LU504271B1 - Method for Defect Detection of LED Wick - Google Patents
Method for Defect Detection of LED Wick Download PDFInfo
- Publication number
- LU504271B1 LU504271B1 LU504271A LU504271A LU504271B1 LU 504271 B1 LU504271 B1 LU 504271B1 LU 504271 A LU504271 A LU 504271A LU 504271 A LU504271 A LU 504271A LU 504271 B1 LU504271 B1 LU 504271B1
- Authority
- LU
- Luxembourg
- Prior art keywords
- region
- led wick
- segmentation
- value
- sub
- Prior art date
Links
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
- G06T7/001—Industrial image inspection using an image reference approach
-
- 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
- G06T7/11—Region-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
- 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
- G06V10/762—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using clustering, e.g. of similar faces in social networks
- G06V10/763—Non-hierarchical techniques, e.g. based on statistics of modelling distributions
-
- 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)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Quality & Reliability (AREA)
- Health & Medical Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
- Led Devices (AREA)
Abstract
The application relates to the technical field of image processing, in particular to a method for defect detection of a Light Emitting Diode wick. The method includes: acquiring an LED wick image with a plurality of LED lamps, clustering the acquired LED wick image based on a grayscale value, and acquiring a highlight region in the LED wick image; determining a sub-region of a single LED wick according to the highlight region; acquiring an initial segmentation threshold corresponding to each sub-region, obtaining an initial binary image according to the initial segmentation threshold, acquiring a gradient value of an edge pixel and the segmentation area of the LED wick based on the initial binary image; obtaining a final binary image through the final segmentation threshold the accuracy of defect detection is improved through local threshold segmentation.
Description
METHOD FOR DEFECT DETECTION OF LED WICK 50427)
The application relates to the technical field of image processing, in particular to a method for defect detection of a Light Emitting Diode (LED) wick.
An LED wick has the advantages of energy saving, long life, free maintenance, easy control, environmental protection, and the like, and may be applied in many occasions, such as lighting, street decoration, landmark buildings and indicator lights. Defect detection of the
LED wick has always been an indispensable part in a wick production process, which involves checking the photoelectric performance and appearance of the LED wick and removing defective products.
At present, defect detection of the LED wick is performed through a computer vision algorithm, that is, defect segmentation is performed on a collected LED wick image by a traditional threshold segmentation method. However, there are a plurality of wicks on one
LED wick image, and different wicks have different defect levels. The traditional threshold segmentation method only has one segmentation threshold, and when the same segmentation threshold is used for defect detection, some wicks may have enlarged defects or inconspicuous defects and thus unable to accurately detect the defects of the LED wick.
In order to solve the above-mentioned technical problem, the application aims to provide a method for defect detection of an LED wick, and the adopted technical solution is specifically as follows.
An embodiment of the application provides a method for defect detection of an LED wick. The method includes the following steps.
An LED wick image with at least two LED lamps is collected, a grayscale image corresponding to the LED wick image is acquired, the grayscale image is clustered based on a grayscale value of each pixel to obtain at least two categories, a highlight region is acquired according to an average value of the grayscale values corresponding to each category, and a sub-region of a single LED wick is determined according to the highlight region.
An initial segmentation threshold corresponding to each sub-region is acquired, an initial binary image corresponding to each sub-region is acquired based on the initial segmentation LUS04271 threshold, the grayscale value of the pixel corresponding to a background in the initial binary image is 0, the grayscale value of the pixel corresponding to the LED wick is 1, and the segmentation area of the LED wick is obtained according to the number of pixels with the grayscale value of 1 in the initial binary image. Edge detection is performed on the initial binary image of the current sub-region to obtain edge pixels of the current sub-region, a gradient value of the corresponding edge pixel is acquired according to the grayscale values of other pixels around each edge pixel, a poor segmentation rate of each edge pixel is acquired according to the gradient value, and the poor segmentation rate of each edge pixel is added to obtain an overall poor segmentation rate of the current sub-region. An adjustment objective function of the initial segmentation threshold of the current sub-region is constructed by combining the overall poor segmentation rate and the segmentation area of the LED wick of the current sub-region. A final segmentation threshold of the current sub-region is acquired by using a simulated annealing algorithm based on the adjustment objective function.
The corresponding sub-region is segmented according to the final segmentation threshold to obtain a final binary image, template image matching is performed on the final binary image and a standard binary image to obtain a matching degree, and defect detection of the LED wick is completed based on the matching degree.
Further, a method for acquiring the sub-region of the single LED wick includes the following operations.
A category corresponding to the maximum average value of the grayscale values is acquired as a target category corresponding to the highlight region, binarization is performed on the LED wick image based on the target category, the highlight region of the LED wick image after binarization is obtained by using morphology, connected domain analysis is performed on the highlight region to obtain a centroid of each connected domain, a distance value of any two of the centroids is acquired, and when the distance value is equal to a standard distance value, it is confirmed that the corresponding two connected domains belong to the same LED wick, and then the sub-region corresponding to the single LED wick is obtained according to the two connected domains. The standard distance value refers to a distance value between two highlight regions of the single LED wick in a standard template.
Further, a method for acquiring the gradient value of the corresponding edge pixel according to the grayscale values of other pixels around each edge pixel includes the following operations.
Edge detection is performed on the initial binary image of the current sub-region to obtain the edge pixels of the current sub-region, a window region of a set size is acquired with LUS04271 each edge pixel as the center, and a difference between the maximum grayscale value and the minimum grayscale value in the window region is taken as the gradient value of the edge pixel.
Further, a method for acquiring the poor segmentation rate of each edge pixel according to the gradient value includes the following operations.
