CN118397011A

CN118397011A - Golden finger defect detection method, golden finger defect detection device, computer equipment and storage medium

Info

Publication number: CN118397011A
Application number: CN202410865189.4A
Authority: CN
Inventors: 戴斌宇; 赵鑫
Original assignee: Suzhou HYC Technology Co Ltd
Current assignee: Suzhou HYC Technology Co Ltd
Priority date: 2024-07-01
Filing date: 2024-07-01
Publication date: 2024-07-26
Anticipated expiration: 2044-07-01
Also published as: CN118397011B

Abstract

The disclosure relates to a golden finger defect detection method, a golden finger defect detection device, computer equipment and a storage medium. The method comprises the following steps: acquiring an image to be detected, and extracting a golden finger gray image of a golden finger area in the image to be detected; determining a first boundary endpoint and a second boundary endpoint of each golden finger according to the reference graph area and the size of each golden finger; dividing a preset number of outlines between a first boundary endpoint and a second boundary endpoint of each golden finger, determining a middle locus of each golden finger based on gray values of the preset number of outlines, and determining a reference value line of a golden finger area based on the middle locus of each golden finger; determining a bright spot area in the golden finger gray level image by using an area growth algorithm; and performing defect detection on the image to be detected based on the distance from the bright point contour point to the reference value line and a distance threshold value to obtain a defect detection result. By adopting the method, the detection precision can be improved, and the golden finger can be accurately subjected to defect detection.

Description

Golden finger defect detection method, golden finger defect detection device, computer equipment and storage medium

Technical Field

The disclosure relates to the technical field of image processing, and in particular relates to a golden finger defect detection method, a golden finger defect detection device, computer equipment and a storage medium.

Background

With the vigorous development of the electronic industry, circuit designs become more complex and fine, and the requirements on the manufacturing process of semiconductor devices such as chips, semiconductor connectors and the like become more and more stringent. The golden finger area is an area on the semiconductor device, which is electroplated with a layer of metal material with high oxidation resistance and conductivity and is used for transmitting signals.

In general, a semiconductor device needs to be subjected to signal testing before being used to test whether the semiconductor device is qualified, a probe is used to contact a golden finger area on the semiconductor device to transmit signals during testing, the probe contact can leave a prick on the golden finger to damage a plating layer, but a mark is not allowed to appear in a drain Zone (KOZ) on the golden finger of the semiconductor device, so that after the semiconductor device is subjected to signal testing, whether the prick appears in the KOZ area needs to be detected.

The conventional detection method is mostly performed manually or by image processing, however, the conventional image processing method generally uses a threshold segmentation method to perform processing, and the method is difficult to obtain a good segmentation effect when facing a rough plating surface, so that the detection precision is reduced, and the defect detection cannot be performed on the golden finger accurately.

Disclosure of Invention

Accordingly, in order to solve the above-described problems, it is necessary to provide a golden finger defect detection method, device, computer apparatus, and storage medium that can improve detection accuracy and accurately detect defects in golden fingers.

In a first aspect, the present disclosure provides a golden finger defect detection method. The method comprises the following steps:

Acquiring an image to be detected, wherein the image to be detected comprises a golden finger area formed by equidistant distribution of a plurality of golden fingers and reference pattern areas positioned at two ends of the golden finger area, and extracting a golden finger gray image of the golden finger area in the image to be detected;

determining a first boundary endpoint and a second boundary endpoint of each golden finger according to the reference graph area and the size of each golden finger;

Dividing a preset number of outlines between a first boundary endpoint and a second boundary endpoint of each golden finger, determining a middle locus of each golden finger based on gray values of the preset number of outlines, and determining a reference value line of the golden finger area based on the middle locus of each golden finger;

determining a bright spot area in the golden finger gray level image by using an area growth algorithm, and calculating the distance from a bright spot contour point of the bright spot area to the reference value line;

and performing defect detection on the image to be detected based on the distance from the bright point contour point to the reference value line and a distance threshold value to obtain a defect detection result, wherein the distance threshold value is determined based on the reference value line and the exclusion area of the golden finger area.

In one embodiment, the determining the median point of each golden finger based on the gray values of the preset number of contours includes:

Performing Gaussian smoothing on the contours of the preset number to obtain contour smoothing results;

calculating a first derivative of the contour smoothing result to obtain a calculation result;

Determining a position point in the golden finger indicated by the calculation result which is larger than a preset derivative threshold;

and sequencing the position points according to the vertical direction of the golden fingers, and determining the middle position point of each golden finger.

In one embodiment, the extracting the golden finger gray level image of the golden finger region in the image to be detected includes:

And carrying out rotation correction on the image to be detected based on the reference pattern areas at the two ends, and extracting a golden finger gray image of the golden finger area in the image to be detected after rotation correction.

In one embodiment, after the extracting the golden finger gray scale image of the golden finger region in the image to be detected, the method further includes:

and processing the golden finger gray level image by adopting an image enhancement method based on gamma transformation.

In one embodiment, the determining the bright spot area in the golden finger gray scale image by using an area growing algorithm includes:

threshold segmentation is carried out on the golden finger gray level image, and a connected domain of a highlight region in the golden finger gray level image is extracted;

determining a pre-selected point according to a gray extreme point of the connected domain of the highlight region in the golden finger gray image and a preset first gray threshold value;

Determining a target point based on the pre-selected point and a predetermined growth criterion, wherein the growth criterion is determined based on a preset gray difference threshold, a second gray threshold, a gradient amplitude and a gradient mean;

Determining a binary image based on the target point and the golden finger gray level image;

And determining a bright spot area in the golden finger gray level image according to the gradient mean value of the connected domain in the binary image.

