CN117314894B - Method for rapidly detecting defects of watch bottom cover - Google Patents
Method for rapidly detecting defects of watch bottom cover Download PDFInfo
- Publication number
- CN117314894B CN117314894B CN202311585982.0A CN202311585982A CN117314894B CN 117314894 B CN117314894 B CN 117314894B CN 202311585982 A CN202311585982 A CN 202311585982A CN 117314894 B CN117314894 B CN 117314894B
- Authority
- CN
- China
- Prior art keywords
- abnormal pixel
- pixel points
- abnormal
- image
- gray
- 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
- 230000007547 defect Effects 0.000 title claims abstract description 114
- 238000000034 method Methods 0.000 title claims abstract description 45
- 230000002159 abnormal effect Effects 0.000 claims abstract description 156
- 230000002950 deficient Effects 0.000 claims abstract description 32
- 238000001514 detection method Methods 0.000 claims abstract description 24
- 238000012216 screening Methods 0.000 claims abstract description 7
- 238000004364 calculation method Methods 0.000 claims description 11
- 238000009826 distribution Methods 0.000 claims description 6
- 238000012545 processing Methods 0.000 abstract description 5
- 238000004519 manufacturing process Methods 0.000 description 8
- 238000013459 approach Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 3
- 238000005286 illumination Methods 0.000 description 3
- 239000002184 metal Substances 0.000 description 3
- 229910052751 metal Inorganic materials 0.000 description 3
- 238000013450 outlier detection Methods 0.000 description 3
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 238000011179 visual inspection Methods 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000003706 image smoothing Methods 0.000 description 1
- 229910052742 iron Inorganic materials 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Quality & Reliability (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Image Processing (AREA)
Abstract
The invention relates to the technical field of image processing, in particular to a method for rapidly detecting defects of a watch bottom cover, which comprises the following steps: acquiring a gray image of the surface of the bottom cover of the watch and marking the gray image as a gray image; obtaining abnormal pixel points in the gray level image; acquiring a plurality of neighborhood blocks according to the distance between abnormal pixel points; acquiring a first defect factor and a second defect factor of an abnormal pixel point according to the neighborhood block; obtaining the defect factors of the abnormal pixel points according to the first defect factors of the abnormal pixel points and the second defect factors of the abnormal pixel points; obtaining the nearest distance between the new abnormal pixel points according to the defect factors of the abnormal pixel points; and screening defective pixel points in the gray level image according to the nearest distance between the new abnormal pixel points. The method effectively distinguishes the noise points from the defective pixel points in the gray level image of the surface of the watch bottom cover, and ensures accurate detection of the defects of the watch bottom cover.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a method for rapidly detecting defects of a watch bottom cover.
Background
With the rapid development of economy, watches are slowly changed from the function of displaying time to a common ornament, and the emphasis is not completely on displaying time but on the added decorative value of the watch. The presence or absence of a flaw in the wristwatch becomes particularly important after the wristwatch becomes an ornament, and the presence or absence of a flaw in the wristwatch, which is the largest part of the wristwatch having a metallic outer surface, directly affects the initial impression of the consumer's watch, but the presence or absence of a flaw in the wristwatch bottom inevitably causes various flaws during the manufacturing process for various reasons.
The traditional defect detection methods such as manual detection and infrared detection detect defects on the metal surface, and because the manual detection is influenced by subjective intention of workers and surrounding environment, the detection result is easily influenced by subjective judgment of people, and secondary damage of products caused by human factors is easily caused in the manual detection process, the detection method based on the machine vision technology can effectively avoid the problems. When visual inspection is carried out, the part is not in contact with the part except the visual inspection system, so that secondary damage caused by movement of the part can be prevented, microscopic defects invisible to naked eyes can be detected, and the influence of subjective factors of people on inspection results is effectively avoided.
Because the acquired image is easily interfered by external noise points and internal noise points in the process of acquiring the surface image of the watch bottom cover. In order to reduce the influence of image noise on the subsequent watch bottom cover defect detection link, a series of operations such as image smoothing and contrast enhancement are required to be performed on the acquired bottom cover image, and the detail information in the image which is usually obtained by a filtering algorithm is excessively smoothed, so that larger errors occur in the detection processing of the image in the later stage, and therefore the noise and defect areas are required to be accurately distinguished, and the subsequent processing of the image can be more accurate.
