CN117495830A - Chip bonding area defect detection method, system and terminal equipment - Google Patents
Chip bonding area defect detection method, system and terminal equipment Download PDFInfo
- Publication number
- CN117495830A CN117495830A CN202311522705.5A CN202311522705A CN117495830A CN 117495830 A CN117495830 A CN 117495830A CN 202311522705 A CN202311522705 A CN 202311522705A CN 117495830 A CN117495830 A CN 117495830A
- Authority
- CN
- China
- Prior art keywords
- image
- area
- chip
- target image
- watershed algorithm
- 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.)
- Pending
Links
- 230000007547 defect Effects 0.000 title claims abstract description 73
- 238000001514 detection method Methods 0.000 title claims abstract description 49
- 238000000034 method Methods 0.000 claims abstract description 47
- 238000003709 image segmentation Methods 0.000 claims abstract description 43
- 230000008569 process Effects 0.000 claims abstract description 24
- 238000012545 processing Methods 0.000 claims abstract description 21
- 239000003550 marker Substances 0.000 claims abstract description 18
- 238000001914 filtration Methods 0.000 claims abstract description 15
- 238000012797 qualification Methods 0.000 claims abstract description 15
- 230000009466 transformation Effects 0.000 claims description 13
- 239000011159 matrix material Substances 0.000 claims description 9
- 238000004458 analytical method Methods 0.000 claims description 8
- 238000000605 extraction Methods 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 7
- 230000000149 penetrating effect Effects 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 20
- 230000000694 effects Effects 0.000 description 17
- 230000011218 segmentation Effects 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 239000011800 void material Substances 0.000 description 3
- 238000005315 distribution function Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 238000004806 packaging method and process Methods 0.000 description 2
- 239000004065 semiconductor Substances 0.000 description 2
- 238000012876 topography Methods 0.000 description 2
- 238000003466 welding Methods 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 239000000853 adhesive Substances 0.000 description 1
- 230000001070 adhesive effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000008094 contradictory effect Effects 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 230000007123 defense Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 239000002356 single layer Substances 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 239000000126 substance Substances 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10116—X-ray image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20112—Image segmentation details
- G06T2207/20152—Watershed segmentation
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20172—Image enhancement details
-
- 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
- G06T2207/30148—Semiconductor; IC; Wafer
-
- 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)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Analysing Materials By The Use Of Radiation (AREA)
Abstract
The invention discloses a method, a system and terminal equipment for detecting defects in a chip bonding area, wherein the method comprises the following steps: acquiring an X-ray image of a chip to be detected, and extracting a target area to obtain a target image; performing image filtering and image enhancement processing on the target image; image segmentation is carried out on the target image based on a watershed algorithm, wherein in the process of image segmentation of the watershed algorithm, a threshold value when the watershed algorithm obtains an internal marker is obtained based on the contrast of the target image; and judging the qualification of the chip bonding area based on the image segmentation result, and outputting a detection result. The invention relates to the technical field of component detection, and aims at improving the quality and accuracy of image segmentation in the image segmentation process by correlating the contrast of a target image with a threshold value when a watershed algorithm acquires an internal marker, so that the detection accuracy of defects of a chip bonding area is improved.
Description
Technical Field
The invention relates to the technical field of component detection, in particular to a method, a system and terminal equipment for detecting defects of a chip bonding area.
Background
The chip is a basic part in various electronic systems and even information systems, is taken as the most fundamental substance basis in the current information age, and has important significance for national folk life. The semiconductor chip provides material guarantee for information systems such as cores, aviation, aerospace, ships, weapons, electronics and the like. Whether they are stable and reliable or not is directly related to the advantages and disadvantages of the information system functions and performance indexes, and even closely related to national defense safety.
Chip defect detection involves appearance detection and internal detection. The appearance inspection is mainly to detect the surface defects of the components by using human eyes or a microscope; internal inspection is required to detect and analyze the internal structure of the component and possible defects by means of an ultrasonic scanning microscope, X-rays, etc.
