Wafer bump height detection method and device based on image arc feature extraction, equipment and medium

Abstract

The invention provides a wafer salient point height detection method and device based on image arc feature extraction, equipment and a medium. The method comprises the following steps: acquiring a light strip image formed by salient points of a wafer to be detected and a substrate plane; performing edge extraction on the light strip image to obtain a light strip edge; performing arc feature detection on the edge of the light bar based on Hough transform to obtain a plurality of circles; removing the redundant solutions of the plurality of detected circles; generating a mask region according to the geometric features of the light strip image, and removing the remaining circle after the redundant solution is removed based on the generated mask region to obtain the arc features of the image; and calculating the height of the wafer salient point based on the image arc feature. Therefore, the Hough circle detection result is optimized through non-maximum suppression and a mask strategy, so that accurate arc feature solving is realized, and height measurement of the wafer salient point is realized.

Description

Technical Field

[0001] The present invention relates to a method, device, equipment and medium for detecting the height of wafer bumps based on image arc feature extraction, and relates to the fields of image processing and precision measurement. Background Art

[0002] The wafer surface detection technology based on the optical triangulation method is a common means for measuring the bump height in the field of advanced packaging. In the prior art, a device for measuring the height of micro-bumps based on the optical triangulation principle is disclosed. This device consists of an illumination module, an imaging module and a motion module. The image is obtained by projecting structured light on the bump projection line, and the height is solved by feature extraction of the image. In the prior art, a method for solving the height based on the optical triangulation principle is also disclosed. The illumination module projects and images the bump at -45°, forming a long strip bright spot and a circular bright spot on the imaging detector. By calculating the height value of the bump top from the base plane, the accurate measurement of the bump height is realized.

[0003] In the method for image feature extraction based on the optical triangulation method disclosed in the prior art, the relationship between the bump height, the arc radius and the center pixel coordinates of the circle is established through the derivation of geometric relations. After extracting the arc features of the image, specifically the arc radius and the center pixel coordinates of the circle, the bump height can be solved by substituting into the formula. Using the traditional circle detection algorithm to extract the arc in the bump light strip image has problems such as poor solution stability, easy generation of multiple solutions and complex parameter settings, thus affecting the accuracy of solving the arc radius and the center coordinates, resulting in a large error in the bump height calculation.

[0004] Therefore, how to optimize the circle detection algorithm in the imaging light strip image to obtain arc features with good stability is an urgent problem to be solved in the measurement of the wafer bump height. Summary of the Invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art. For this reason, in view of the above problems, the object of the present invention is to provide a method, device, equipment and medium for detecting the height of wafer bumps based on image arc feature extraction, which can overcome the deficiency of the prior art in the extraction of light strip arc features, can realize accurate solution of arc features, and then accurately solve the height of the bumps.

[0006] In order to achieve the above object of the invention, the technical solution adopted by the present invention is as follows:

[0007] In a first aspect, the present invention provides a method for detecting the height of wafer bumps based on image arc feature extraction, the method comprising:

[0008] Obtaining a light strip image formed by the bumps of the wafer to be measured and the base plane;

[0009] Perform edge extraction on the light strip image to obtain the light strip edge;

[0010] Based on Hough transform, arc feature detection is performed on the edge of the light strip to obtain several circles;

[0011] Remove redundant solutions of several detected circles;

[0012] Generate a mask area according to the geometric features of the light strip image, and remove the remaining circles after removing the redundant solutions based on the generated mask area to obtain the image arc features;

[0013] The height of the wafer bump is calculated based on the arc features of the image.

[0014] In a possible implementation, the Canny operator is used to extract the edge of the light stripe image.

[0015] In a possible implementation, arc feature detection is performed on the edge of the light strip based on Hough transform to obtain several circles including two types: positive examples detected correctly and negative examples detected incorrectly, wherein the positive examples are circles that can better match the light strip concavity in the Hough circle detection.