In one embodiment, the growth criteria include: the gray level difference value between the candidate point and the preselected point is smaller than a preset first gray level difference value; the second criterion is that the gray value of the candidate point is larger than a second gray threshold value, and the second gray threshold value is determined according to the average gray of the golden finger gray image; the third criterion is that the gradient amplitude of the candidate point is larger than a gradient average value threshold value, and the gradient amplitude threshold value is determined according to the gradient average value of each pixel point in the golden finger gray level image; the fourth criterion is that the gradient mean value of the second and fourth neighborhoods of the candidate points is larger than the gradient mean value threshold; the determining a target point based on the pre-selected point and a predetermined growth criterion comprises:

For each pre-selected point, determining eight neighborhood pixel points of the pre-selected point in the golden finger gray level image;

Taking the pixel points in the eight neighborhood pixel points as candidate points, and determining candidate points which are indicated by the pre-selected points and meet the conditions in response to the condition that any one of the candidate points meets the first criterion, the second criterion and the third criterion simultaneously or meets the conditions of the first criterion, the second criterion and the fourth criterion simultaneously;

And determining the target point according to the candidate points which are indicated by each pre-selected point and meet the condition.

In a second aspect, the present disclosure further provides a golden finger defect detection device. The device comprises:

The image processing module is used for acquiring an image to be detected, wherein the image to be detected comprises a plurality of golden finger areas distributed at equal intervals and reference pattern areas positioned at two ends of the image to be detected, and extracting golden finger gray images of the golden finger areas in the image to be detected;

The boundary point determining module is used for determining a first boundary endpoint and a second boundary endpoint of each golden finger according to the reference graph areas positioned at the two ends of the image to be detected and the size of each golden finger;

The reference line determining module is used for dividing a preset number of outlines between a first boundary endpoint and a second boundary endpoint of each golden finger, determining a middle position point of each golden finger based on gray values of the preset number of outlines, and determining a reference value line of the golden finger area based on the middle position point of each golden finger;

the distance calculation module is used for determining a bright point area in the golden finger gray level image by utilizing a target segmentation algorithm of area growth and calculating the distance from a bright point contour point of the bright point area to the reference value line;

And the defect detection module is used for carrying out defect detection on the image to be detected based on the distance from the bright point contour point to the reference value line and a distance threshold value to obtain a defect detection result, wherein the distance threshold value is determined based on the reference value line and the exclusion area of the golden finger area.

In a third aspect, the present disclosure also provides a computer device. The computer device comprises a memory storing a computer program and a processor implementing the steps of any of the method embodiments described above when the processor executes the computer program.

In a fourth aspect, the present disclosure also provides a computer-readable storage medium. The computer readable storage medium has stored thereon a computer program which, when executed by a processor, implements the steps of any of the method embodiments described above.

In a fifth aspect, the present disclosure also provides a computer program product. The computer program product comprises a computer program which, when executed by a processor, implements the steps of any of the method embodiments described above.

In the above embodiments, the image to be detected is obtained, and the golden finger gray level image of the golden finger area in the image to be detected is extracted; determining a first boundary endpoint and a second boundary endpoint of each golden finger according to the reference graph area and the size of each golden finger; dividing a preset number of outlines between a first boundary endpoint and a second boundary endpoint of each golden finger, determining a middle locus of each golden finger based on gray values of the preset number of outlines, and determining a reference value line of a golden finger area based on the middle locus of each golden finger. The position of each golden finger can be accurately positioned by utilizing the datum line of the golden finger area, so that a standard is provided for subsequent judgment, and the detection accuracy is ensured. In addition, a bright spot area in the golden finger gray level image is determined by utilizing an area growth algorithm, and adjacent pixels are gradually expanded and combined into an area according to a predefined growth criterion, so that the bright spot area contacted by a probe is accurately determined, the detection accuracy is further ensured, the defect detection is carried out on the detection image based on the distance from a bright spot contour point of the bright spot area to a reference value line and a distance threshold value, a defect detection result is obtained, and the detection accuracy can be improved compared with a traditional segmentation detection mode.

Drawings

In order to more clearly illustrate the embodiments of the present disclosure or the prior art, the drawings that are required in the detailed description or the prior art will be briefly described, it will be apparent to those skilled in the art that the drawings in the following description are some embodiments of the present disclosure and that other drawings may be obtained according to the drawings without inventive effort.

FIG. 1 is a flow chart of a golden finger defect detection method according to one embodiment;

FIG. 2 is a schematic diagram of an image to be detected in one embodiment;

FIG. 3 is a schematic diagram of a first boundary point, a second boundary point and a contour in a golden finger image according to one embodiment;

FIG. 4 is a schematic diagram between a bright point contour spot and a reference value line in a bright point area in one embodiment;

FIG. 5 is a schematic diagram of one embodiment between a bright spot area and an exclusion area;

FIG. 6 is a flow chart of determining a mid-site in step S106 according to one embodiment;

FIG. 7 is a flowchart of determining a bright spot area in step S108 according to an embodiment;

FIG. 8 is a flow chart of step S306 in one embodiment;

FIG. 9 is a flowchart of a golden finger defect detection method according to another embodiment;

FIG. 10 is a schematic block diagram of a golden finger defect detection device according to one embodiment;

FIG. 11 is a schematic diagram of the internal architecture of a computer device in one embodiment;

FIG. 12 is a schematic diagram of the internal structure of a computer device in one embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present disclosure more apparent, the present disclosure will be further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present disclosure.