At present, the chip X-ray detection is mainly manually identified, and a tester realizes manual interpretation analysis by browsing images. However, as the processing and manufacturing levels of semiconductor chips are increasingly increased over time, chip packaging forms are increasingly abundant and packaging densities are also gradually increasing. The manual identification speed and accuracy are greatly influenced by subjective factors such as experience and physical conditions of testers, the detection efficiency and reliability are not high, the detection quality of the chip is difficult to be practically guaranteed, and therefore risks and hidden dangers in the use process of the chip are brought, and the requirements of people on an electronic system are not met. Therefore, a defect detection method is urgently needed, which fully automatizes a series of processes such as X-ray image processing and defect detection of a detected part, replaces a great deal of manual labor of detection personnel, and reduces interpretation errors caused by human factors so as to meet detection tasks of different types and models of components.
Disclosure of Invention
Aiming at the current situations of increasing the complexity of components and increasing the detection task amount in the prior art, the invention provides a chip bonding area defect detection method, system and terminal equipment based on an improved watershed algorithm for a chip bonding area, and solves the problems of low efficiency and poor precision of traditional manual detection.
In order to achieve the above object, the present invention provides a chip bonding area defect detection method based on an improved watershed algorithm, comprising the following steps:
step 1, acquiring a chip X-ray image to be detected, and extracting a target area of the chip X-ray image based on template matching to obtain a target image;
step 2, carrying out image filtering and image enhancement processing on the target image;
step 3, image segmentation is carried out on the target image based on a watershed algorithm, wherein in the process of image segmentation of the watershed algorithm, a threshold value when the watershed algorithm acquires an internal marker is obtained based on the contrast of the target image;
and 4, judging the qualification of the chip bonding area based on the image segmentation result, and outputting a detection result.
In one embodiment, in step 1, the target area extraction is performed on the chip X-ray image based on template matching, specifically:
sequentially sliding on the chip X-ray images based on the template images, traversing the chip X-ray images, and extracting the area, covered by the template images, of the chip X-ray images as a subgraph in the process of sliding the template images;
and taking the subgraph with the largest phase relation number as the target image based on the correlation coefficient of each subgraph and the template image.
In one embodiment, the calculating process of the correlation coefficient is:
wherein R (i, j) is sub-graph S ij The correlation coefficients with the template image T are m and n which are the numbers of pixels of the subgraph and the template image in the width direction and the height direction, m and n are the numbers of pixels of the template image in the width direction and the height direction, and W, H is the number of pixels of the chip X-ray image in the width direction and the height direction.
In one embodiment, in step 2, the target image is image filtered based on gaussian filtering.
In one embodiment, in step 2, the image enhancement processing is performed on the target image based on histogram equalization, specifically:
wherein r (x, y) is the pixel value of the point with the coordinates (x, y) on the original target image, M is the line number of the target image, N is the column number of the target image, L is the number of gray levels, s (x, y) is the pixel value of the point with the coordinates (x, y) on the target image after the image enhancement processing, r j Is the number of pixels in the target image with a gray level j.
In one embodiment, in step 3, the image segmentation is performed on the target image based on a watershed algorithm, specifically:
step 3.1, calculating the contrast of the target image;
step 3.2, calculating a threshold value when the watershed algorithm obtains an internal marker based on the contrast;
step 3.3, converting the target image into a gray image, and detecting only edges of the gray image based on the threshold value to obtain a binary image;
step 3.4, performing distance transformation on the binary image to obtain a distance transformation matrix, wherein the value in the distance transformation matrix is Euclidean distance between a corresponding pixel point in the binary image and the nearest foreground pixel point;
step 3.5, searching all local extremum points on the distance transformation matrix, and taking the local extremum points as watershed lines;
step 3.6, grouping the watershed lines so that the watershed lines in the same group are connected with each other to form a region;
and 3.7, dividing the distinction to obtain a divided image.
In one embodiment, in step 3.2, the threshold is specifically:
θ=k·C
where θ is a threshold, C is the contrast of the target image, and k is a threshold coefficient.
In one embodiment, in step 4, the determining the eligibility of the die attach area based on the image segmentation result specifically includes:
and according to the image segmentation result of the target image, obtaining the size of the defect through connected domain analysis and calculation, respectively calculating the areas of the defect area and the bonding area based on a pixel statistics method, and judging the qualification of the chip bonding area according to the area ratio.