[0016] In a possible implementation, removing redundant solutions of the detected circles includes:

[0017] The circle detected by Hough transform is generated into a circumscribed square, and whether to suppress it is determined according to the degree of overlap between the circumscribed squares of adjacent circles. Specifically, the degree of overlap is described based on the image intersection-over-union (IOU) of the circumscribed squares of adjacent circles. The non-maximum suppression algorithm is based on comparing the image intersection-over-union (IOU) of the circumscribed squares of the circle with the set threshold to eliminate redundant circles.

[0018] In a possible implementation, a mask area is generated according to the geometric features of the light strip image, and the circles remaining after removing the redundant solutions are eliminated based on the generated mask area to obtain the image arc features, including:

[0019] The grayscale centroid pixel points of the light strip image are calculated row by row to obtain the centroid line segment of the light strip area;

[0020] A parallelogram region is generated with the centroid line segment as the center as the mask region of the image. If the remaining circles after removing the redundant solutions are not completely within the mask region, they are removed.

[0021] In one possible implementation, the height of the wafer bump is calculated based on the image arc feature by measuring the height value H of the top of the bump from the base surface through the information of the distance between the center of the bright spot at the top of the bump and the lowest point of the rectangular bright spot depression in the light strip image and the radius of the arc:

[0022]

[0023] Wherein, x is the distance between the lowest point of the depression of the rectangular bright spot and the center of the circular bright spot on the light bar image, β is the magnification of the imaging module, θ is the incident angle, and r' represents the radius of the circular arc in the light bar image after introducing the magnification.

[0024] In a possible implementation manner, before edge extraction of the light bar image, it further includes steps of performing median filtering and binarization processing on the acquired light bar image and / or performing image dilation and erosion on the light bar image.

[0025] In a second aspect, the present invention provides a wafer bump height detection device based on image circular arc feature extraction, and the device includes:

[0026] An image acquisition unit configured to acquire a light bar image formed by the bumps and the substrate plane of the wafer to be measured;

[0027] An edge extraction unit configured to perform edge extraction on the light bar image to obtain a light bar edge;

[0028] A Hough transform unit configured to perform circular arc feature detection on the light bar edge based on Hough transform to obtain a plurality of circles;

[0029] A redundant circle removal unit configured to remove redundant solutions of the plurality of detected circles;

[0030] A mask unit configured to generate a mask region according to the geometric features of the light bar image, and perform rejection on the remaining circles after removing redundant solutions based on the generated mask region to obtain image circular arc features;

[0031] A height calculation unit configured to calculate the height of the wafer bump based on the image circular arc features.

[0032] In a third aspect, the present invention further provides an electronic device, including:

[0033] At least one processor; and

[0034] A memory communicatively connected to the processor; wherein,

[0035] The memory stores instructions executable by the processor, and the instructions are executed by the processor so that the processor can execute the described method.

[0036] In a fourth aspect, the present invention further provides a computer-readable storage medium storing one or more programs, and the one or more programs include computer instructions for causing a computer to execute any one of the described methods.

[0037] The present invention is applicable to the extraction of circular arc features in images. By optimizing the results of Hough circle detection through non-maximum suppression and masking strategies, circular arc features can be accurately extracted, and it has the following specific advantages:

[0038] 1. The method for image feature extraction and height calculation of wafer bumps in the advanced packaging field using structured light imaging of lines in the present invention avoids the multi-solution problem that occurs in traditional circular arc detection algorithms. Almost no situation of multiple solutions corresponding to the same circular arc will occur, improving the accuracy and stability of circle detection.

[0039] 2. The present invention simplifies the steps of parameter adjustment in traditional circular arc detection algorithms, improving the execution efficiency of the algorithm.

[0040] 3. The present invention has no strict requirements on the size and angle of the imaging light strip. The wafer can be placed arbitrarily within the field of view, and the method will adaptively process it.