It should be noted that the terms "first," "second," and the like in the description and claims herein and in the foregoing figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments described herein may be capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, apparatus, article, or device that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed or inherent to such process, method, article, or device.

In this document, the term "and/or" is merely one association relationship describing the associated object, meaning that three relationships may exist. For example, a and/or B may represent: a exists alone, A and B exist together, and B exists alone. In addition, the character "/" herein generally indicates that the front and rear associated objects are an "or" relationship.

In one embodiment, as shown in fig. 1, the embodiment of the disclosure provides a method for detecting a golden finger defect, where the method is applied to a terminal for illustrating, it can be understood that the method may also be applied to a server, and may also be applied to a system including the terminal and the server, and implemented through interaction between the terminal and the server. In this embodiment, the method includes the steps of:

S102, an image to be detected comprises a golden finger area formed by equidistant distribution of a plurality of golden fingers and reference pattern areas positioned at two ends of the golden finger area, and a golden finger gray level image of the golden finger area in the image to be detected is extracted.

The image to be detected may be an image of a device to be tested, which contains a gold finger and is tested by the gold finger, the device to be tested may include, but is not limited to, various electronic components, semiconductor devices, mechanical parts, and the like, the semiconductor devices may include, but are not limited to, various chips, printed circuit boards, semiconductor connectors, and the like, and the image to be detected may include, but is not limited to, images of various chips, images of various printed circuit boards, images of various semiconductor connectors, and the like. As shown in fig. 2, the image to be detected may include a plurality of gold fingers, which are generally distributed at equal intervals. All equally spaced golden fingers can form a golden finger region. Reference pattern areas are typically provided at both ends of the gold finger area. The reference pattern area is generally used for locating the target, and the accuracy of the locating position is high. The golden finger gray scale image can be a ROI (Region of Interest) gray scale image corresponding to each golden finger.

Specifically, optical hardware may be used to capture images of electronic components containing golden fingers. The optical hardware may be a CCD (charge coupled device) or CMOS (complementary metal oxide semiconductor) sensor, or may be a camera. The reference pattern areas at two ends of the golden finger area exist in the image to be detected, so that the golden finger area can be accurately positioned by utilizing the reference pattern areas, and in addition, the golden fingers are distributed at equal intervals in the golden finger area, so that the golden finger gray level image of each golden finger can be extracted in the golden finger area by utilizing the characteristic of equal-interval distribution of the golden fingers.

In some exemplary embodiments, a template matching algorithm may be used to locate the reference pattern areas at both ends. Template matching algorithms are a computer vision technique for finding specific patterns or objects in an image. The basic principle of the algorithm is to compare a small reference image (template) with the input image to determine if there is a region in the input image that matches the template.

S104, determining a first boundary endpoint and a second boundary endpoint of each golden finger according to the reference graph area and the size of each golden finger.

The size of the golden finger can include the horizontal size of the golden finger and the vertical size of the golden finger. The first boundary end point and the second boundary end point may be points where the two ends of the left and right boundary of the golden finger are symmetrical. As shown in fig. 3, the first boundary point may be the A1 point, and the second boundary point may be the A2 point.

Specifically, since the reference pattern area is determined. And a golden finger area is arranged between the two reference pattern areas, and according to the characteristic of equidistant distribution of golden fingers in the golden finger area, the distance between the two reference pattern areas and the size of each golden finger, two endpoints of each golden finger, namely a first boundary endpoint and a second boundary endpoint, can be calculated.

In some exemplary embodiments, taking a certain golden finger as an example, the center coordinates (x 0, y 0) of the certain golden finger can be calculated according to the characteristic of equidistant distribution of the golden fingers in the golden finger area and the distance between the two reference graph areas, if the horizontal dimension of each golden finger is known to be w0 and the vertical dimension is known to be h0, for a certain golden finger, the coordinates of the first boundary end point of the golden finger are (x 0-w0/2, y0+h0/2), and the coordinates of the second boundary end point are (x0+w0/2, y0+h0/2).

S106, dividing a preset number of outlines between the first boundary end point and the second boundary end point of each golden finger, determining the middle position point of each golden finger based on the gray value of the preset number of outlines, and determining the reference value line of the golden finger area based on the middle position point of each golden finger.

The reference line of the golden finger area may be a straight line for determining the golden finger position. The shape of the profile is not limited in some embodiments of the present disclosure and may be rectangular, circular, square, etc. The spacing between each profile may be the same. With continued reference to FIG. 3, L1, L2, and L3 in FIG. 3 may be contours.

Specifically, burrs, bends and the like exist at the edges of the upper end and the lower end of the golden finger due to process problems, and all factors influence the defect detection process. Therefore, in order to accurately determine the position of the gold finger, the reference value line of the gold finger area may be determined first. The first boundary end point and the second boundary end point of each golden finger have been determined in the above steps, so that a preset number of contours can be divided between the first boundary end point and the second boundary end point. The median point of the golden finger is then determined based on the gray value in each contour. In general, the outline with a larger gray value may be the median of the golden finger. After the median point of each golden finger is determined, straight line fitting can be performed on the median point of each golden finger, so that the reference value line of the golden finger area is determined.

S108, determining a bright spot area in the golden finger gray level image by using an area growing algorithm, and calculating the distance from a bright spot contour point of the bright spot area to the reference value line.