In one embodiment, the determining the qualification of the die attach area according to the area ratio specifically includes:
when the area of the defect area is greater than 50% of the total area of the bonding area, judging that the bonding area of the chip is unqualified;
when the area of the defect area is less than or equal to 50% of the total area of the bonding area and is greater than or equal to 10% of the total area of the bonding area, judging whether the defect penetrating through the bonding area of the chip exists or not through the connected domain analysis, if so, judging that the bonding area of the chip is unqualified, otherwise, judging that the bonding area of the chip is qualified;
and when the area of the defect area is smaller than 10% of the total area of the bonding area, judging that the chip bonding area is qualified.
In order to achieve the above object, the present invention further provides a system for detecting defects in a die bonding area based on an improved watershed algorithm, the method for detecting defects in a die bonding area, the system for detecting defects in a die bonding area comprising:
the image acquisition unit is used for acquiring an X-ray image of the chip to be detected;
the target extraction unit is used for extracting a target area of the chip X-ray image according to template matching to obtain a target image;
an image processing unit for performing image filtering and image enhancement processing on the target image;
the image segmentation unit is used for carrying out image segmentation on the target image by adopting a watershed algorithm, wherein in the process of carrying out image segmentation by the watershed algorithm, a threshold value when the watershed algorithm acquires an internal marker is obtained based on the contrast of the target image;
and the defect detection unit is used for judging the qualification of the chip bonding area according to the image segmentation result and outputting a detection result.
In order to achieve the above object, the present invention further provides a terminal device, where the terminal device is provided with:
a memory for storing a program;
and a processor for executing the program stored in the memory, the processor being configured to perform some or all of the steps of the method as described above when the program is executed.
Compared with the prior art, the invention has the following beneficial technical effects:
in the process of defect detection of a chip bonding area, the invention adopts the watershed algorithm with improved points to divide the target image, and the improvement mode is that the contrast of the target image is related to the threshold value when the watershed algorithm acquires the internal marker, so that the effect that the threshold value can be dynamically changed according to different target images is achieved. And the contrast of the target image is related to the threshold value when the watershed algorithm acquires the internal marker, so that the watershed algorithm has more pertinence in the process of image segmentation, the quality and the accuracy of image segmentation are improved, and the detection precision of the defects of the chip bonding area is improved.
Drawings
In order to more clearly illustrate the embodiments of the present 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, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to the structures shown in these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a method for detecting defects in a die attach area based on an improved watershed algorithm in example 1 of the present invention;
fig. 2 is a schematic diagram of the matching effect of the mold in embodiment 1 of the present invention, wherein: (a) A template image schematic diagram, and (b) a matched image schematic diagram;
fig. 3 is a schematic diagram of gaussian filtering effect in embodiment 1 of the present invention, wherein: (a) A target image schematic diagram before filtering, (b) a target image schematic diagram after filtering;
fig. 4 is a schematic diagram of the histogram equalization effect in embodiment 1 of the present invention, wherein: (a) is an original target image schematic diagram, (b) is a histogram of the original target image, (c) is an equalized target image schematic diagram, and (d) is a histogram of the equalized target image;
FIG. 5 is a schematic diagram showing the segmentation effect of the modified watershed algorithm in the embodiment 1 of the present invention, wherein: (a) A traditional watershed segmentation effect schematic diagram is shown, and (b) an improved watershed segmentation effect schematic diagram is shown;
fig. 6 is a schematic diagram showing the effect of detecting voids in an adhesion area in embodiment 1 of the present invention, wherein: (a) An original target image schematic diagram and (b) a cavity defect schematic diagram of an adhesion area;
fig. 7 is a schematic diagram of the detection effect on other kinds of chips 1 in embodiment 1 of the present invention, wherein: (a) is a schematic view of another type of chip 1, (b) is a schematic view of an image of an adhesion region of the chip 1, and (c) is a schematic view of a detection effect of the chip 1;
fig. 8 is a schematic diagram of the detection effect on other kinds of chips 2 in embodiment 1 of the present invention, wherein: (a) is a schematic view of other kinds of chips 2, (b) is a schematic view of an image of an adhesion region of the chip 2, and (c) is a schematic view of a detection effect of the chip 2;
FIG. 9 is a block diagram showing a chip bonding area defect detection system based on the modified watershed algorithm in example 2 of the present invention;
fig. 10 is a block diagram of a terminal device in embodiment 3 of the present invention.
The achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be noted that all directional indicators (such as up, down, left, right, front, and rear … …) in the embodiments of the present invention are merely used to explain the relative positional relationship, movement, etc. between the components in a particular posture (as shown in the drawings), and if the particular posture is changed, the directional indicator is changed accordingly.
Furthermore, descriptions such as those referred to as "first," "second," and the like, are provided for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implying an order of magnitude of the indicated technical features in the present disclosure. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include at least one such feature. In the description of the present invention, the meaning of "plurality" means at least two, for example, two, three, etc., unless specifically defined otherwise.
In the present invention, unless specifically stated and limited otherwise, the terms "connected," "affixed," and the like are to be construed broadly, and for example, "affixed" may be a fixed connection, a removable connection, or an integral body; the device can be mechanically connected, electrically connected, physically connected or wirelessly connected; either directly or indirectly, through intermediaries, or both, may be in communication with each other or in interaction with each other, unless expressly defined otherwise. The specific meaning of the above terms in the present invention can be understood by those of ordinary skill in the art according to the specific circumstances.
In addition, the technical solutions of the embodiments of the present invention may be combined with each other, but it is necessary to be based on the fact that those skilled in the art can implement the technical solutions, and when the technical solutions are contradictory or cannot be implemented, the combination of the technical solutions should be considered as not existing, and not falling within the scope of protection claimed by the present invention.
Example 1
Fig. 1 shows a chip bonding area defect detection method based on an improved watershed algorithm, which mainly includes the following steps 1 to 4.
Step 1, acquiring a chip X-ray image to be detected, and extracting a target area from the chip X-ray image based on template matching to obtain a target image.
Referring to FIG. 2, using a correlation matching algorithm, let the original chip X-ray image be S (W, H), the template image be T (m, n), and the part of the template image T (m, n) covered by the chip X-ray image S (W, H) be called sub-image S ij Wherein, (i, j) is the coordinates of the lower left end point of the sub-image in the chip X-ray image, m, n are the number of pixels of the template image in the width direction and the height direction, and W, H is the number of pixels of the chip X-ray image in the width direction and the height direction.
Specifically, the template image sequentially moves on the chip X-ray image from left to right and from top to bottom, the chip X-ray image is traversed, the area covered by the template image is extracted as sub-images in the process of sliding the template image, the sub-images are based on the correlation coefficient between each sub-image and the template image, and the sub-image with the largest correlation number is taken as a target image, so that the target area extraction is completed. Wherein, the searching range of template matching is that i is more than or equal to 1 and less than or equal to W-n, and j is more than or equal to 1 and less than or equal to H-m.
In this embodiment, the calculation process of the correlation coefficient specifically includes:
wherein R (i, j) is sub-graph S ij (m, n) correlation coefficient with the template image, and when the subgraph is identical to the template image, the correlation coefficient R (i, j) =1.
And 2, performing image filtering and image enhancement processing on the target image.
In this embodiment, the target image is image-filtered based on gaussian filtering. The basic idea of gaussian filtering is to perform a convolution operation with a gaussian function, centering on a certain pixel, applying a two-dimensional normal distribution function as weights to pixels around it, and then weighted-averaging these values and taking them as new values for that pixel. The gaussian filtering can smooth noise generated at the edges of the image by eliminating high frequency components in the image, thereby achieving the effect of suppressing noise. When digital image processing is performed, a two-dimensional Gaussian function is generally used, and the formula of the two-dimensional Gaussian function is as follows:
as shown in fig. 3, the gaussian filtering algorithm can effectively smooth noise generated at the edges of the image, suppress noise of the image and maintain the overall characteristics of the image.
In this embodiment, image enhancement processing is performed on the target image based on histogram equalization. The histogram equalization balances the occurrence frequency of each gray level in the histogram by carrying out nonlinear stretching on the image and adjusting the distribution of the pixel values of the image, so that each gray level is fully utilized, detail information can be better displayed, and the image quality is improved.