Disclosure of Invention
In order to solve the above problems, the present invention provides a method for rapidly detecting defects of a bottom cover of a wristwatch, comprising the steps of:
acquiring a watch bottom cover surface image, and carrying out graying treatment on the acquired watch bottom cover surface image to obtain a watch bottom cover surface gray image which is recorded as a gray image;
obtaining abnormal pixel points in a gray level image;
dividing the abnormal pixel points in the gray image according to the distance between the abnormal pixel points in the gray image to obtain a plurality of neighborhood blocks; acquiring a first defect factor of an abnormal pixel point according to the distribution of the abnormal pixel point in the neighborhood block; obtaining a second defect factor of the abnormal pixel point according to the gray value of the abnormal pixel point in the neighborhood block; obtaining the defect factors of the abnormal pixel points according to the first defect factors of the abnormal pixel points and the second defect factors of the abnormal pixel points;
obtaining the nearest distance between the new abnormal pixel points according to the defect factors of the abnormal pixel points; and screening defective pixel points in the gray level image according to the nearest distance between the new abnormal pixel points.
Preferably, the step of acquiring the surface image of the bottom cover of the watch comprises the following specific steps:
the camera is arranged right above the watch bottom cover, and an annular low-angle LED light source is adopted to illuminate the watch bottom cover; and acquiring an image of the surface of the bottom cover of the watch.
Preferably, the step of acquiring the abnormal pixel point in the gray image includes the following specific steps:
obtaining gradients of all pixel points in the gray level image by using a sobel operator; the pixel point with the gradient of not 0 in the gray image is marked as an abnormal pixel point.
Preferably, the dividing the abnormal pixel points in the gray image according to the distance between the abnormal pixel points in the gray image to obtain a plurality of neighborhood blocks includes the following specific steps:
recording any abnormal pixel point as a central pixel point, and counting the nearest distance between the abnormal pixel point and the central pixel pointAbnormal pixels->For the preset range threshold value, the center pixel point and the nearest +.>The abnormal pixels are marked as neighborhood blocks.
Preferably, the obtaining the first defect factor of the abnormal pixel point includes the following specific calculation formula:
in the method, in the process of the invention,indicate->A first defect factor for each abnormal pixel; />Representing the mean value of all neighborhood block distances in the gray scale image, is->Indicate->The neighborhood block distance.
Preferably, the specific obtaining steps of the neighborhood block distance are as follows:
marking all abnormal pixel points except the central pixel point in the neighborhood block as partial pixel points; and calculating the distance between the central pixel point in the neighborhood block and each partial pixel point, and recording the average value of the distances between the central pixel point in the neighborhood block and all the partial pixel points as the neighborhood block distance.
Preferably, the obtaining the second defect factor of the abnormal pixel point includes the following specific calculation formula:
in the method, in the process of the invention,indicate->A second defect factor for each abnormal pixel; />Indicate->Variance of gray values of abnormal pixel points in the neighborhood blocks; />And representing the mean value of the variances of the gray values of the abnormal pixel points in all the neighborhood blocks in the gray image.
Preferably, the specific calculation formula for obtaining the defect factor of the abnormal pixel point is as follows:
in the method, in the process of the invention,indicate->Defect factors for each abnormal pixel; />Indicate->A first defect factor for each abnormal pixel; />Indicate->And a second defect factor for each abnormal pixel.
Preferably, the obtaining the nearest distance between the new abnormal pixel points according to the defect factor of the abnormal pixel points includes the following specific steps:
for the first in gray scale imageAbnormal pixels, will be +.>The nearest distance between each abnormal pixel and other abnormal pixels is recorded as +.>Will->The defect factor of each abnormal pixel point is marked as +.>Nearest distance between new abnormal pixel pointsIs that。
Preferably, the step of screening defective pixel points in the gray scale image according to the nearest distance between new abnormal pixel points includes the following specific steps:
and detecting the abnormal pixel points in the gray image by using an LOF abnormal value detection algorithm according to the nearest distance between the new abnormal pixel points to obtain the abnormal pixel points in the gray image.