An average gradient value of the current sub-region is obtained according to the gradient value of each edge pixel in the current sub-region, an absolute value of a difference between the gradient value of the i-th edge pixel and the average gradient value of the sub-region where the i-th edge pixel is located is acquired, an opposite number of the gradient value of the i-th edge pixel is acquired and substituted into a value corresponding to an exponential function with a constant e as a base number, the absolute value of the difference is multiplied by the value corresponding to the exponential function, and an obtained result is taken as the poor segmentation rate of the i-th edge pixel.
Further, a method for constructing the adjustment objective function of the initial segmentation threshold of the current sub-region by combining the overall poor segmentation rate and the segmentation area of the LED wick of the current sub-region includes the following operation.
A calculation formula of the adjustment objective function is: » Fuss fa, + IM — Mi where t is a segmentation threshold corresponding to the current solution; fg' is the overall poor segmentation rate of the current sub-region under the segmentation threshold corresponding to the current solution; M is the segmentation area of the corresponding LED wick under the initial segmentation threshold of the current sub-region, M; is the segmentation area of the LED wick of the current sub-region under the segmentation threshold corresponding to the current solution; and |*| is an absolute value function.
Further, a method for completing defect detection of the LED wick based on the matching degree includes the following operations.
A matching degree value threshold is set, and when the matching degree is lower than the matching degree value threshold, it is confirmed that the LED lamp corresponding to the
LED wick image has a defect.
The application has the following beneficial effects that: in the application, the LED wick image corresponding to the plurality of LED lamps is collected, the pixels in the LED wick image are clustered to obtain the highlight region based on the highlight feature LUS04271 generated by the LED wick, and the sub-region of the single LED wick in the LED wick image is determined by analyzing the highlight region based on the standard distance between two highlight regions corresponding to the single LED lamp, so that the division of the sub-region corresponding to the single LED wick is more accurate; the adjustment objective function of the segmentation threshold is constructed based on the initial segmentation threshold of each sub-region, and the final segmentation threshold of each sub-region, namely, the optimal segmentation threshold of each sub-region, is adaptively acquired by using the adjustment objective function; and the corresponding sub-region in the LED wick image is segmented according to the final segmentation threshold to obtain the final binary image, which improves the effect of threshold segmentation, so that the defect detection effect confirmed by template image matching between the final binary image and the standard binary image is better, and the accuracy of a defect detection result is improved.
In order to more clearly illustrate the embodiments of the application or the technical solutions and advantages in the related art, the drawings used in the description of the embodiments or the related art will be briefly described below. It is apparent that the drawings described below are only some embodiments of the application. Other drawings may further be obtained by those of ordinary skill in the art according to these drawings without creative efforts.
Fig. 1 is a flowchart of steps of a method for defect detection of LED wick according to an embodiment of the application.
Fig. 2 is a schematic diagram of a result of performing binarization on a grayscale image based on a target category according to an embodiment of the application.
Fig. 3 is a schematic diagram of a result of preprocessing a grayscale image after binarization by using morphology according to an embodiment of the application.
In order to further describe the technical means and effects of the application for achieving the predetermined objects, specific implementation modes, structures, features and effects of a method for defect detection of an LED wick of the application will be described below in detail in conjunction with the drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not always refer to the same embodiment. In addition, specific features, structures or characteristics of one or LU504271 more embodiments may be combined in any suitable way.
Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art of the application. 5 À specific solution of the method for defect detection of the LED wick provided by the application will be described below in detail in conjunction with the drawings.
Referring to Fig. 1, which shows a flowchart of steps of a method for defect detection of
LED wick according to an embodiment of the application, the method includes the following steps.
At S1, an LED wick image with at least two LED lamps is collected, a grayscale image corresponding to the LED wick image is acquired, the grayscale image is clustered based on a grayscale value of each pixel to obtain at least two categories, a highlight region is acquired according to an average value of the grayscale values corresponding to each category, and a sub-region of a single LED wick is determined according to the highlight region.
Specifically, the LED wick image with the plurality of LED lamps is collected by an
LED wick detection device.
After the LED wick image is collected, a conventional method may obtain the segmented wick part for defect detection of the LED wick through an Otsu threshold segmentation method, but an Otsu threshold is a threshold, which may result in that the segmentation effect is not necessarily good for all the wicks, for example, when the threshold is too large, although the segmentation effect is obvious, over-segmentation may occur, so that it is necessary to locally adjust a local grayscale threshold of the LED wick image to obtain a better image segmentation result.
The premise of acquiring the local grayscale threshold is to first perform region division on the LED wick image to obtain the sub-region of the single LED wick, and a specific division process thereof is that: the corresponding grayscale image is obtained by performing grayscale processing on the LED wick image, and all the pixels are clustered according to the grayscale value of each pixel in the grayscale image. In this solution, clustering is performed by a k-mean algorithm, and k=10, so that all the pixels in the grayscale image are classified into 10 categories, the category with the largest average value of the grayscale values in the 10 categories is selected as a target category, and a region corresponding to the target category is the highlight region in the LED wick image. As shown in Fig. 2, binarization is performed on the grayscale image based on the target category, where the grayscale value of the pixel in the target category is set as 1, and the grayscale values of other pixels corresponding to a non-target category are set as 0. The grayscale image after binarization is preprocessed by LUS04271 using morphology to obtain the image as shown in Fig. 3. Then, in order to acquire a connected domain corresponding to the more accurate highlight region in the LED wick image, the connected domain of the highlight region in the grayscale image after binarization is acquired through a connected domain extraction algorithm, a centroid of each connected domain is extracted so as to acquire centroid coordinates, and a distance between any two connected domains is obtained according to the centroid coordinates. The distance between any two connected domains obtained by calculation is compared with a standard distance between the connected domains corresponding to upper and lower positive and negative electrodes of a standard LED wick, if the distances are equal, it indicates that the two connected domains are in the same LED wick, and then according to the two connected domains belonging to the same LED wick confirmed in the grayscale image, the grayscale image is divided to obtain the sub-region corresponding to the single LED wick.