The region growing algorithm is a target segmentation algorithm, and is used for combining pixels with similar characteristics in an image into a continuous region or object. The basic principle of the algorithm is to start with seed points and gradually grow or merge pixels with similar features until a complete target area is formed.

The bright spot area may be an area contacted by the probe when the test is performed.

Specifically, a bright spot area in the golden finger gray scale image can be determined by using an area growing algorithm. And then calculating the distance from the bright point contour point of the bright point area to the reference value line. As shown in fig. 4, S is a bright point area, B1 and B2 are two bright point contour points of the bright point area, and a distance D1 from B1 to a reference line may be calculated, and a distance D2 from B2 to the reference line may be calculated. In addition, in general, in order to accurately detect a defect of a gold finger, a maximum value of a distance from a bright point contour point to a reference line is generally selected for processing, and if D1 is greater than D2 and D1 is the maximum value, D1 may be selected for subsequent processing.

S110, performing defect detection on the image to be detected based on the distance from the bright point contour point to the reference value line and a distance threshold value to obtain a defect detection result, wherein the distance threshold value is determined based on the reference value line and the exclusion area of the golden finger area. The exclusion area may be a KOZ area, and the exclusion area (Keep Out Zone, KOZ) on the gold finger refers to an area where other components or wires need to be avoided when designing the gold finger. KOZ is typically a region that defines the area where placement of any other components or wires is prohibited to ensure proper operation and connection of the gold finger. In designing a circuit board, a designer needs to leave enough space for the KOZ to ensure stable connection of the gold fingers while avoiding interference or damage to other components or wires. As shown in fig. 5, K is an exclusion area, and the distance between K and the reference value line can be calculated as a distance threshold.

Specifically, the defect detection of the image to be detected can be performed according to the size relation between the distance from the bright point contour point to the reference value line and the distance threshold value, so that the defect detection result of the golden finger can be obtained. When the distance from the outline point of the bright point to the reference value line is larger than the distance threshold value, the bright point area can be determined to be located in the exclusion area, and in this case, the defect of the golden finger in the electronic component corresponding to the image to be detected is determined. When the distance from the two outline points to the reference value line is smaller than the distance threshold value, the fact that the bright point area is not located in the exclusion area can be determined, and the fact that the golden finger in the electronic component corresponding to the image to be detected has no defect is determined.

In the golden finger defect detection method, an image to be detected is obtained, and a golden finger gray image of a golden finger area in the image to be detected is extracted; determining a first boundary endpoint and a second boundary endpoint of each golden finger according to the reference graph area and the size of each golden finger; dividing a preset number of outlines between a first boundary endpoint and a second boundary endpoint of each golden finger, determining a middle locus of each golden finger based on gray values of the preset number of outlines, and determining a reference value line of a golden finger area based on the middle locus of each golden finger. The position of each golden finger can be accurately positioned by utilizing the datum line of the golden finger area, so that a standard is provided for subsequent judgment, and the detection accuracy is ensured. In addition, a bright spot area in the golden finger gray level image is determined by utilizing an area growth algorithm, and adjacent pixels are gradually expanded and combined into an area according to a predefined growth criterion, so that the bright spot area contacted by a probe is accurately determined, the detection accuracy is further ensured, the defect detection is carried out on the detection image based on the distance from a bright spot contour point of the bright spot area to a reference value line and a distance threshold value, a defect detection result is obtained, and the detection accuracy can be improved compared with a traditional segmentation detection mode.

In one embodiment, as shown in fig. 6, the determining the median point of each golden finger based on the gray values of the preset number of contours includes:

S202, performing Gaussian smoothing on the contours of the preset number to obtain contour smoothing results.

Among them, gaussian smoothing is an image processing technique for removing noise and details in an image, so that the image becomes smoother. Smoothing is performed based on a gaussian function, and noise in the image is suppressed by weighted averaging of each pixel value in the image, while preserving the overall structure and characteristics of the image.

Specifically, gaussian smoothing can be performed on a preset number of contours, and noise and details in the contours of the regions enable the contours to be smoother, so that a contour smoothing result after gaussian smoothing processing is obtained.

S204, calculating a first derivative of the contour smoothing result to obtain a calculation result.

S206, determining the position point in the golden finger indicated by the calculation result which is larger than a preset derivative threshold value.

S208, sorting the position points according to the vertical direction of the golden fingers, and determining the middle position point of each golden finger.

The derivative threshold value can be calculated according to the average value of the derivative at each position of the golden finger area. The location point may typically be a certain pixel point in the contour.

Specifically, a first derivative may be calculated for the smoothed contour smoothing result to obtain a calculation result, where the first derivative is calculated according to the calculation result to calculate a gray level difference between the contour smoothing result and surrounding points, and the larger the gray level difference, the larger the calculation result. Typically, the calculation result is compared with a preset derivative threshold value, a calculation result greater than the derivative threshold value is found, and a location point in the golden finger indicated by the calculation result greater than the derivative threshold value is determined. Typically, one or more location points are defined in each golden finger. When there are a plurality of position points, since the coordinates in the X direction have no influence on the slope of the reference line when the line fitting is performed (when the reference line is determined), the Y direction (vertical direction) may have an influence on the slope of the reference line, and thus the plurality of position points may be sorted in the vertical direction, for example, the position points having small Y coordinate values may be arranged in front, or the position points having large Y coordinate values may be arranged in front, and the middle point of each gold finger may be determined from the sorted position points.