The technical scheme of the invention has the beneficial effects that: in the process of detecting the surface image of the watch bottom cover, noise points in the image are needed to be filtered firstly, so that the subsequent calculation is convenient and accurate, but some noise points randomly appear in the image; the invention provides a watch bottom cover defect rapid detection method, which comprises the steps of firstly detecting abnormal pixels in an image, outputting an optimal value by using an objective function according to the distribution characteristics of the noise and the defect area, then introducing an LOF abnormal point detection algorithm, correcting the LOF algorithm through the output value of a constructed characteristic model, screening out pixels belonging to the defect area in the image, and then determining the noise intensity in the image, thereby providing important data support for the subsequent operation of image processing, ensuring more accurate detection of the watch bottom cover defect.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of steps of a method for rapidly detecting defects of a bottom cover of a wristwatch according to the present invention.
Fig. 2 is a schematic view of the invention for capturing an image of the surface of the bottom cover of a wristwatch.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following description refers to the specific implementation, structure, characteristics and effects of the method for rapidly detecting the defects of the bottom cover of the watch according to the invention, which are provided by the invention, with reference to the accompanying drawings and the preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following specifically describes a specific scheme of the method for rapidly detecting defects of the bottom cover of the watch provided by the invention with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of steps of a method for rapidly detecting defects of a bottom cover of a wristwatch according to an embodiment of the invention is shown, the method comprises the following steps:
step S001: and acquiring a watch bottom cover surface image, and carrying out graying treatment on the acquired watch bottom cover surface image to obtain a watch bottom cover surface gray image which is recorded as a gray image.
It should be noted that, because the watch bottom cover is a metal product, the watch bottom cover has a certain light reflection property, which affects the image acquisition function of the watch bottom cover defect detection system, and the proper light source can highlight the important characteristics of the watch bottom cover part to be detected, and inhibit the secondary characteristics; uniform incidence of the light source is therefore required to avoid specular reflection. In addition, it is ensured that a sufficiently large illumination range is provided and that other external light is restricted as much as possible from entering the part inspection area.
In particular, in the process of collecting the surface image of the bottom cover of the watch, a proper light source is required to be selected for reducing the influence caused by metal reflection, and the embodiment mainly adopts an annular low-angle LED light source for illumination, as shown in fig. 2; acquiring an image of the surface of the bottom cover of the watch; and carrying out graying treatment on the watch bottom cover surface image to obtain the watch bottom cover surface gray image.
Thus, the surface image of the watch bottom cover is obtained, and the gray image of the surface of the watch bottom cover is recorded as the gray image.
Step S002: and acquiring abnormal pixel points in the gray level image.
It should be noted that, because there may be some defects or some iron filings in the mold for manufacturing the bottom cover of the watch, there may be some depressions on the surface of the bottom cover in the process of manufacturing the bottom cover of the watch, that is, there may be some defects of pits on the surface, and the defects are large or small, and the volume of the bottom cover of the watch is small in the process of manufacturing; meanwhile, defects and noise points in the watch bottom cover surface image are difficult to distinguish due to the existence of the noise points in the image acquisition process, the defects are easily identified as the noise points when the image is subjected to filtering operation, partial information of a defect area can be smoothed, and certain errors can be caused to detection of the defect area. It is necessary to acquire noise and defective pixels in the watch bottom surface image.
It should be further noted that, because the noise and the defective pixel in the gray image are different from the surrounding pixels, that is, the gray gradient of the noise and the defective pixel in the gray image is not 0, and because the illumination of the collected image is uniform and the watch bottom cover is smooth, the gray gradient of the normal pixel in the gray image is 0, the noise and the defective pixel in the gray image can be obtained by the gray gradient of the pixel in the gray image.
Specifically, the sobel operator is used to detect the gradient pixel points in the gray image to obtain the gradients of all the pixel points in the gray image, and the sobel operator is used as a well-known technology, so that the description is omitted in the embodiment; and obtaining gradients of all pixel points in the gray level image, wherein the pixel points with the gradients of not 0 in the gray level image are noise points and defective pixel points, and marking the noise points and the defective pixel points in the gray level image as abnormal pixel points.
Step S003: dividing the abnormal pixel points in the gray image according to the distance between the abnormal pixel points in the gray image to obtain a plurality of neighborhood blocks; acquiring a first defect factor of an abnormal pixel point according to the distribution of the abnormal pixel point in the neighborhood block; obtaining a second defect factor of the abnormal pixel point according to the gray value of the abnormal pixel point in the neighborhood block; and obtaining the defect factors of the abnormal pixel points according to the first defect factors of the abnormal pixel points and the second defect factors of the abnormal pixel points.