Grayscale processing, the k-mean algorithm and connected domain analysis are all known technologies, which will not be elaborated here.
It is to be noted that the size of an image region of the single wick on a captured image, namely, the complete sub-region of the single LED wick, may be accurately obtained through the standard size of the LED wick image.
At S2, an initial segmentation threshold corresponding to each sub-region is acquired, an initial binary image corresponding to each sub-region is acquired based on the initial segmentation threshold, the grayscale value of the pixel corresponding to a background in the initial binary image is 0, the grayscale value of the pixel corresponding to the LED wick is 1, and the segmentation area of the LED wick is obtained according to the number of pixels with the grayscale value of 1 in the initial binary image. Edge detection is performed on the initial binary image of the current sub-region to obtain edge pixels of the current sub-region, a gradient value of the corresponding edge pixel is acquired according to the grayscale values of other pixels around each edge pixel, a poor segmentation rate of each edge pixel is acquired according to the gradient value, and the poor segmentation rate of each edge pixel is added to obtain an overall poor segmentation rate of the current sub-region. An adjustment objective function of the initial segmentation threshold of the current sub-region is constructed by combining the overall poor segmentation rate and the segmentation area of the LED wick of the current sub-region. A final segmentation threshold of the current sub-region is acquired by using a simulated annealing algorithm based on the adjustment objective function.
Specifically, since the defect of the LED wick is obtained by a gradient difference, if the
LED wick is a complete and better wick, when the initial segmentation threshold LUS04271 segmentation is performed, the segmented wick edge and the background of the wick have more obvious difference information, and the difference information between the segmented wick edge and the background of the wick is more uniform. If the LED wick has the defect, the local image of the LED wick may be blurred, resulting in a smaller gradient value of the local image of the LED wick, but it does not necessarily indicate that the decrease in gradient is the defect of the LED wick and may be caused by the noise when the image is collected.
However, when the segmentation threshold is used for segmentation, segmentation is performed only according to the difference of grayscale distribution, which does not meet the requirements for segmentation of the defect of the LED wick, an LED wick that may not be regarded as defective may still have the wick defect after segmentation, which results in defect enlargement of the LED wick and causes energy waste. In order to avoid this situation, the following measures are taken.
The initial segmentation threshold of each sub-region is acquired by an Otsu method, the corresponding sub-region is segmented by using the initial segmentation threshold to obtain the initial binary image corresponding to each segmented sub-region, the grayscale value of the pixel corresponding to the background in the initial binary image is 0, and the grayscale value of the pixel corresponding to the LED wick is 1. Edge detection is performed on the initial binary image corresponding to each sub-region by using a canny operator to obtain the edge pixels of the LED wick, a difference between the maximum grayscale value and the minimum grayscale value in a 3*3 window is acquired as the gradient value of the edge pixel with each edge pixel as the center, and the average gradient value of each sub-region may be obtained through the gradient value of each edge pixel.
The Otsu method and the canny operator are known technologies, which will not be elaborated here.
If the gradient value of the edge pixel of the sub-region is smaller and has a larger difference from the average gradient value, it indicates that poor segmentation is likely to occur in the segmentation effect of the edge pixel of the current sub-region, and the initial segmentation threshold needs to be adjusted. An adjustment process of the initial segmentation threshold is as follows.
Taking one sub-region as an example, a difference Cz between the gradient value T of each edge pixel and the average gradient value T' corresponding to the sub-region is acquired, and the poor segmentation rate fg; of the i-th edge pixel of the sub-region is obtained by using the difference.
fac [Cr | * exp (ST) where Czi is the difference between the i-th edge pixel and the average gradient value; is an absolute value function; exp is an exponential function with a constant e as a bottom number; and Ti is the gradient value of the i-th edge pixel.
It is to be noted that the poor segmentation rate fg; indicates the degree of poor segmentation of the i-th edge pixel of the sub-region, the smaller the gradient value of the i-th edge pixel of the sub-region, that is, the smaller the Tj, the less obvious grayscale change, the worse the segmentation effect, and the larger the poor segmentation rate fgi, so that the negative correlation mapping is performed on Ti, and therefore the smaller the Ti, the larger the fg. The smaller the Cad the more uniform the effect during segmentation, and it may be considered that the current segmentation effect is good. On the contrary, it is considered that the segmentation effect on the LED wick is poor.
The larger the value of fg;, the greater the possibility of poor segmentation of the i-th edge pixel of the LED wick, and the greater the necessity of reference when the initial gray threshold is adjusted. The initial segmentation threshold is adjusted to reduce the poor segmentation rate fg; of the edge pixels to the minimum, and then the overall poor segmentation rate fg' of the edge pixels corresponding to the sub-region may be acquired, that is, the overall poor segmentation rate is obtained by accumulating the poor segmentation rates of the edge pixels corresponding to the sub-region. A calculation formula of the overall poor segmentation rate is as follows: where n is the number of edge pixels.
With the continuous adjustment and change of the segmentation threshold, the segmentation result of the LED wick may change continuously. If the segmentation result of the LED wick changes too much, it may cause the LED wick to be partially interfered, resulting in poor segmentation result. Furthermore, this method of adjusting the segmentation threshold is very inefficient, and the result is also inaccurate, so that the adjustment objective function is constructed based on the overall poor segmentation rate. Through the adjustment objective function, the optimal segmentation threshold, namely, the final segmentation threshold of the sub-region, is obtained, and a process of constructing the adjustment objective function is as follows. LU504271
The segmentation area M of the sub-region, under the initial segmentation threshold is acquired, which corresponds to a region where the grayscale value of the pixel in the initial binary image corresponding to the single LED wick is 1, and the adjustment objective function of the initial segmentation threshold of the sub-region is constructed by combining the overall poor segmentation rate and the segmentation area of the sub-region. A calculation formula of the adjustment objective function is:
F, fae IM — Mi where t is the segmentation threshold corresponding to the current solution, fg' is the overall poor segmentation rate of the current sub-region under the segmentation threshold corresponding to the current solution, and M4 is the segmentation area of the LED wick of the current sub-region under the segmentation threshold corresponding to the current solution.