In some exemplary embodiments, for example, 3 contours are obtained, namely, an a contour, a B contour and a C contour, where the a contour is subjected to gaussian smoothing and a first derivative is calculated to obtain an A1 calculation result, the B contour is subjected to gaussian smoothing and a first derivative is calculated to obtain a B1 calculation result, and the C contour is subjected to gaussian smoothing and a first derivative is calculated to obtain a C1 calculation result. If the result greater than the derivative threshold exists in the A1 calculation result, a position point indicated by the result greater than the derivative threshold in the A1 calculation result can be determined. If the position points are A11, A12 and A13, if the position points indicated by the result greater than the derivative threshold value do not exist in the B1 calculation result. And if the position points indicated by the result larger than the derivative threshold in the calculation result of C1 are C11 and C12, the A11, A12, A13, C11 and C12 can be sequenced in the vertical direction, the sequenced sequences A12, A13, A11, C12 and C11 are sequenced, and the position point A11 is taken as the middle position point of the golden finger.

In other exemplary embodiments, points in the calculation result of calculating the first derivative whose absolute value is greater than the derivative threshold may be placed in the boundary point set N; and sorting the points in the point set N according to the y coordinate (vertical direction) to obtain a point set N ', and taking the median point of the N' to obtain the median point of the golden finger.

In this embodiment, the contour is processed through gaussian smoothing, so that the influence caused by noise can be reduced, and the accuracy of calculation is improved. In addition, the first derivative of the contour smoothing result is calculated to obtain a calculation result, and position points in the golden finger indicated by the calculation result which is larger than a preset derivative threshold value are determined, so that the difference positions in the contour can be determined more accurately, and finally the difference positions are sorted according to the vertical direction.

Specifically, since the positions of the electronic components are not fixed, the positions of the gold fingers in the acquired image to be detected are not necessarily horizontal, so that the accuracy of subsequent detection is ensured. The angle of the golden finger region can be positioned based on the reference pattern regions at the two ends, then the rotation correction is carried out on the image to be detected according to the angle, and then the golden finger gray level image of the golden finger region in the image to be detected after the rotation correction is extracted.

In the present embodiment, by correcting an image to be detected using a reference image area, the accuracy of detection can be improved.

The image enhancement method based on gamma conversion is a common image enhancement technology, and the contrast and brightness of the image are adjusted by performing gamma conversion on the pixel value of the image so as to improve the visual effect of the image. The gamma transformation is a nonlinear transformation, and can adjust the gray level of an image, so that the image is more clear and vivid in vision.

Specifically, in general, a trace is left on the golden finger after the signal test, and the gray level of the trace position is higher than that of other regions in the golden finger. But the probe cannot be pressed too deep due to the image acquisition environment and the depth of contact of the probe with the semiconductor device during signal testing. The probes are pressed down and contacted with the golden finger, but the probes are compatible with the fact that a row of probes are contacted with the golden finger, so that the probes on some golden fingers have overvoltage, but the overvoltage cannot be too much, the coating on the golden finger can be damaged due to the too much overvoltage, the contrast between the pricking position and other areas of the golden finger is not particularly obvious, and in order to highlight the pricking position, an image enhancement method based on gamma transformation can be used for processing the gray level image of the golden finger.

In some exemplary embodiments, the golden finger gray scale image may be processed using the following formula:

Wherein r is an input value of the golden finger gray level image after normalization, and the value range is [0,1]. s is the gray output value after gamma conversion. Gamma is a gamma factor that controls the degree of scaling of the overall transform. c is the gray scale factor.

In this embodiment, the image enhancement method based on gamma transformation is used to process the golden finger gray level image, so that the visual effect of the prick in the golden finger gray level image can be improved, and the subsequent defect detection task can be facilitated.

In one embodiment, as shown in fig. 7, the determining the bright spot area in the golden finger gray scale image by using the area growing algorithm includes:

s302, threshold segmentation is carried out on the golden finger gray level image, and a connected domain of a highlight region in the golden finger gray level image is extracted.

Specifically, a proper threshold value can be selected to perform binarization processing on the golden finger gray level image, and a highlight region and a low-light region are separated. The threshold may be determined using a global threshold, an adaptive threshold, or a histogram-based threshold selection method. And carrying out connected domain analysis on the binarized image, and identifying and marking different connected domains in the image. Connected domain analysis may be implemented using functions (e.g., connected Components WITH STATS) provided by an image processing library such as OpenCV. And screening out the connected domain of the highlight region according to the marking information of the connected domain. The filtering can be performed according to the size, shape and other characteristics of the connected domain, so as to ensure that the extracted connected domain accords with the characteristics of the highlight region.

S304, determining a pre-selected point according to the gray extreme point of the connected domain of the highlight region in the golden finger gray image and a preset first gray threshold value.

Wherein the gray-scale extremum point is typically the gray-scale maximum. The pre-selected points may be seed pre-selected points, which are initial points for the region growing algorithm in image processing. In the region growing algorithm, a seed pre-selection point is used as a starting point of growth, and a region or a connected region is formed by gradually expanding the similarity of pixels around the seed point or other rules.

Specifically, the gray value of the connected domain of the highlight region in the golden finger gray image may be calculated, thereby determining the gray extreme point. And comparing the gray extreme point with a preset first gray threshold value, and taking the gray extreme point larger than the first gray threshold value as a preselected point.

S306, determining a target point based on the pre-selected point and a predetermined growth criterion, wherein the growth criterion is determined based on a preset gray level difference threshold, a second gray level threshold, a gradient amplitude and a gradient mean value.