[0041] In summary, the present invention provides an efficient circular arc detection method with stable solution and no redundant solutions, thereby realizing accurate solution of circular arc features. These circular arc features can be used to solve the height measurement of bumps and are suitable for the scenario of wafer bump height measurement in industrial production. Description of the Drawings

[0042] By reading the detailed description of the preferred embodiments below, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only for the purpose of showing the preferred embodiments and are not considered to be a limitation of the present invention. Throughout the drawings, the same reference numerals are used to represent the same components. In the drawings:

[0043] Figure 1 is the optical path imaging of wafer bumps based on the optical triangulation method in the embodiment of the present invention.

[0044] Figure 2 is the flowchart of the method in the embodiment of the present invention.

[0045] Figure 3 is a schematic diagram of extracting circular features in an image based on the Hough transform in the embodiment of the present invention.

[0046] Figure 4 is to use non-maximum suppression to eliminate redundant solutions in the embodiment of the present invention. Figure (a) shows the principle of non-maximum suppression, and Figure (b) shows the effect of non-maximum suppression.

[0047] Figure 5 is to use the masking strategy to eliminate the detected negative examples in the example of the present invention. Figure (a) shows the generated parallelogram masking area, and Figure (b) shows the final circle detection result.

[0048] Figure 6In the embodiments of the present invention, the height of the convex point is solved from the circular feature parameters according to the geometric relationship.

[0049] Figure 7 The structure diagram of the electronic device according to the embodiment of the present invention. Specific embodiments

[0050] It should be understood that the terms used herein are for the purpose of describing specific example embodiments only and are not intended to be limiting. Unless the context clearly dictates otherwise, the singular forms "a", "an", and "the" as used herein may also include the plural forms. The terms "comprising", "including", "containing", and "having" are inclusive and thus specify the presence of the stated features, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, steps, operations, elements, components, and / or combinations thereof. The method steps, processes, and operations described herein are not to be construed as necessarily requiring them to be performed in the particular order described or illustrated, unless the order of performance is explicitly stated. It should also be understood that additional or alternative steps may be used.

[0051] Although the terms first, second, third, etc. may be used in this document to describe multiple elements, components, regions, layers, and / or sections, these elements, components, regions, layers, and / or sections should not be limited by these terms. These terms may be used only to distinguish one element, component, region, layer, or section from another. Unless the context clearly indicates otherwise, terms such as "first", "second", and other numerical terms do not imply an order or sequence when used in this document. Thus, the first element, component, region, layer, or section discussed below may be referred to as the second element, component, region, layer, or section without departing from the teachings of the example embodiments.

[0052] For ease of description, spatial relative relationship terms may be used in this document to describe the relationship of one element or feature shown in the figure to another element or feature, such as "inside", "outside", "inner side", "outer side", "below", "above", etc. Such spatial relative relationship terms are intended to include different orientations of the device in use or operation in addition to the orientation depicted in the figure.

[0053] How to optimize the circle detection algorithm in the imaging light bar image to obtain arc features with good stability is an urgent problem to be solved in the measurement of the height of wafer bumps. The wafer bump height detection method, device, equipment and medium based on image arc feature extraction provided by the present invention include: obtaining a light bar image formed by the bumps of the wafer to be measured and the substrate plane; performing edge extraction on the light bar image to obtain the light bar edge; performing arc feature detection on the light bar edge based on the Hough transform to obtain a plurality of circles; removing redundant solutions of the detected plurality of circles; generating a mask region according to the geometric features of the light bar image, and removing the remaining circles after removing the redundant solutions based on the generated mask region to obtain image arc features; calculating the height of the wafer bumps based on the image arc features. Therefore, the present invention optimizes the results of the Hough circle detection through non-maximum suppression and mask strategies, accurately extracts arc features, and realizes the measurement of the height of wafer bumps.

[0054] The exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present invention can be more thoroughly understood and the scope of the present invention can be fully conveyed to those skilled in the art.