The smaller the overall poor segmentation rate, the smaller the possibility of poor segmentation of the edge pixels. The smaller the IM a Mil the smaller the change of the segmentation result of the LED wick, and the more the poor segmentation caused by the excessive change of the segmentation result may be avoided. Therefore, the smaller the Fy, the more suitable the corresponding solution is as the final segmentation threshold, so that the t value corresponding to the minimum value of Ft is the final segmentation threshold.
In the application, the final segmentation threshold is calculated by the adjustment objective function based on the initial segmentation threshold through a simulated annealing algorithm. When simulated annealing is performed, an initial solution is usually a random solution. However, in the application, the initial segmentation threshold is taken as the initial solution. Since the purpose of acquiring the optimal segmentation threshold is to obtain the better segmentation effect, the application preferably sets that an initial temperature T=100, an attenuation parameter K=0.9, the attenuation parameter is an annealing rate, and an iteration stopping temperature is 0.01, so that the difference between the initial segmentation threshold and the final segmentation threshold is not too large, the number of iterations may be reduced and the efficiency may be improved.
The final segmentation threshold of each sub-region is adaptively obtained through the simulated annealing algorithm with the set parameters, then the adjustment objective function corresponding to each sub-region is constructed according to the initial segmentation threshold of each sub-region based on the above-mentioned method for acquiring the final segmentation threshold, and the final segmentation threshold of each sub-region is obtained LUS04271 by using the simulated annealing algorithm based on the adjustment objective function.
The simulated annealing algorithm is a known technology, which will not be elaborated here.
At S3, the corresponding sub-region is segmented according to the final segmentation threshold to obtain a final binary image, template image matching is performed on the final binary image and a standard binary image to obtain a matching degree, and defect detection of the LED wick is completed based on the matching degree.
Specifically, the corresponding sub-region in the LED wick image is segmented through the acquired final segmentation threshold to obtain a final segmentation result of the LED wick image, namely, the final binary image, skeletonization extraction is performed on the final binary image, and template image matching is performed on the extracted final binary image and the standard binary image to obtain the matching degree. The standard binary image is a binary image of the LED wick image without any defect. A matching degree threshold is set, when the matching degree is less than or equal to the matching degree threshold, it indicates that the current wick has a defect, and when the matching degree is greater than the matching degree threshold, it indicates that the current wick is qualified.
Preferably, in the embodiment of the application, the matching degree threshold takes an empirical value of 0.95, and the matching degree threshold may be adjusted according to a specific scenario.
It is to be noted that the skeletonization extraction is performed first and then the image template matching is performed due to the fact that the defect features may be preserved and the area of the segmentation result of the LED wick may be reduced, so that there may also be more obvious matching degree in image matching results in the case of a smaller defect.
It is to be noted that the sequence of the embodiments of the application does not represent superiority-inferiority of the embodiments but only for description. The processes depicted in the drawings do not necessarily require a specific sequence or successive sequences shown to achieve the desired results. In some implementation modes, multi-tasking and parallel processing are also possible or may be advantageous.
The various embodiments in the present specification are described in a progressive manner, the same or similar parts between the various embodiments can be referred to each other, and each embodiment focuses on differences from other embodiments.
The description above is only the preferred embodiment of the application and is not intended to limit the application. Any modifications, equivalent replacements, improvements
. 1e ._. __, 1: LU504271 and the like made within the principle of the application shall fall within the scope of protection of the application.
Claims (6)
1. A method for defect detection of a Light Emitting Diode (LED) wick, wherein the method comprises the following steps: collecting an LED wick image with at least two LED lamps, acquiring a grayscale image corresponding to the LED wick image, clustering the grayscale image based on a grayscale value of each pixel to obtain at least two categories, acquiring a highlight region according to an average value of the grayscale values corresponding to each category, and determining a sub-region of a single LED wick according to the highlight region; acquiring an initial segmentation threshold of each sub-region, acquiring an initial binary image corresponding to each sub-region based on the initial segmentation threshold, the grayscale value of the pixel corresponding to a background in the initial binary image being 0, the grayscale value of the pixel corresponding to the LED wick being 1, and obtaining the segmentation area of the LED wick according to the number of pixels with the grayscale value of 1 in the initial binary image; performing edge detection on the initial binary image of the current sub-region to obtain edge pixels of the current sub-region, acquiring a gradient value of the corresponding edge pixel according to the grayscale values of other pixels around each edge pixel, acquiring a poor segmentation rate of each edge pixel according to the gradient value, and adding the poor segmentation rate of each edge pixel to obtain an overall poor segmentation rate of the current sub-region; constructing an adjustment objective function of the initial segmentation threshold of the current sub-region by combining the overall poor segmentation rate and the segmentation area of the LED wick of the current sub-region; acquiring a final segmentation threshold of the current sub-region by using a simulated annealing algorithm based on the adjustment objective function; and segmenting the corresponding sub-region according to the final segmentation threshold to obtain a final binary image, performing template image matching on the final binary image and a standard binary image to obtain a matching degree, and completing defect detection of the LED wick based on the matching degree.