It should be noted that, in this embodiment, defect detection is performed on the watch bottom cover based on the LOF outlier detection algorithm; the LOF abnormal value detection algorithm is an abnormal value detection algorithm based on distance; the distance model of the algorithm needs to be optimized according to the degree of abnormality of the outlier. Since in the present embodiment, since the noise is uniformly distributed in the gray image, it is not concentrated in a certain area; the defect area formed by the defect pixel points in the gray level image is an area with a certain shape; and, the defect of the watch bottom cover is caused by the mold for manufacturing the watch bottom cover. The distribution of defective pixel points in the watch bottom cover is relatively concentrated, and the defective area has certain edge characteristics; the noise points have no obvious characteristics in shape, so that the noise points can be distinguished from the defective pixel points through the distribution condition of the abnormal pixel points.
Specifically, any one abnormal pixel point is marked as a central pixel point, and the distance between the abnormal pixel point and the central pixel point is counted to be nearestAbnormal pixels, said +.>A range threshold value preset for the present embodiment, < > is given>The specific value of (2) can be set according to specific conditions, and the embodiment does not do hard requirementIn this embodiment +.>To describe, the center pixel and the nearest +.>The abnormal pixel points are marked as neighborhood blocks, and all abnormal pixel points except the central pixel point in the neighborhood blocks are marked as partial pixel points. Calculating the distance between the central pixel point in the neighborhood block and each partial pixel point, and recording the average value of the distances between the central pixel point in the neighborhood block and all the partial pixel points as the neighborhood block distance, wherein a specific calculation formula is as follows:
in the method, in the process of the invention,indicate->The average value of the distances between the central pixel point and the partial pixel points in the neighborhood blocks is marked as +.>A number of neighborhood block distances; />Representing a distance operation between two pixel points; />Indicate->Coordinates of central pixel points in the neighborhood blocks; />Indicate->Intra-block->Coordinates of the partial pixels; />A range threshold value preset for the present embodiment.
It should be further noted that, because the noise points are uniformly distributed in the gray image, the nearest distances between the noise points tend to be equal, i.e. if no defective pixel point exists in the gray image, the distances between each neighborhood block in the gray image tend to be equal, and if a defective pixel point exists in the gray image, the distances between each neighborhood block in the gray image do not tend to be equal; and in general, the number of noise points in the gray level image is far greater than the number of defective pixel points, so that the first defect factor of the abnormal pixel points can be obtained through the difference between the average value of all neighborhood block distances in the gray level image and each neighborhood block in the gray level image.
The calculation formula of the first defect factor of the specific abnormal pixel point is as follows:
in the method, in the process of the invention,indicate->The first defect factor of the neighboring block, i.e.>A first defect factor for each abnormal pixel;representing the mean value of all neighborhood block distances in the gray scale image, is->Indicate->The neighborhood block distance.
It should be further noted that, when the first defect factor of the neighborhood block approaches 1, the difference between the neighborhood block distance and the whole neighborhood block distance is smaller, that is, the center pixel of the neighborhood block is less likely to be a defective pixel; when the first defect factor of the neighborhood block is less than 1, the difference between the distance of the neighborhood block and the distance of the whole neighborhood block is larger, namely the center pixel point of the neighborhood block is more likely to be a defective pixel point.
So far, a first defect factor of the abnormal pixel point is obtained.
It should be noted that, in the gray image, there are only two gray values of the noise point, and the gray value range of the defective pixel point is large; when the neighborhood blocks have no defective pixel points, the variance of gray values of abnormal pixel points in the neighborhood blocks is extremely small and approaches to 0; when the neighborhood blocks have defective pixel points, the variance of gray values of abnormal pixel points in the neighborhood blocks is large; so the second defect factor of the neighborhood block can be obtained according to the gray value variance of the abnormal pixel points in each neighborhood block.
The calculation formula of the second defect factor of the specific abnormal pixel point is as follows:
in the method, in the process of the invention,indicate->Second defect factor of the neighboring block, i.e.>A second defect factor for each abnormal pixel;indicate->Variance of gray values of abnormal pixel points in the neighborhood blocks; />And representing the mean value of the variances of the gray values of the abnormal pixel points in all the neighborhood blocks in the gray image.