Let r be k Is the number of pixels of gray level k in the original image, p (r k ) For its probability of occurrence, there are:
wherein M is the number of lines of the target image, and N is the number of columns of the target image. Thus, a Cumulative Distribution Function (CDF) of the target image can be calculated as:
in order to map in the new gray level range, it is necessary to linearly transform the CDF to map to the gray level range between 0 and 255. Specifically, let s k Representing the value of gray level k in the new mapping table, there are:
where L represents the number of gray levels (typically 256). The meaning of this formula is to map all pixel values of k-level or less in the original image onto the new gray level, and L-1 is used in order to scale the mapped gray level to a suitable range. Therefore, the target image after histogram equalization is:
where r (x, y) is the pixel value of the point on the original target image with coordinates (x, y), r j The number of pixels with the gray level j in the target image is M, the number of rows of the target image is M, the number of columns of the target image is N, the number of gray levels is L, and x (x, y) is the pixel value of a point with coordinates (x, y) on the target image after the image enhancement processing.
Referring to fig. 4, it can be seen that the contrast between the hole defect and the background in the central bonding area of the chip X-ray image after histogram equalization is obviously enhanced, and the detail of the defect area is more obvious.
And 3, image segmentation is carried out on the target image based on a watershed algorithm, wherein in the process of image segmentation by the watershed algorithm, a threshold value when the watershed algorithm acquires the internal marker is obtained based on the contrast of the target image.
The basic idea of the watershed algorithm is to consider the image as a topography in which areas of high brightness correspond to peaks and areas of low brightness correspond to valleys. The image is divided into a plurality of regions by adding virtual water to this topography, starting from the lowest point, until it merges into a watershed line.
In this embodiment, a watershed segmentation based on distance transformation is used, and the metric is Euclidean distance. And combining the image contrast with a threshold value when the watershed algorithm acquires the internal marker, and improving the watershed algorithm with a fixed threshold value into a watershed algorithm with a dynamic threshold value controlled by the image contrast. The contrast of the target image is related to the threshold value when the watershed algorithm acquires the internal marker, so that the watershed algorithm has more pertinence in the process of image segmentation, and the quality and accuracy of image segmentation are improved. Reference is made to fig. 5, which is a schematic diagram showing the comparison of the segmentation effect of the conventional watershed algorithm and the modified watershed algorithm.
In this embodiment, the specific implementation process for image segmentation of the target image based on the watershed algorithm is as follows:
step 3.1, calculating the contrast of the target image;
step 3.2, obtaining a threshold value when the watershed algorithm obtains the internal marker based on contrast calculation, specifically:
the contrast can be directly used as a threshold value when the watershed algorithm obtains the internal marker, or the threshold value when the watershed algorithm obtains the internal marker is obtained through the calculation of the contrast and the threshold value coefficient, and the threshold value is as follows:
θ=k·C
wherein θ is a threshold value, C is a contrast of the target image, and k is a threshold coefficient;
step 3.3, converting the target image into a gray image, and detecting only edges of the gray image based on a threshold value to obtain a binary image;
step 3.4, carrying out distance transformation on the binary image to obtain a distance transformation matrix, wherein the value in the distance transformation matrix is Euclidean distance between the corresponding pixel point in the binary image and the nearest foreground pixel point;
step 3.5, searching all local maximum extremum points on the distance transformation matrix, and taking the local maximum extremum points as a watershed line;
step 3.6, grouping the watershed lines so that the watershed lines in the same group are connected with each other to form a region;
and 3.7, dividing the distinction to obtain a divided image.
As shown in fig. 6, the black area is the detected void defect in the bonding area. As can be seen from FIG. 6, by the modified watershed algorithm, a large portion of voids can be detected from the image, wherein bubbles having a greater impact on the eligibility determination have been detected, and the desired effect is achieved.
In a specific implementation process, the calculation process of the threshold coefficient is as follows:
firstly, carrying out image segmentation on a plurality of sample images by adopting a traditional watershed algorithm, and adjusting a threshold value when the watershed algorithm acquires an internal marker according to an image segmentation result in the process of segmenting each sample image until the optimal image segmentation result exists, and marking the current threshold value as an optimal threshold value;
secondly, dividing the optimal threshold value of each sample image by the similarity to obtain a sample threshold value coefficient of the sample image;
finally, calculating the arithmetic average or geometric average of the sample threshold coefficients corresponding to all the sample images as the final threshold coefficient.
And 4, judging the qualification of the chip bonding area based on the image segmentation result, and outputting a detection result.