2. The method for defect detection of the LED wick as claimed in claim 1, wherein a method for acquiring the sub-region of the single LED wick comprises: acquiring a category corresponding to the maximum average value of the grayscale values as a target category corresponding to the highlight region, performing binarization on the LED wick image based on the target category, obtaining the highlight region of the LED wick image after binarization by using morphology, performing connected domain analysis on the highlight region to obtain a centroid of each connected domain, acquiring a distance value of any two of the centroids, and when the distance value is equal to a standard distance value, LUS04271 confirming that the corresponding two connected domains belong to the same LED wick, and then obtaining the sub-region corresponding to the single LED wick according to the two connected domains, wherein the standard distance value refers to a distance value between two highlight regions of the single LED wick in a standard template.
3. The method for defect detection of the LED wick as claimed in claim 1, wherein a method for acquiring the gradient value of the corresponding edge pixel according to the grayscale values of other pixels around each edge pixel comprises: performing edge detection on the initial binary image of the current sub-region to obtain the edge pixels of the current sub-region, acquiring a window region of a set size with each edge pixel as the center, and taking a difference between the maximum grayscale value and the minimum grayscale value in the window region as the gradient value of the edge pixel.
4. The method for defect detection of the LED wick as claimed in claim 1, wherein a method for acquiring the poor segmentation rate of each edge pixel according to the gradient value comprises: obtaining an average gradient value of the current sub-region according to the gradient value of each edge pixel in the current sub-region, acquiring an absolute value of a difference between the gradient value of the i-th edge pixel and the average gradient value of the sub-region where the i-th edge pixel is located, acquiring an opposite number of the gradient value of the i-th edge pixel and substituting same into a value corresponding to an exponential function with a constant e as a base number, multiplying the absolute value of the difference by the value corresponding to the exponential function, and taking an obtained result as the poor segmentation rate of the i-th edge pixel.
5. The method for defect detection of the LED wick as claimed in claim 1, wherein a method for constructing the adjustment objective function of the initial segmentation threshold of the current sub-region by combining the overall poor segmentation rate and the segmentation area of the LED wick of the current sub-region comprises: a calculation formula of the adjustment objective function being: PF, = fa, + IM — My where t is a segmentation threshold corresponding to the current solution; fg' is the overall poor segmentation rate of the current sub-region under the segmentation threshold corresponding to the current solution; M is the segmentation area of the corresponding LED wick under the initial segmentation threshold of the current sub-region, M; is the segmentation area of the LED wick of the current sub-region under the segmentation LUS04271 threshold corresponding to the current solution; and | * | is an absolute value function.
6. The method for defect detection of the LED wick as claimed in claim 1, wherein a method for completing defect detection of the LED wick based on the matching degree comprises: setting a matching degree value threshold, and when the matching degree is lower than the matching degree value threshold, confirming that the LED lamp corresponding to the LED wick image has a defect.
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211611872.2A CN115601367B (en) | 2022-12-15 | 2022-12-15 | LED lamp wick defect detection method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| LU504271B1 true LU504271B1 (en) | 2023-07-31 |
Family
ID=84854357
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| LU504271A LU504271B1 (en) | 2022-12-15 | 2023-04-03 | Method for Defect Detection of LED Wick |
Country Status (3)
| Country | Link |
|---|---|
| CN (1) | CN115601367B (en) |
| LU (1) | LU504271B1 (en) |
| WO (1) | WO2023134792A2 (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117649391A (en) * | 2023-12-11 | 2024-03-05 | 东莞市中钢模具有限公司 | Method and system for detecting defects of die casting die based on image processing |
Families Citing this family (90)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115601367B (en) * | 2022-12-15 | 2023-04-07 | 苏州迈创信息技术有限公司 | LED lamp wick defect detection method |
| CN116612126B (en) * | 2023-07-21 | 2023-09-19 | 青岛国际旅行卫生保健中心(青岛海关口岸门诊部) | Container disease vector biological detection early warning method based on artificial intelligence |
| CN116630321B (en) * | 2023-07-24 | 2023-10-03 | 铁正检测科技有限公司 | Intelligent bridge health monitoring system based on artificial intelligence |
| CN116630315B (en) * | 2023-07-24 | 2023-09-29 | 山东东阿亿福缘阿胶制品有限公司 | Intelligent beverage packaging defect detection method based on computer vision |
| CN116664567B (en) * | 2023-07-26 | 2023-09-29 | 山东艾迈科思电气有限公司 | Solid insulation switch cabinet quality assessment method and system |