[0055] This embodiment provides a wafer bump height detection method based on image arc feature extraction. Before the method is implemented, it is necessary to use an imaging device based on the optical triangulation method to project a line structured light onto the surface of the wafer. After the reflected light is modulated by the bumps on the wafer surface, a light bar image is formed on the photosensitive surface: the line structured light is projected onto the wafer surface at -45°, and there are different reflection situations for the bumps and the substrate on the wafer surface. The imaging system receives the reflected light at 45° and forms a reflected image of the top of the bumps and the wafer substrate on the area array camera. Specifically, an optical triangulation bump height measuring device as shown in Figure 1 can be used. The object to be measured has spherical micro bumps and a substrate plane. The device includes a light source module, an imaging module, and an image detector. The light source module uses an LED and places a diaphragm to form a line structured light, which is obliquely irradiated from the -45° direction onto the spherical micro bumps and the adjacent base surface. The light bars irradiated onto the spherical micro bumps and the adjacent base surface are reflected by the top part of the spherical micro bumps and the adjacent base surface and imaged by the imaging module, and are detected by the image detector. The image detector can use an area array camera.

[0056] In this embodiment, the object to be measured is a wafer, the top of the object to be measured is a hemisphere, and the bottom is a cylinder, and the diameters of the hemisphere and the cylinder are the same. It should be noted that the present invention is also applicable to the detection of the height difference between different shapes and different height positions by the triangulation method. For example, for the detection of the height difference when the top is circular or arc-shaped and the bottom is flat, this is taken as an example and is not limited thereto.

[0057] As shown Figure 1 in the figure, a hemisphere and a cylinder are combined as the object to be measured. The radius of the hemisphere is r and the total height is H. The origin O of the coordinate system is the intersection of the axis of the cylinder and the bottom surface. The bottom plane is the XY plane and the Z axis is consistent with the axis of the cylinder. Both the hemisphere surface and the base surface have good reflectivity. The line structured light is obliquely irradiated on the hemisphere from the -45° direction with a width of d. The optical axis of the microscopic objective lens is located in the XOZ plane and the angle with the X axis is 45°. Assume that one side edge of the light strip is L1, the other side edge is L2, and the intersection edges of the light strip edge and the base are D1 and D2 respectively. The positions of the imaging edges of the light strip on the area array camera are L1' and L2' respectively. The light strip projected onto the base is reflected by the bottom surface and forms a long bright spot on the imaging plane of the area array camera. Due to the light blocking of the hemisphere, an arc-shaped notch appears in the long bright spot, as Figure 1 shown in the image in the upper right area array camera plane in. The light ray l3 is tangent to the hemisphere at T and the intersection point with the XY plane is A, Figure 1 and A" and A' in Figure 1 correspond to the arc vertex and the bottom edge of the rectangle on the area array camera respectively. Due to the limitation of the entrance pupil of the imaging optical path, only the reflected light beams within a small circular area at the top of the hemisphere can enter the imaging optical path, where P is the vertex of the convex point. The light rays formed by the parallel light irradiating the hemisphere surface are imaged onto the area array camera through the imaging optical path, and a circular light spot is formed in the

[0058] As shown Figure 2 in the figure, the wafer height detection method based on arc feature extraction provided in this embodiment includes:

[0059] S1. Obtain the light strip image formed by the spherical micro convex points of the wafer and the base plane.

[0060] S2. Median filtering and binarization.

[0061] In this embodiment, due to mechanical jitter, wafer surface contamination, poor light source uniformity, etc., there are strong noise points in the extracted light strip image, and the pixel gray values in the light strip area fluctuate, increasing the difficulty of subsequent feature extraction. The median filtering method is used to remove the noise areas on the light strip image where the pixel area is smaller than the filter area, and at the same time, pixel binarization is used to set the pixel values higher than the threshold to 255 and those lower than the threshold to 0.

[0062] S3. Perform image dilation and erosion on the filtered light strip image.

[0063] In this embodiment, due to the poor uniformity of the light source, there are small holes in the light bar area, and there are many burrs on the edge of the light bar, which is not conducive to feature extraction. Perform a closing operation on the light bar image to fill the small holes in the light bar and smooth the light bar boundary at the same time. Further denoise the light bar image. Specifically: first perform a closing operation of dilation followed by erosion on the light bar image to be processed, remove the holes in the light bar area, improve the uniformity of the light bar area, and improve the edge of the light bar to make it smoother.