The eligibility of the single-layer chip welding area void is mainly judged according to the percentage of the void area to the total area of the welding contact area. For an input chip X-ray image, firstly, a watershed algorithm is adopted to divide a cavity area, meanwhile, the size of a defect is calculated through connected domain analysis, the areas of the cavity defect area and an adhesive area are respectively calculated through a pixel statistics method, and finally, the qualification of the chip X-ray image is judged according to a corresponding judgment criterion through an area ratio, and a corresponding judgment result is output. In this embodiment, the specific embodiment for determining the eligibility of the die attach area based on the image division result is as follows:
and according to the image segmentation result of the target image, obtaining the size of the defect through connected domain analysis and calculation, respectively calculating the areas of the defect area and the bonding area based on a pixel statistics method, and judging the qualification of the chip bonding area according to the area ratio.
In the specific implementation process, the qualification of the chip bonding area is judged according to the area ratio, specifically:
when the area of the defect area is greater than 50% of the total area of the bonding area, judging that the bonding area of the chip is unqualified;
when the area of the defect area is less than or equal to 50% of the total area of the bonding area and is greater than or equal to 10% of the total area of the bonding area, judging whether the defect penetrating through the bonding area of the chip exists or not through the connected domain analysis, if so, judging that the bonding area of the chip is unqualified, otherwise, judging that the bonding area of the chip is qualified;
and when the area of the defect area is smaller than 10% of the total area of the bonding area, judging that the chip bonding area is qualified.
As an example of the hole defect in the bonding area, as shown in fig. 6 (b), the number of pixels in the hole area and the number of pixels in the chip bonding area are detected, the number of pixels in the hole area is 5474, the number of pixels in the bonding area is 22137, and the hole defect ratio is 24.73% or less than 50%; further, whether or not a through hole exists is determined, and since the size of the single hole defect is smaller than the size of the die bonding region, no through hole exists in the die, and the final determination result is qualified. The method for detecting defects in the die attach area in this embodiment is also applicable to other types of chips, and the detection effect is as shown in fig. 7 and 8
Example 2
Based on the chip bonding area defect detection method in embodiment 1, this embodiment discloses a chip bonding area defect detection system based on an improved watershed algorithm. Referring to fig. 9, the chip bonding region defect detection system includes an image acquisition unit, a target extraction unit, an image processing unit, an image segmentation unit, and a defect detection unit. The system for detecting defects in the die attach area is used for performing part or all of the steps of the method for detecting defects in the die attach area in embodiment 1, and further performing defect detection on the die attach area. Specifically:
the image acquisition unit is used for acquiring an X-ray image of the chip to be detected;
the target extraction unit is used for extracting a target area of the chip X-ray image according to template matching to obtain a target image;
the image processing unit is used for carrying out image filtering and image enhancement processing on the target image;
the image segmentation unit is used for carrying out image segmentation on the target image by adopting a watershed algorithm, wherein a threshold value when the watershed algorithm obtains an internal marker is obtained based on the contrast of the target image in the process of carrying out image segmentation by the watershed algorithm;
the defect detection unit is used for judging the qualification of the chip bonding area according to the image segmentation result and outputting the detection result.
In this embodiment, the specific working processes and working principles of the image acquisition unit, the target extraction unit, the image processing unit, the image segmentation unit and the defect detection unit are the same as those of the method in embodiment 1, so that the description thereof will not be repeated in this embodiment.
Example 3
Fig. 10 shows a terminal device disclosed in this embodiment, which includes a transmitter, a receiver, a memory, and a processor. Wherein the transmitter is used for transmitting instructions and data, the receiver is used for receiving instructions and data, the memory is used for storing computer-executable instructions, and the processor is used for executing the computer-executable instructions stored in the memory so as to realize part or all of the steps executed by the defect detection of the chip bonding area in the embodiment 1. The specific implementation procedure is the same as that of the defect detection of the die bonding area in the foregoing embodiment 1.
It should be noted that the memory may be separate or integrated with the processor. When the memory is provided separately, the terminal device further comprises a bus for connecting the memory and the processor.
Example 4
The present embodiment discloses a computer-readable storage medium in which computer-executable instructions are stored, which when executed by a processor, implement part or all of the steps performed for detecting defects in a die attach area in embodiment 1 described above.