| CN116664574B (en) * | 2023-07-31 | 2023-10-20 | 山东罗斯夫新材料科技有限公司 | Visual detection method for acrylic emulsion production wastewater |
| CN116704177B (en) * | 2023-08-01 | 2023-11-17 | 东莞市零越传动科技有限公司 | Gear box fault detection method based on image data |
| CN116721099B (en) * | 2023-08-09 | 2023-11-21 | 山东奥洛瑞医疗科技有限公司 | Image segmentation method of liver CT image based on clustering |
| CN117115197B (en) * | 2023-08-09 | 2024-05-17 | 幂光新材料科技（上海）有限公司 | Intelligent processing method and system for design data of LED lamp bead circuit board |
| CN116721106B (en) * | 2023-08-11 | 2023-10-20 | 山东明达圣昌铝业集团有限公司 | Profile flaw visual detection method based on image processing |
| CN116721107B (en) * | 2023-08-11 | 2023-11-03 | 青岛胶州电缆有限公司 | Intelligent monitoring system for cable production quality |
| CN116740058B (en) * | 2023-08-11 | 2023-12-01 | 深圳市金胜电子科技有限公司 | Quality detection method for solid state disk matched wafer |
| CN116740653A (en) * | 2023-08-14 | 2023-09-12 | 山东创亿智慧信息科技发展有限责任公司 | Distribution box running state monitoring method and system |
| CN116740061B (en) * | 2023-08-14 | 2023-11-21 | 山东淼珠生物科技有限公司 | Visual detection method for production quality of explosive beads |
| CN116758071B (en) * | 2023-08-17 | 2023-11-03 | 青岛冠宝林活性炭有限公司 | Intelligent detection method for carbon electrode dirt under visual assistance |
| CN116758086B (en) * | 2023-08-21 | 2023-10-20 | 山东聚宁机械有限公司 | Bulldozer part quality detection method based on image data |
| CN116777916B (en) * | 2023-08-24 | 2023-10-27 | 济宁安泰矿山设备制造有限公司 | Defect detection method based on metal shell of pump machine |
| CN116778431B (en) * | 2023-08-25 | 2023-11-10 | 青岛娄山河水务有限公司 | Automatic sludge treatment monitoring method based on computer vision |
| CN116843757B (en) * | 2023-08-29 | 2023-12-01 | 山东千颐科技有限公司 | Intelligent labeling machine positioning method based on computer vision |
| CN116883446B (en) * | 2023-09-08 | 2023-11-21 | 鲁冉光电(微山)有限公司 | Real-time monitoring system for grinding degree of vehicle-mounted camera lens |
| CN116883415B (en) * | 2023-09-08 | 2024-01-05 | 东莞市旺佳五金制品有限公司 | Thin-wall zinc alloy die casting quality detection method based on image data |
| CN116883412B (en) * | 2023-09-08 | 2023-11-17 | 浙江中骏石墨烯科技有限公司 | Graphene far infrared electric heating equipment fault detection method |
| CN116912260B (en) * | 2023-09-15 | 2023-11-28 | 沂水友邦养殖服务有限公司 | Broiler chicken breeding health state detection method based on artificial intelligence |
| CN116977333B (en) * | 2023-09-22 | 2023-12-22 | 山东恒海钢结构有限公司 | Image data-based quality detection method for assembled steel structural member |
| CN116993724B (en) * | 2023-09-26 | 2023-12-08 | 卡松科技股份有限公司 | Visual detection method for coal mine industrial gear oil based on image filtering |
| CN116993742B (en) * | 2023-09-28 | 2023-12-22 | 中色(天津)新材料科技有限公司 | Nickel alloy rolling defect detection method based on machine vision |
| CN117058130B (en) * | 2023-10-10 | 2024-01-09 | 威海威信光纤科技有限公司 | Visual inspection method for coating quality of optical fiber drawing surface |
| CN117078667B (en) * | 2023-10-13 | 2024-01-09 | 山东克莱蒙特新材料科技有限公司 | Mineral casting detection method based on machine vision |
| CN117094993B (en) * | 2023-10-18 | 2024-03-26 | 山东聚亨源环保科技有限公司 | Precipitation detection method in sewage treatment process |
| CN117095009B (en) * | 2023-10-20 | 2024-01-12 | 山东绿康装饰材料有限公司 | PVC decorative plate defect detection method based on image processing |
| CN117115153B (en) * | 2023-10-23 | 2024-02-02 | 威海坤科流量仪表股份有限公司 | Intelligent printed circuit board quality detection method based on visual assistance |
| CN117152137B (en) * | 2023-10-30 | 2024-01-26 | 江苏高特高金属科技有限公司 | Welded pipe corrosion state detection method based on image processing |
| CN117152444B (en) * | 2023-10-30 | 2024-01-26 | 山东泰普锂业科技有限公司 | Equipment data acquisition method and system for lithium battery industry |
| CN117152180B (en) * | 2023-10-31 | 2024-01-26 | 山东克莱蒙特新材料科技有限公司 | Mineral casting defect detection method based on artificial intelligence |
| CN117152447B (en) * | 2023-11-01 | 2024-02-02 | 东莞市京品精密模具有限公司 | Intelligent management method and system for punching die |
| CN117152160B (en) * | 2023-11-01 | 2024-01-26 | 湖南辰东科技有限公司 | Airport lamp detection method and system based on image processing |
| CN117197130B (en) * | 2023-11-03 | 2024-01-26 | 山东太阳耐磨件有限公司 | Driving tooth angle defect identification method based on machine vision |
| CN117173184B (en) * | 2023-11-03 | 2024-01-26 | 济宁市市政园林养护中心 | Road construction quality detection method and system based on artificial intelligence |
| CN117197141B (en) * | 2023-11-07 | 2024-01-26 | 山东远盾网络技术股份有限公司 | Method for detecting surface defects of automobile parts |
| CN117351001B (en) * | 2023-11-16 | 2024-05-28 | 肇庆市大正铝业有限公司 | Surface defect identification method for regenerated aluminum alloy template |
| CN117237396B (en) * | 2023-11-16 | 2024-02-06 | 山东华盛中天工程机械有限责任公司 | Rail bolt rust area segmentation method based on image characteristics |
| CN117252893B (en) * | 2023-11-17 | 2024-02-23 | 科普云医疗软件(深圳)有限公司 | Segmentation processing method for breast cancer pathological image |
| CN117274293B (en) * | 2023-11-17 | 2024-03-15 | 广东省农业科学院动物科学研究所 | Accurate bacterial colony dividing method based on image features |
| CN117252877B (en) * | 2023-11-17 | 2024-02-02 | 济南界龙科技有限公司 | Diode lead frame quality detection method based on image characteristics |