Wherein in the region growing algorithm, the growing criterion is a rule or condition that determines whether a pixel should be added to the current growing region. The growth criterion typically determines whether to add a pixel to the growth area based on similarity between pixels or other specific rules. In some embodiments of the present disclosure, the growth criterion may be determined based on a gray difference threshold, a preset second gray threshold, a gradient magnitude, and a gradient mean.

Specifically, the target point may be determined according to the pre-selected point and the pre-set growth accuracy.

S308, determining a binary image based on the target point and the golden finger gray level image.

S310, determining a bright spot area in the golden finger gray level image according to the gradient mean value of the connected domain in the binary image.

Specifically, after the target point is determined, the target point may be marked in the golden finger gray image, thereby determining a binary image. And searching the connected domain in the binary image, and extracting and analyzing the characteristics of the connected domain. And calculating the gradient mean value of the connected domain, if the gradient mean value is smaller than a set threshold value, determining the connected domain as a noise region, and if the gradient mean value is larger than the set threshold value, determining the connected domain as a bright spot region.

In this embodiment, the pre-selected point is determined according to the gray extreme point of the connected domain of the highlight region in the golden finger gray image and the preset first gray threshold value, and then the target point is determined according to the pre-selected point and the predetermined growth criterion, so that the target point is processed, the accuracy of the finally obtained bright point region is ensured, and the accuracy of defect detection is further ensured.

In one embodiment, the growth criteria include: the gray level difference value between the candidate point and the preselected point is smaller than a preset first gray level difference value. The first gray difference value is obtained according to the segmentation effect when the golden finger gray image is actually obtained, the first gray difference values obtained from different application scenes are unequal, and specific numerical values of the gray difference values are not limited in some embodiments of the present disclosure.

The second criterion is that the gray value of the candidate point is larger than a second gray threshold, the second gray threshold is usually determined according to the average gray in the golden finger gray image, further, the average gray in the golden finger gray image can be calculated, and the average gray and a preset first constant are added to obtain the second gray threshold.

The third criterion is that the gradient magnitude of the candidate point is greater than the gradient mean threshold. Gradient magnitude refers to the magnitude of the gradient at each pixel point in the image, representing the rate of change of gray scale or the rate of change of color of the image at that point. Gradient refers to the rate of change of the gray or color of an image, and the magnitude of the gradient indicates the magnitude of this change. The gradient average value threshold value can be determined according to the gradient average value of each pixel point in the golden finger gray level image. Furthermore, the gradient mean value of each pixel point in the golden finger gray level image can be calculated, and then the gradient mean value is added with a preset second constant to obtain a gradient mean value threshold value.

And the fourth criterion is that the gradient mean value of the second four neighborhoods of the candidate points is larger than the gradient mean value threshold value. The second four-neighborhood refers to four adjacent pixels around one pixel in the image, i.e., pixels in the up, down, left, and right directions. In a two-dimensional image, each pixel has a two-four neighborhood, and the four adjacent pixels and the center pixel together form four neighbors in a two-dimensional plane. Gradient mean generally refers to the average of the gradient magnitudes over a region in an image. The gradient magnitude represents the gradient magnitude at each pixel point in the image, and the gradient mean value is a value obtained by averaging the gradient magnitudes of all pixels in a certain region. The gradient mean value can be used to describe the overall gradient change of a region in the image.

As shown in fig. 8, the determining the target point based on the pre-selected point and a predetermined growth criterion includes:

s402, determining eight neighborhood pixel points of the preselected points in the golden finger gray scale image aiming at each preselected point.

Wherein, eight neighborhood refers to eight adjacent pixels around a pixel in the image, including pixels in eight directions of up, down, left, right, up left, up right, down left, and down right. In a two-dimensional image, each pixel has an eight neighborhood, and the eight adjacent pixels and the center pixel together form eight neighbors in a two-dimensional plane.

Specifically, for each pre-selected point, eight neighborhood pixel points of the pre-selected point in the golden finger gray scale image can be determined.

And S404, taking the pixel points in the eight neighborhood pixel points as candidate points, and determining the candidate points which are indicated by the pre-selected points and meet the conditions in response to the condition that any one of the candidate points meets the first criterion, the second criterion and the third criterion simultaneously or meets the conditions of the first criterion, the second criterion and the fourth criterion simultaneously.

Specifically, each of the eight neighborhood pixel points may be taken as a candidate point. And when a certain candidate point simultaneously meets the conditions of the first criterion, the second criterion and the third criterion or simultaneously meets the conditions of the first criterion, the second criterion and the fourth criterion, determining the candidate point meeting the conditions. And so on until the candidate points meeting the condition indicated by each of the pre-selected points are determined.

S406, determining target points according to the candidate points which are indicated by the each pre-selected point and meet the conditions.

In some exemplary embodiments, a point P1 may be selected from the preselected points, P1 may be placed in the point set Pfg, and the eight neighborhood pixels of P1 may be searched in the golden finger gray scale image. And determining the eight neighborhood pixel points which simultaneously meet the conditions of the first criterion, the second criterion and the third criterion or simultaneously meet the conditions of the first criterion, the second criterion and the fourth criterion as target points. Wherein the initial point does not include the pre-selected point. And so on, the above process is repeated until all the pre-selected points are contained in the point set Pfg, thereby screening out the target points.

In this embodiment, the target point is screened by the first criterion, the second criterion, the third criterion and the fourth criterion, so that a more accurate bright spot area can be obtained, and the accuracy of subsequent defect detection is ensured.