[0064] S4. Extract the edge of the circular light spot after dilation and erosion.

[0065] In this embodiment, the arc feature in the image depends on the edge of the light bar and has nothing to do with the pixels inside the light bar and the background. Use the Canny operator to extract the edge feature information of the light bar, simplify the complex image into a set of edges, accurately locate the edge position while suppressing noise interference, and avoid the influence of the points inside the light bar on the arc detection.

[0066] S5. Hough circle detection: Extract the arc feature based on the Hough transform.

[0067] In this embodiment, the circle detection method based on the Hough transform can detect the circles existing in the target image. The circle detection method based on the Hough transform requires accurately setting the range of the arc radius parameter to be fitted. When there is no prior knowledge of the geometric size of the arc, it is often difficult to set an ideal radius parameter range, resulting in the appearance of a large number of redundant circles. The arc in the image is formed by the occlusion of the convex points to the light bar, and the detected arc contains the geometric information for solving the height of the convex points.

[0068] Furthermore, the circle detection method based on the Hough transform can detect the circles existing in the target image, including the following two types: the positive examples correctly detected and the negative examples wrongly detected. However, in practical applications, the Hough circle detection has problems such as difficult parameter adjustment, easy generation of multiple solutions, and unstable solution, as Figure 3 shown.

[0069] S6. Non-maximum suppression: For the circles detected in S5, set the intersection over union threshold for determining the overlapping area, and use the non-maximum suppression algorithm to remove the redundant solutions in the detection results.

[0070] In this embodiment, an external circumscribed square is generated for the circles detected in S5, and it is judged whether to suppress according to the overlapping degree of the external circumscribed squares of adjacent circles. In order to reduce the situation of multiple solutions, the non-maximum suppression algorithm is introduced to eliminate the redundant solutions of the Hough circle detection. The non-maximum suppression algorithm first calculates the overlap of each pair of detection frames, and then performs affine inference clustering to select an example for each class representing the final detection frame.

[0071] As Figure 4(As shown in (a)), the Hough circle detection contains positive examples that can fit well with the depression of the light strip, and also contains redundant circles as negative examples. The redundant circles obtained by the Hough circle detection generally appear adjacent to the positive examples. The intersection over union (IOU) of the circumscribed squares of the two circles describes the degree of overlap. As Figure 4 (a) The set of squares in the upper right. The middle square has a large overlapping area with the other squares. The positive and negative examples can be judged according to the visualization of the image. As Figure 4 (As shown in (a)), the positive example can fit well with the depression of the light strip, while the negative example cannot. It is eliminated by the non-maximum suppression algorithm. As Figure 4 (a) The set of squares below. The IOU of the two squares is less than the set threshold, and the inscribed circle corresponding to one of them belongs to the correctly detected positive example. The non-maximum suppression algorithm will not eliminate it. The larger the threshold set by the non-maximum suppression algorithm, the higher the tolerance for the overlap between adjacent squares. However, too small a threshold will eliminate the detected positive examples. In this embodiment, the set threshold is 0.1, and the non-maximum suppression algorithm can effectively eliminate some redundant solutions. As Figure 4 (As shown in (b)), taking this as an example, not limited to this.

[0072] S7. Mask elimination: Generate a mask region according to the geometric features of the light strip image, and eliminate the circles located outside the mask region from the remaining circles after executing S6.

[0073] In this embodiment, the centroid line of the light strip is obtained by solving the centroid of the pixel gray value row by row according to the pixel distribution of the image. A parallelogram region is generated with the centroid line as the center as the mask of the image. If the circle is not completely located within the mask region, it is eliminated.