The foregoing description is only of the preferred embodiments of the present invention and is not intended to limit the scope of the invention, and all equivalent structural changes made by the description of the present invention and the accompanying drawings or direct/indirect application in other related technical fields are included in the scope of the invention.
Claims (10)
1. The chip bonding area defect detection method based on the improved watershed algorithm is characterized by comprising the following steps of:
step 1, acquiring a chip X-ray image to be detected, and extracting a target area of the chip X-ray image based on template matching to obtain a target image;
step 2, carrying out image filtering and image enhancement processing on the target image;
step 3, image segmentation is carried out on the target image based on a watershed algorithm, wherein in the process of image segmentation of the watershed algorithm, a threshold value when the watershed algorithm acquires an internal marker is obtained based on the contrast of the target image;
and 4, judging the qualification of the chip bonding area based on the image segmentation result, and outputting a detection result.
2. The method for detecting defects in a die attach area based on an improved watershed algorithm according to claim 1, wherein in step 1, the target area extraction is performed on the die X-ray image based on template matching, specifically:
sequentially sliding on the chip X-ray images based on the template images, traversing the chip X-ray images, and extracting the area, covered by the template images, of the chip X-ray images as a subgraph in the process of sliding the template images;
and taking the subgraph with the largest phase relation number as the target image based on the correlation coefficient of each subgraph and the template image.
3. The method for detecting defects in a die attach area based on an improved watershed algorithm as claimed in claim 2, wherein the correlation coefficient is calculated by:
wherein R (i, j) is sub-graph S ij The correlation coefficients with the template image T are m and n which are the numbers of pixels of the subgraph and the template image in the width direction and the height direction, m and n are the numbers of pixels of the template image in the width direction and the height direction, and W, H is the number of pixels of the chip X-ray image in the width direction and the height direction.
4. The method for detecting defects in a die attach area based on an improved watershed algorithm according to claim 1, wherein in step 2, the image enhancement processing is performed on the target image based on histogram equalization, specifically:
wherein r (x, y) is the pixel value of the point with the coordinates (x, y) on the original target image, M is the line number of the target image, N is the column number of the target image, L is the number of gray levels, s (x, y) is the pixel value of the point with the coordinates (x, y) on the target image after the image enhancement processing, r j Is the number of pixels in the target image with a gray level j.
5. The method for detecting defects in a die attach area based on an improved watershed algorithm according to any one of claims 1 to 4, wherein in step 3, the watershed algorithm performs image segmentation on the target image, specifically:
step 3.1, calculating the contrast of the target image;
step 3.2, calculating a threshold value when the watershed algorithm obtains an internal marker based on the contrast;
step 3.3, converting the target image into a gray image, and detecting only edges of the gray image based on the threshold value to obtain a binary image;
step 3.4, performing distance transformation on the binary image to obtain a distance transformation matrix, wherein the value in the distance transformation matrix is Euclidean distance between a corresponding pixel point in the binary image and the nearest foreground pixel point;
step 3.5, searching all local extremum points on the distance transformation matrix, and taking the local extremum points as watershed lines;
step 3.6, grouping the watershed lines so that the watershed lines in the same group are connected with each other to form a region;
and 3.7, dividing the distinction to obtain a divided image.
6. The method for detecting defects in a die attach area based on an improved watershed algorithm as in claim 5, wherein in step 3.2, the threshold is specifically:
θ=k·C
where θ is a threshold, C is the contrast of the target image, and k is a threshold coefficient.
7. The method for detecting defects in a die attach area based on the modified watershed algorithm as claimed in any one of claims 1 to 4, wherein in the step 4, the qualification of the die attach area is determined based on the image segmentation result, specifically:
and according to the image segmentation result of the target image, obtaining the size of the defect through connected domain analysis and calculation, respectively calculating the areas of the defect area and the bonding area based on a pixel statistics method, and judging the qualification of the chip bonding area according to the area ratio.