It should be further noted that, when the second defect factor of the neighboring block is smaller, the gray values representing the abnormal pixel points in the neighboring block are more equal, i.e. the center pixel point of the neighboring block is less likely to be the defective pixel point; when the second defect factor of the neighborhood block is larger, the gray values representing the abnormal pixel points in the neighborhood block are not more equal, namely the center pixel point of the neighborhood block is more likely to be a defective pixel point; when the gray values of the abnormal pixels in the gray image are the same, the method can result inAppears to occurIn order to avoid the situation that the denominator of (2) is 0 +.>The case where the denominator of (2) is 0 occurs, so in +.>+1 at the denominator of (c).
So far, the second defect factor of the abnormal pixel point is obtained.
And finally, adding the first defect factor of the abnormal pixel point and the second defect factor of the abnormal pixel point to obtain the defect factor of the abnormal pixel point, wherein a specific calculation formula is as follows:
in the method, in the process of the invention,indicate->Defect factors for each abnormal pixel; />Indicate->A first defect factor for each abnormal pixel; />Indicate->And a second defect factor for each abnormal pixel.
It should be further noted that, as the result of the first defect factor of the abnormal pixel approaches to 1, the abnormal pixel is less likely to be a defective pixel; the more the result of the second defect factor of the abnormal pixel approaches 0, the less likely the abnormal pixel is to be a defective pixel, so the less likely the abnormal pixel is to be a defective pixel when the defect factor of the abnormal pixel approaches 1, whereas the less likely the defect factor of the abnormal pixel approaches 1, the more likely the abnormal pixel is to be a defective pixel.
So far, the defect factor of the abnormal pixel point is obtained.
Step S004: obtaining the nearest distance between the new abnormal pixel points according to the defect factors of the abnormal pixel points; and screening defective pixel points in the gray level image according to the nearest distance between the new abnormal pixel points.
Note that, the defect factor due to noise in the gray image tends to be 1; the defect factor of the defect pixel point is not close to 1, so that the nearest distance between the new abnormal pixel points can be obtained according to the defect factor of the abnormal pixel point, and the nearest distance between the new abnormal pixel points can distinguish the defect pixel point from the noise point in the abnormal pixel point.
It should be further noted that, in the conventional LOF outlier detection algorithm, anomaly detection is performed according to the euclidean distance between the outlier pixels, and in this embodiment, anomaly detection is performed by using the nearest distance between the new outlier pixels, so as to achieve the purpose of distinguishing the defective pixel from the noisy pixel in the outlier pixels.
Specifically, for the first in gray scale imageAbnormal pixels, will be +.>The nearest distance between each abnormal pixel and other abnormal pixels is recorded as +.>Will->The defect factor of each abnormal pixel point is marked as +.>The nearest distance between the new outlier pixels is +.>。
It should be further noted that, because the defect factor of the noise point approaches 1, the distance metric of the LOF outlier detection algorithm of the optimized noise point is basically unchanged from the original distance metric; and the defect pixel point is larger than 1 due to the defect factor, so that the distance measurement between the defect pixel point and the abnormal pixel point is exponentially increased. At this time, according to the nearest distance between the new abnormal pixels, the abnormal pixels in the gray image are detected by using the LOF abnormal value detection algorithm, so as to obtain the abnormal pixels in the gray image, and the LOF abnormal value detection algorithm is used as a known technology, so that details are not repeated in this embodiment.
When the filter is used for smoothing the image, defective pixel points in the gray image are reserved, noise points in the gray image are filtered, the problem of noise point interference can be solved, and defect information is reserved.
And extracting the defect area, recording the information such as the area, the shape and the like of the defect area for the optimization and improvement of a subsequent production line, and finally removing the dial with the production quality problem by using a mechanical arm to ensure the production quality.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.
