent detection and mark system and method for parts | |
| CN113129260B (en) | Automatic detection method and device for internal defects of lithium battery cell | |
| CN109668897A (en) | The vision detection system and its detection method of cyclic annular precision component surface micro defect | |
| CN115532620B (en) | Pulp molding product quality inspection device and method | |
| CN207397296U (en) | A kind of object identification device based on figure viewed from behind imaging | |
| CN116245793A (en) | Color ring resistor resistance value detection method based on vision | |
| CN108171691A (en) | A kind of detection method of plastic containers |
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 |