In one embodiment, as shown in fig. 9, the disclosure further provides another golden finger defect detection method, including:

extracting an image:

s502, acquiring an image to be detected, wherein the image to be detected comprises a golden finger area formed by equidistant distribution of a plurality of golden fingers and reference pattern areas positioned at two ends of the golden finger area.

And S504, carrying out rotation correction on the image to be detected based on the reference pattern areas at the two ends, and extracting a golden finger gray image of the golden finger area in the image to be detected after rotation correction.

Confirm the reference value line:

S506, determining a first boundary endpoint and a second boundary endpoint of each golden finger according to the reference graph area and the size of each golden finger.

S508, dividing a preset number of outlines between the first boundary end point and the second boundary end point of each golden finger.

S510, performing Gaussian smoothing on the contours of the preset number to obtain a contour smoothing result.

S512, calculating a first derivative of the contour smoothing result to obtain a calculation result.

S514, determining the position point in the golden finger indicated by the calculation result which is larger than the preset derivative threshold value.

S516, sorting the position points according to the vertical direction of the golden fingers, and determining the middle position point of each golden finger.

Image enhancement:

S518, processing the golden finger gray level image by adopting an image enhancement method based on gamma transformation.

Region growth:

S520, threshold segmentation is carried out on the golden finger gray level image, and a connected domain of a highlight region in the golden finger gray level image is extracted.

S522, determining a pre-selected point according to the gray extreme point of the connected domain of the highlight region in the golden finger gray image and a preset first gray threshold value.

S524, for each pre-selected point, eight neighborhood pixel points of the pre-selected point in the golden finger gray scale image are determined.

S526, taking the pixel points in the eight neighborhood pixel points as candidate points, and determining the candidate points which are indicated by the pre-selected points and meet the conditions in response to the condition that any candidate point meets the first criterion, the second criterion and the third criterion simultaneously or meets the conditions of the first criterion, the second criterion and the fourth criterion simultaneously.

S528, determining the target point according to the candidate points which are indicated by each pre-selected point and meet the condition.

S530, determining a binary image based on the target point and the processed golden finger gray level image.

S532, according to the gradient mean value of the connected domain in the binary image, the bright spot area in the golden finger gray level image is determined.

S534, calculating the distance from the bright point contour point of the bright point area to the reference value line, and carrying out defect detection on the image to be detected based on the distance from the bright point contour point to the reference value line and the distance threshold value to obtain a defect detection result.

Reference may be made to the foregoing embodiments for specific implementation and limitation in this embodiment, and the detailed description is not repeated here.

It should be understood that, although the steps in the flowcharts related to the embodiments described above are sequentially shown as indicated by arrows, these steps are not necessarily sequentially performed in the order indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts described in the above embodiments may include a plurality of steps or a plurality of stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of the steps or stages is not necessarily performed sequentially, but may be performed alternately or alternately with at least some of the other steps or stages.

Based on the same inventive concept, the embodiment of the disclosure also provides a golden finger defect detection device for implementing the above mentioned golden finger defect detection method. The implementation of the solution provided by the device is similar to the implementation described in the above method, so the specific limitation of the embodiment of the device for detecting a golden finger defect provided in the following may be referred to the limitation of the method for detecting a golden finger defect hereinabove, and will not be repeated here.

In one embodiment, as shown in fig. 10, there is provided a golden finger defect detection apparatus 600, comprising: an image processing module 602, a boundary point determination module 604, a reference line determination module 606, a distance calculation module 608, and a defect detection module 610, wherein:

The image processing module 602 is configured to obtain an image to be detected, where the image to be detected includes a plurality of golden finger areas distributed at equal intervals and reference pattern areas located at two ends of the image to be detected, and extract a golden finger gray image of the golden finger areas in the image to be detected;

The boundary point determining module 604 is configured to determine a first boundary endpoint and a second boundary endpoint of each golden finger according to the reference pattern areas located at two ends of the image to be detected and the size of each golden finger;

A reference line determining module 606, configured to divide a preset number of contours between a first boundary endpoint and a second boundary endpoint of each golden finger, determine a middle position point of each golden finger based on a gray value of the preset number of contours, and determine a reference line of the golden finger region based on the middle position point of each golden finger;

the distance calculating module 608 is configured to determine a bright point area in the golden finger gray level image by using a target segmentation algorithm of area growth, and calculate a distance from a bright point contour point of the bright point area to the reference value line;

and a defect detection module 610, configured to detect a defect of the image to be detected based on a distance from the bright point contour point to the reference line and a distance threshold, to obtain a defect detection result, where the distance threshold is determined based on the reference line and an exclusion area of the golden finger area.

In one embodiment of the apparatus, the reference line determination module 606 includes: the middle locus determining module is used for carrying out Gaussian smoothing on the contours of the preset quantity to obtain contour smoothing results; calculating a first derivative of the contour smoothing result to obtain a calculation result; determining a position point in the golden finger indicated by the calculation result which is larger than a preset derivative threshold; and sequencing the position points according to the vertical direction of the golden fingers, and determining the middle position point of each golden finger.

In an embodiment of the device, the image processing module 602 is further configured to perform rotation correction on the image to be detected based on the reference pattern areas at the two ends, and extract a golden finger gray image of a golden finger area in the image to be detected after the rotation correction.

In one embodiment of the apparatus, the apparatus further comprises: and the image enhancement module is used for processing the golden finger gray level image by adopting an image enhancement method based on gamma transformation.