| CN117252876B (en) * | 2023-11-17 | 2024-02-09 | 江西斯迈得半导体有限公司 | LED support defect detection method and system |
| CN117274405B (en) * | 2023-11-22 | 2024-02-02 | 深圳市蓝方光电有限公司 | LED lamp working color detection method based on machine vision |
| CN117292328B (en) * | 2023-11-24 | 2024-02-02 | 山东新中鲁建设有限公司 | Safety management and monitoring method and system for construction quality of assembled building |
| CN117291922B (en) * | 2023-11-27 | 2024-01-30 | 浙江日井泵业股份有限公司 | Visual detection method for defects of stainless steel multistage pump impeller |
| CN117291924B (en) * | 2023-11-27 | 2024-02-09 | 惠州市鑫晖源科技有限公司 | Main machine power supply appearance quality detection method based on machine vision |
| CN117314949B (en) * | 2023-11-28 | 2024-02-20 | 山东远硕上池健康科技有限公司 | Personnel injury detection and identification method based on artificial intelligence |
| CN117314924B (en) * | 2023-11-30 | 2024-02-09 | 湖南西欧新材料有限公司 | Image feature-based electroplated product surface flaw detection method |
| CN117372435B (en) * | 2023-12-08 | 2024-02-06 | 智联信通科技股份有限公司 | Connector pin detection method based on image characteristics |
| CN117437229B (en) * | 2023-12-20 | 2024-03-15 | 山东晨光胶带有限公司 | High-strength flame-retardant turning conveyor belt defect detection method based on image analysis |
| CN117437600B (en) * | 2023-12-20 | 2024-03-26 | 山东海纳智能装备科技股份有限公司 | Coal flow monitoring system based on image recognition technology |
| CN117576086B (en) * | 2024-01-11 | 2024-06-21 | 普森美微电子技术(苏州)有限公司 | Method and system for detecting metal electron beam welding defects |
| CN117576089B (en) * | 2024-01-15 | 2024-03-22 | 山东恒力源精密机械制造有限公司 | Piston ring defect detection method and system |
| CN117635610B (en) * | 2024-01-25 | 2024-04-09 | 青岛正大正电力环保设备有限公司 | Visual detection method for oil leakage of oil pipe of hydraulic tensioning mechanism |
| CN117649412B (en) * | 2024-01-30 | 2024-04-09 | 山东海天七彩建材有限公司 | Aluminum material surface quality detection method |
| CN117689663B (en) * | 2024-02-04 | 2024-04-26 | 电科科知(成都)科技集团有限公司 | Visual detection method and system for welding robot |
| CN117745724B (en) * | 2024-02-20 | 2024-04-26 | 高唐县瑞景精密机械有限公司 | Stone polishing processing defect region segmentation method based on visual analysis |
| CN117764992B (en) * | 2024-02-22 | 2024-04-30 | 山东乔泰管业科技有限公司 | Plastic pipe quality detection method based on image processing |
| CN117764989B (en) * | 2024-02-22 | 2024-04-30 | 深圳优色专显科技有限公司 | Visual-aided display screen defect detection method |
| CN117788464B (en) * | 2024-02-26 | 2024-04-30 | 卡松科技股份有限公司 | Industrial gear oil impurity visual detection method |
| CN117788829B (en) * | 2024-02-27 | 2024-05-07 | 长春师范大学 | Image recognition system for invasive plant seed detection |
| CN117789132B (en) * | 2024-02-27 | 2024-05-07 | 深圳市深创高科电子有限公司 | Electronic product shell defect monitoring method and system based on computer vision |
| CN117830300B (en) * | 2024-03-04 | 2024-05-14 | 新奥新能源工程技术有限公司 | Visual-based gas pipeline appearance quality detection method |
| CN117853483B (en) * | 2024-03-05 | 2024-06-14 | 济宁市市政园林养护中心 | Intelligent analysis method for concrete stirring quality |
| CN117853487B (en) * | 2024-03-07 | 2024-05-14 | 浙江合丰科技有限公司 | FPC connector crack detection method and system based on image processing technology |
| CN117876360B (en) * | 2024-03-08 | 2024-07-02 | 卡松科技股份有限公司 | Intelligent detection method for lubricating oil quality based on image processing |
| CN117893532B (en) * | 2024-03-14 | 2024-05-24 | 山东神力索具有限公司 | Die crack defect detection method for die forging rigging based on image processing |
| CN118014988B (en) * | 2024-03-28 | 2024-06-07 | 浙江康鹏半导体有限公司 | Intelligent gallium arsenide substrate wafer defect detection method |
| CN117974646B (en) * | 2024-03-29 | 2024-06-14 | 山东太平洋光纤光缆有限公司 | Visual inspection method for coating quality of optical fiber surface |
| CN117994258B (en) * | 2024-04-07 | 2024-07-09 | 深圳市松旭机电设备有限公司 | PCB quality detection method based on computer vision |
| CN118014993B (en) * | 2024-04-08 | 2024-06-11 | 中机凯博表面技术江苏有限公司 | Superfine coating powder screening quality detection method |
| CN118037718B (en) * | 2024-04-11 | 2024-07-02 | 海门裕隆光电科技有限公司 | Electrical terminal production defect detection method |
| CN118097305B (en) * | 2024-04-16 | 2024-06-28 | 深圳市呈泰半导体科技有限公司 | Method and system for detecting quality of semiconductor light-emitting element |
| CN118096726B (en) * | 2024-04-18 | 2024-06-25 | 江苏裕荣光电科技有限公司 | Cable temperature anomaly identification method |
| CN118071757B (en) * | 2024-04-25 | 2024-06-25 | 江苏裕荣光电科技有限公司 | Defect detection method for pipe gallery cable protective film |
| CN118134921B (en) * | 2024-05-07 | 2024-08-30 | 太仓瑞鼎精密机械科技有限公司 | Real-time reamer fracture defect detection method and system |
| CN118134928B (en) * | 2024-05-08 | 2024-06-28 | 江苏太湖锅炉股份有限公司 | Slag layer thickness accurate detection method for boiler water wall |
| CN118212478B (en) * | 2024-05-22 | 2024-07-30 | 大连博讯科技有限公司 | Construction engineering quality detection method based on image processing |
| CN118314133B (en) * | 2024-06-07 | 2024-08-09 | 深圳鼎智通讯股份有限公司 | Intelligent terminal defect rapid detection method based on machine vision |
| CN118351116B (en) * | 2024-06-17 | 2024-09-06 | 浙江省水利水电勘测设计院有限责任公司 | Thermal insulation performance detection method for low-carbon building material |