8. The method for detecting defects in a die attach area based on an improved watershed algorithm according to claim 7, wherein the determining of the qualification of the die attach area according to the area ratio is specifically:
when the area of the defect area is greater than 50% of the total area of the bonding area, judging that the bonding area of the chip is unqualified;
when the area of the defect area is less than or equal to 50% of the total area of the bonding area and is greater than or equal to 10% of the total area of the bonding area, judging whether the defect penetrating through the bonding area of the chip exists or not through the connected domain analysis, if so, judging that the bonding area of the chip is unqualified, otherwise, judging that the bonding area of the chip is qualified;
and when the area of the defect area is smaller than 10% of the total area of the bonding area, judging that the chip bonding area is qualified.
9. A chip bonding area defect detection system based on an improved watershed algorithm, characterized in that the method of any one of claims 1 to 8 is adopted to detect the defects of the chip bonding area, and the chip bonding area defect detection system comprises:
the image acquisition unit is used for acquiring an X-ray image of the chip to be detected;
the target extraction unit is used for extracting a target area of the chip X-ray image according to template matching to obtain a target image;
an image processing unit for performing image filtering and image enhancement processing on the target image;
the image segmentation unit is used for carrying out image segmentation on the target image by adopting a watershed algorithm, wherein in the process of carrying out image segmentation by the watershed algorithm, a threshold value when the watershed algorithm acquires an internal marker is obtained based on the contrast of the target image;
and the defect detection unit is used for judging the qualification of the chip bonding area according to the image segmentation result and outputting a detection result.
10. A terminal device, characterized in that the terminal device is provided with:
a memory for storing a program;
a processor for executing the program stored in the memory, the processor being adapted to perform part or all of the steps of the method according to any one of claims 1 to 8 when the program is executed.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311522705.5A CN117495830A (en) | 2023-11-15 | 2023-11-15 | Chip bonding area defect detection method, system and terminal equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311522705.5A CN117495830A (en) | 2023-11-15 | 2023-11-15 | Chip bonding area defect detection method, system and terminal equipment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN117495830A true CN117495830A (en) | 2024-02-02 |
Family
ID=89682648
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311522705.5A Pending CN117495830A (en) | 2023-11-15 | 2023-11-15 | Chip bonding area defect detection method, system and terminal equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117495830A (en) |
-
2023
- 2023-11-15 CN CN202311522705.5A patent/CN117495830A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111681197B (en) | Remote sensing image unsupervised change detection method based on Siamese network structure | |
CN108564085B (en) | Method for automatically reading of pointer type instrument | |
CN109087274A (en) | Electronic device defect inspection method and device based on multidimensional fusion and semantic segmentation | |
CN106170819A (en) | A kind of diameter radar image Ship Target method for quick | |
CN111626993A (en) | Image automatic detection counting method and system based on embedded FEFnet network | |
CN104463870A (en) | Image salient region detection method | |
CN106340010B (en) | A kind of angular-point detection method based on second order profile difference | |
CN110245600B (en) | Unmanned aerial vehicle road detection method for self-adaptive initial quick stroke width | |
CN110866926A (en) | Infrared remote sensing image rapid and fine sea-land segmentation method | |
CN111709964B (en) | PCBA target edge detection method | |
CN113971669A (en) | Three-dimensional detection system applied to pipeline damage identification | |
CN111325728B (en) | Product defect detection method, device, equipment and storage medium | |
CN105354547A (en) | Pedestrian detection method in combination of texture and color features | |
CN107944497A (en) | Image block method for measuring similarity based on principal component analysis | |
CN107369163B (en) | Rapid SAR image target detection method based on optimal entropy dual-threshold segmentation | |
CN113781413A (en) | Electrolytic capacitor positioning method based on Hough gradient method | |
CN113012127A (en) | Cardiothoracic ratio measuring method based on chest medical image | |
CN113139549A (en) | Parameter self-adaptive panorama segmentation method based on multitask learning | |
US20230386023A1 (en) | Method for detecting medical images, electronic device, and storage medium | |
CN117495830A (en) | Chip bonding area defect detection method, system and terminal equipment | |
Ortiz-Jaramillo et al. | Computing contrast ratio in images using local content information | |
CN107123105A (en) | Images match defect inspection method based on FAST algorithms | |
Kong et al. | Adaptive image edge detection model using improved Canny algorithm | |
CN116012283B (en) | Full-automatic ultrasonic image measurement method, equipment and storage medium | |
CN117274788B (en) | Sonar image target positioning method, system, electronic equipment and storage medium |
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 |