In one embodiment of the apparatus, the distance calculation module 608 includes: the bright point area calculation module is used for carrying out threshold segmentation on the golden finger gray level image and extracting a connected domain of a highlight area in the golden finger gray level image; determining a pre-selected point according to a gray extreme point of the connected domain of the highlight region in the golden finger gray image and a preset first gray threshold value; determining a target point based on the pre-selected point and a predetermined growth criterion, wherein the growth criterion is determined based on a preset gray difference threshold, a second gray threshold, a gradient amplitude and a gradient mean; determining a binary image based on the target point and the golden finger gray level image; and determining a bright spot area in the golden finger gray level image according to the gradient mean value of the connected domain in the binary image.

In one embodiment of the apparatus, the growth criteria include: the gray level difference value between the candidate point and the preselected point is smaller than a preset first gray level difference value; the second criterion is that the gray value of the candidate point is larger than a second gray threshold value, and the second gray threshold value is determined according to the average gray of the golden finger gray image; the third criterion is that the gradient amplitude of the candidate point is larger than a gradient average value threshold value, and the gradient amplitude threshold value is determined according to the gradient average value of each pixel point in the golden finger gray level image; the fourth criterion is that the gradient mean value of the second and fourth neighborhoods of the candidate points is larger than the gradient mean value threshold; the bright spot area calculation module includes: the target point determining module is used for determining eight neighborhood pixel points of the preselected points in the golden finger gray level image aiming at each preselected point; taking the pixel points in the eight neighborhood pixel points as candidate points, and determining candidate points which are indicated by the pre-selected points and meet the conditions in response to the condition that any one of the candidate points meets the first criterion, the second criterion and the third criterion simultaneously or meets the conditions of the first criterion, the second criterion and the fourth criterion simultaneously; and determining the target point according to the candidate points which are indicated by each pre-selected point and meet the condition.

All or part of the modules in the golden finger defect detection device can be realized by software, hardware and a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.

In one embodiment, a computer device is provided, which may be a server, and the internal structure of which may be as shown in fig. 11. The computer device includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is used for storing the image to be detected. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program when executed by the processor is used for realizing a golden finger defect detection method.

In one embodiment, a computer device is provided, which may be a terminal, and the internal structure thereof may be as shown in fig. 12. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless mode can be realized through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. The computer program when executed by the processor is used for realizing a golden finger defect detection method. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, can also be keys, a track ball or a touch pad arranged on the shell of the computer equipment, and can also be an external keyboard, a touch pad or a mouse and the like.

It will be appreciated by those skilled in the art that the structures shown in fig. 11 or 12 are merely block diagrams of portions of structures related to the disclosed aspects and do not constitute a limitation of the computer device on which the disclosed aspects may be applied, and that a particular computer device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.

In one embodiment, a computer device is provided, comprising a memory and a processor, the memory having stored therein a computer program, the processor implementing the steps of any of the method embodiments described above when the computer program is executed.

In one embodiment, a computer readable storage medium is provided, having stored thereon a computer program which, when executed by a processor, implements the steps of any of the method embodiments described above.

In an embodiment, a computer program product is provided comprising a computer program which, when executed by a processor, implements the steps of any of the method embodiments described above.

Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, database, or other medium used in embodiments provided by the present disclosure may include at least one of non-volatile and volatile memory, among others. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high density embedded nonvolatile Memory, resistive random access Memory (ReRAM), magneto-resistive random access Memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric Memory (Ferroelectric Random Access Memory, FRAM), phase change Memory (PHASE CHANGE Memory, PCM), graphene Memory, and the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, and the like. By way of illustration, and not limitation, RAM can be in various forms such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), etc. The databases referred to in the various embodiments provided by the present disclosure may include at least one of a relational database and a non-relational database. The non-relational database may include, but is not limited to, a blockchain-based distributed database, and the like. The processors involved in the embodiments provided by the present disclosure may be general-purpose processors, central processing units, graphics processors, digital signal processors, programmable logic, quantum computing-based data processing logic, etc., without limitation thereto.

The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.

The foregoing examples have expressed only a few embodiments of the present disclosure, which are described in more detail and detail, but are not to be construed as limiting the scope of the present disclosure. It should be noted that variations and modifications can be made by those skilled in the art without departing from the spirit of the disclosure, which are within the scope of the disclosure. Accordingly, the scope of the present disclosure should be determined from the following claims.

Claims

1. The golden finger defect detection method is characterized by comprising the following steps of:

2. The method of claim 1, wherein determining the median point of each golden finger based on the gray values of the predetermined number of contours comprises:

3. The method according to claim 1, wherein the extracting the golden finger gray level image of the golden finger region in the image to be detected includes:

4. The method according to claim 1, wherein after extracting the golden finger gray scale image of the golden finger region in the image to be detected, the method further comprises:

5. The method of claim 1, wherein the determining a bright spot area in the golden finger gray scale image using an area growing algorithm comprises:

6. The method of claim 5, wherein the growth criteria comprises: the gray level difference value between the candidate point and the preselected point is smaller than a preset first gray level difference value; the second criterion is that the gray value of the candidate point is larger than a second gray threshold value, and the second gray threshold value is determined according to the average gray of the golden finger gray image; the third criterion is that the gradient amplitude of the candidate point is larger than a gradient average value threshold value, and the gradient amplitude threshold value is determined according to the gradient average value of each pixel point in the golden finger gray level image; the fourth criterion is that the gradient mean value of the second and fourth neighborhoods of the candidate points is larger than the gradient mean value threshold; the determining a target point based on the pre-selected point and a predetermined growth criterion comprises:

7. A golden finger defect detection device, the device comprising:

8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 6 when the computer program is executed.

9. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 6.

10. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 6.