| CN118365966B (en) * | 2024-06-17 | 2024-08-27 | 陕西华昱太阳能科技有限公司 | Insect pest identification method based on image processing insecticidal lamp |
| CN118378648B (en) * | 2024-06-20 | 2024-08-30 | 山东渔郎食品有限公司 | Canned fish production information tracing method and system |
| CN118379317B (en) * | 2024-06-24 | 2024-09-13 | 常州市宏发纵横新材料科技股份有限公司 | Glass fiber cloth edge identification method, equipment and storage medium |
| CN118397027B (en) * | 2024-06-26 | 2024-09-06 | 宝鸡市宏远特种金属材料有限公司 | Titanium wire uniformity detection method and device based on image recognition |
| CN118411383B (en) * | 2024-07-04 | 2024-08-27 | 尽开科技(大连)有限公司 | Livestock growth state monitoring method based on machine vision |
| CN118445335B (en) * | 2024-07-08 | 2024-09-20 | 山东惠通科技有限公司 | Data management method and system for intelligent Internet of things platform |
| CN118470006A (en) * | 2024-07-10 | 2024-08-09 | 北京奥力斯特投资管理有限公司 | Novel heat exchange tube quality visual detection method and system |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111028202B (en) * | 2019-11-14 | 2021-07-13 | 苏州斯玛维科技有限公司 | X-ray bubble defect image processing method, device, storage medium and system for LED chip |
| DE102021116177A1 (en) * | 2020-06-24 | 2022-01-05 | Metralabs Gmbh Neue Technologien Und Systeme | System and method for the detection and avoidance of image defects |
| CN111815630B (en) * | 2020-08-28 | 2020-12-15 | 歌尔股份有限公司 | Defect detection method and device for LCD screen |
| CN112784847B (en) * | 2021-01-28 | 2022-03-04 | 中国空气动力研究与发展中心超高速空气动力研究所 | Segmentation and identification method for ultra-high-speed impact damage detection image |
| CN114723701B (en) * | 2022-03-31 | 2023-04-18 | 厦门力和行自动化有限公司 | Gear defect detection method and system based on computer vision |
| CN115018828B (en) * | 2022-08-03 | 2022-10-25 | 深圳市尹泰明电子有限公司 | Defect detection method for electronic component |
| CN115018853B (en) * | 2022-08-10 | 2022-10-25 | 南通市立新机械制造有限公司 | Mechanical component defect detection method based on image processing |
| CN115601367B (en) * | 2022-12-15 | 2023-04-07 | 苏州迈创信息技术有限公司 | LED lamp wick defect detection method |
-
2022
- 2022-12-15 CN CN202211611872.2A patent/CN115601367B/en active Active
-
2023
- 2023-04-03 LU LU504271A patent/LU504271B1/en active IP Right Grant
- 2023-04-03 WO PCT/CN2023/086006 patent/WO2023134792A2/en unknown
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117649391A (en) * | 2023-12-11 | 2024-03-05 | 东莞市中钢模具有限公司 | Method and system for detecting defects of die casting die based on image processing |
| CN117649391B (en) * | 2023-12-11 | 2024-05-10 | 东莞市中钢模具有限公司 | Method and system for detecting defects of die casting die based on image processing |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115601367A (en) | 2023-01-13 |
| WO2023134792A3 (en) | 2023-09-07 |
| WO2023134792A2 (en) | 2023-07-20 |
| CN115601367B (en) | 2023-04-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| LU504271B1 (en) | Method for Defect Detection of LED Wick | |
| CN115082683B (en) | Injection molding defect detection method based on image processing | |
| CN116758061B (en) | Casting surface defect detection method based on computer vision | |
| CN104700085B (en) | A kind of chip positioning method based on template matches | |
| CN104794491B (en) | Based on the fuzzy clustering Surface Defects in Steel Plate detection method presorted | |
| CN115249246B (en) | Optical glass surface defect detection method | |
| CN109242853B (en) | PCB defect intelligent detection method based on image processing | |
| CN114419025A (en) | Fiberboard quality evaluation method based on image processing | |
| CN110610496B (en) | Fluorescent glue defect segmentation method with robust illumination change | |
| CN115861320B (en) | Intelligent detection method for automobile part machining information | |
| CN106097368A (en) | A kind of recognition methods in veneer crack | |
| CN115294410B (en) | Plastic product molding control method based on pattern recognition | |
| WO2023159985A1 (en) | Video detection technology-based identification method for abnormal behaviors of sorting center | |
| CN115222735B (en) | Metal mold quality detection method based on pockmark defects | |
| CN116152242B (en) | Visual detection system of natural leather defect for basketball | |
| CN109120230A (en) | A kind of solar battery sheet EL image detection and defect identification method | |
| CN115359237A (en) | Gear broken tooth identification method based on pattern identification | |
| CN105447489A (en) | Character and background adhesion noise elimination method for image OCR system | |
| CN118097305A (en) | Method and system for detecting quality of semiconductor light-emitting element | |
| CN114140416A (en) | Glass edge detection method and system based on machine vision | |
| CN115311443B (en) | Oil leakage identification method for hydraulic pump | |
| CN115578390A (en) | Welding control method for deaerator | |
| CN114862786A (en) | Retinex image enhancement and Ostu threshold segmentation based isolated zone detection method and system | |
| CN116681707B (en) | Cornea fluorescein staining image identification grading method | |
| CN117152148B (en) | Method for detecting defect of wool spots of textile |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| FG | Patent granted |
Effective date: 20230731 |