Claims (3)
1. The method for rapidly detecting the defects of the bottom cover of the watch is characterized by comprising the following steps of:
acquiring a watch bottom cover surface image, and carrying out graying treatment on the acquired watch bottom cover surface image to obtain a watch bottom cover surface gray image which is recorded as a gray image;
obtaining abnormal pixel points in a gray level image;
dividing the abnormal pixel points in the gray image according to the distance between the abnormal pixel points in the gray image to obtain a plurality of neighborhood blocks; acquiring a first defect factor of an abnormal pixel point according to the distribution of the abnormal pixel point in the neighborhood block; obtaining a second defect factor of the abnormal pixel point according to the gray value of the abnormal pixel point in the neighborhood block; obtaining the defect factors of the abnormal pixel points according to the first defect factors of the abnormal pixel points and the second defect factors of the abnormal pixel points;
dividing the abnormal pixel points in the gray image according to the distance between the abnormal pixel points in the gray image to obtain a plurality of neighborhood blocks, wherein the method comprises the following specific steps:
recording any abnormal pixel point as a central pixel point, and counting the nearest distance between the abnormal pixel point and the central pixel pointAbnormal pixels->For the preset range threshold value, the center pixel point and the nearest +.>Marking the abnormal pixel points as neighborhood blocks;
the specific calculation formula for obtaining the first defect factor of the abnormal pixel point is as follows:
in the method, in the process of the invention,indicate->A first defect factor for each abnormal pixel; />Representing the mean value of all neighborhood block distances in the gray scale image, is->Indicate->A number of neighborhood block distances;
the specific acquisition steps of the neighborhood block distance are as follows:
marking all abnormal pixel points except the central pixel point in the neighborhood block as partial pixel points; calculating the distance between the central pixel point in the neighborhood block and each partial pixel point, and recording the average value of the distances between the central pixel point in the neighborhood block and all the partial pixel points as the neighborhood block distance;
the specific calculation formula for obtaining the second defect factor of the abnormal pixel point is as follows:
in the method, in the process of the invention,indicate->A second defect factor for each abnormal pixel; />Indicate->Variance of gray values of abnormal pixel points in the neighborhood blocks; />Representing the mean value of the variance of the gray values of the abnormal pixel points in all neighborhood blocks in the gray image;
the specific calculation formula for obtaining the defect factor of the abnormal pixel point is as follows:
in the method, in the process of the invention,indicate->Defect factors for each abnormal pixel;
obtaining the nearest distance between the new abnormal pixel points according to the defect factors of the abnormal pixel points; screening out defective pixel points in the gray level image according to the nearest distance between the new abnormal pixel points;
the method for obtaining the nearest distance between the new abnormal pixel points according to the defect factors of the abnormal pixel points comprises the following specific steps:
for the first in gray scale imageAbnormal pixels, will be +.>The nearest distance between each abnormal pixel and other abnormal pixels is recorded as +.>Will->The defect factor of each abnormal pixel point is marked as +.>The nearest distance between the new outlier pixels is +.>；
The method for screening the defective pixel points in the gray level image according to the nearest distance between the new abnormal pixel points comprises the following specific steps:
and detecting the abnormal pixel points in the gray image by using an LOF abnormal value detection algorithm according to the nearest distance between the new abnormal pixel points to obtain the abnormal pixel points in the gray image.
2. The method for rapidly detecting defects of a bottom cover of a wristwatch according to claim 1, wherein the step of capturing the surface image of the bottom cover of the wristwatch comprises the following specific steps:
the camera is arranged right above the watch bottom cover, and an annular low-angle LED light source is adopted to illuminate the watch bottom cover; and acquiring an image of the surface of the bottom cover of the watch.
3. The method for rapidly detecting defects of a bottom cover of a wristwatch according to claim 1, wherein the step of obtaining abnormal pixels in a gray level image comprises the following specific steps:
obtaining gradients of all pixel points in the gray level image by using a sobel operator; the pixel point with the gradient of not 0 in the gray image is marked as an abnormal pixel point.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311585982.0A CN117314894B (en) | 2023-11-27 | 2023-11-27 | Method for rapidly detecting defects of watch bottom cover |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311585982.0A CN117314894B (en) | 2023-11-27 | 2023-11-27 | Method for rapidly detecting defects of watch bottom cover |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117314894A CN117314894A (en) | 2023-12-29 |
| CN117314894B true CN117314894B (en) | 2024-03-29 |
Family
ID=89285089
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311585982.0A Active CN117314894B (en) | 2023-11-27 | 2023-11-27 | Method for rapidly detecting defects of watch bottom cover |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117314894B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111931704A (en) * | 2020-09-15 | 2020-11-13 | 北京嘀嘀无限科技发展有限公司 | Method, device, equipment and computer readable storage medium for evaluating map quality |
| CN113240721A (en) * | 2021-03-31 | 2021-08-10 | 煤炭科学研究总院 | Navigation vanishing point detection method applied to complex scene of roadway |
| CN115239738A (en) * | 2022-09-26 | 2022-10-25 | 南通鑫生派智能科技有限公司 | Intelligent detection method for automobile part configuration |
| CN115423807A (en) * | 2022-11-04 | 2022-12-02 | 山东益民服饰有限公司 | Cloth defect detection method based on outlier detection |
| CN116777907A (en) * | 2023-08-18 | 2023-09-19 | 深圳市福山自动化科技有限公司 | Sheet metal part quality detection method |
| CN117095008A (en) * | 2023-10-20 | 2023-11-21 | 深圳市高进实业有限公司 | Intelligent detection method for defects of steel bar pipe of clock |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112348773B (en) * | 2020-09-28 | 2022-09-23 | 歌尔股份有限公司 | Screen defect detection method and device and electronic equipment |
| CN115082467B (en) * | 2022-08-22 | 2022-11-18 | 山东亿昌装配式建筑科技有限公司 | Building material welding surface defect detection method based on computer vision |
-
2023
- 2023-11-27 CN CN202311585982.0A patent/CN117314894B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111931704A (en) * | 2020-09-15 | 2020-11-13 | 北京嘀嘀无限科技发展有限公司 | Method, device, equipment and computer readable storage medium for evaluating map quality |
| CN113240721A (en) * | 2021-03-31 | 2021-08-10 | 煤炭科学研究总院 | Navigation vanishing point detection method applied to complex scene of roadway |
| CN115239738A (en) * | 2022-09-26 | 2022-10-25 | 南通鑫生派智能科技有限公司 | Intelligent detection method for automobile part configuration |
| CN115423807A (en) * | 2022-11-04 | 2022-12-02 | 山东益民服饰有限公司 | Cloth defect detection method based on outlier detection |
| CN116777907A (en) * | 2023-08-18 | 2023-09-19 | 深圳市福山自动化科技有限公司 | Sheet metal part quality detection method |
| CN117095008A (en) * | 2023-10-20 | 2023-11-21 | 深圳市高进实业有限公司 | Intelligent detection method for defects of steel bar pipe of clock |
Non-Patent Citations (1)
| Title |
|---|
| 基于机器视觉的金属表面缺陷检测方法研究;翟伟良;李有煊;黄浩湄;黄茂发;陈俊杰;莫锦超;;科技传播;20170508(第09期);第65-66+68页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117314894A (en) | 2023-12-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109141232B (en) | Online detection method for disc castings based on machine vision | |
| CN116758061B (en) | Casting surface defect detection method based on computer vision | |
| CN109685760B (en) | MATLAB-based SLM powder bed powder laying image convex hull depression defect detection method | |
| CN109242853A (en) | A kind of PCB defect intelligent detecting method based on image procossing | |
| CN115082451B (en) | Stainless steel soup ladle defect detection method based on image processing | |
| CN113610850B (en) | Decorative paper texture abnormity detection method based on image processing | |
| CN111272768B (en) | Ceramic tube detection method | |
| CN110111301A (en) | Metal based on frequency-domain transform aoxidizes surface defect visible detection method | |
| CN115205290A (en) | Online detection method and system for PCB production process | |
| CN115063423A (en) | Mechanical casting cold and hot crack self-adaptive identification method based on computer vision | |
| CN116993742B (en) | Nickel alloy rolling defect detection method based on machine vision | |
| CN111047556A (en) | Strip steel surface defect detection method and device | |
| CN116363126B (en) | Welding quality detection method for data line USB plug | |
| CN116612112B (en) | Visual inspection method for surface defects of bucket | |
| CN116309416A (en) | Visual inspection method for surface of part | |
| CN112669265A (en) | Method for realizing surface defect detection based on Fourier transform and image gradient characteristics | |
| CN116051543A (en) | Defect identification method for peeling steel | |
| CN117437219A (en) | Magnetic glue inductance irregular appearance defect detection method based on computer vision | |
| CN117152129B (en) | Visual detection method and system for surface defects of battery cover plate | |
| CN117314894B (en) | Method for rapidly detecting defects of watch bottom cover | |
| CN117011291B (en) | Watch shell quality visual detection method | |
| CN111815575B (en) | Bearing steel ball part detection method based on machine vision | |
| CN115330770B (en) | Cloth area type defect identification method | |
| CN115294119B (en) | Machine vision-based method for detecting stains in inner grooves of heads of plum-blossom-shaped threads | |
| CN110857919A (en) | Tail yarn defect detection method for package filament |
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 |