[0074] Furthermore, after non-maximum suppression of the detection results, there are still some cases where the misdetected negative examples are not eliminated. Considering the geometric relationship between the positive examples and the bright spot distribution of the light strip, specifically, the positive examples are distributed between the bright spots at the top of the convex point and the bright spots reflected by the base. As Figure 5 (As shown in (a)), calculate the centroid line segment of the light strip region by calculating the centroid pixel points of the gray value row by row for the image, including:

[0075] Extract the centroid value of the gray value column by column for the image. The ordinate v of the centroid point is the column ordinal number being processed. Set the image height L. The abscissa u of the centroid point is determined by the formula and the obtained centroid points are linearly fitted to obtain the centroid line segment of the gray value:

[0076]

[0077] Further, a parallelogram mask region with a certain redundancy range is generated with the center of gravity line as the geometric center. Specifically, the geometric pixel distance between the bright spot at the top of the convex point on the image and the side of the base bright spot far from the bright spot at the top of the convex point is calculated to determine the width W and height H of the mask region. A tolerance of α is set, and the center of gravity line is translated left and right respectively to obtain a pair of parallel side line segments of the parallelogram, and the upper and lower ends of the two sides are respectively connected to obtain another pair of parallel side line segments, thereby generating the mask region. The positive examples are completely located within the mask region. As Figure 1 and Figure 5 (b), the parallelogram mask strategy can effectively remove the negative examples that are not eliminated by the non-maximum suppression algorithm. In this embodiment, for example, the designed algorithm can effectively obtain 4 stable circular arc solutions. The circular arc solutions are specifically the radius of the circular arc and the pixel coordinates of the center of the circle, which means obtaining the necessary image feature information for solving the spatial height of the convex point, that is, the radius r′ of the circular arc and the pixel coordinates of the center of the circle. The geometric pixel distance between the center P″ of the top bright spot in the image and the center of the circle is added to r′ to obtain the distance P″A″ between the center of the top bright spot and the lowest point of the depression of the base bright spot. This method can reduce the complexity of parameter debugging while ensuring the accuracy of P″A″ and r′.

[0078] S8. Calculation of the height of the wafer bump: According to the image circular arc features obtained in S7, combined with the geometric relationship between the height from the top of the bump to the base surface and the image circle feature parameters, the information on the distance between the center of the top bright spot and the lowest point of the depression of the rectangular bright spot and the radius of the circular arc is accurately extracted, thereby realizing the accurate measurement of the height value of the top of the bump from the base surface and realizing the detection of the height of the wafer bump.

[0079] In this embodiment, as Figure 6 shown, in the XZ plane, based on the geometric derivation of the optical triangulation method, the relationship between the height H of the bump and the distance x between the lowest point A″ of the depression of the rectangular bright spot and the center P″ of the circular bright spot on the image can be obtained. The incident angle is θ. The distance between P″ and A″ in the imaging system is actually the projection of QA, and QA = x / sin2θ. PM is the auxiliary line drawn for calculating PO, which coincides with the incident light of the highest point P of the bump and intersects ET at point S. PM = QA - QR - NA, ST = ET - ES = r - r·sinθ, and ET is the radius r of the bump. QR = PR·tan(90° - 2θ), NA = MN·tanθ, H = PM·cosθ. The relationship between the height H of the bump and x and r is:

[0080]

[0081] When the magnification of the imaging module is β, we have r = r' / β and x = x' / β, where x' represents the distance between the center of the light spot and the lowest point of the depression of the rectangular bright spot in the reflected light image of the bump surface after introducing the magnification, and r' represents the radius of the arc in the light strip image after introducing the magnification. Therefore, we have:

[0082]

[0083] Among them, k = 1 / 2β·sinθ can be obtained through calibration.

[0084] Through the above embodiments, the bump height measurement method provided by the present invention can obtain the bump height value H from the distance P''A'' between the center position of the light spot on the camera imaging surface and the lowest point of the light strip depression and the radius value r' of the circle. This method is simple, has high execution efficiency, can accurately solve the position of the bump height relative to the bottom of the chip, is not limited by the width of the light strip, and has strong operability.

[0085] In summary, in view of the problems existing in the existing optical triangulation system, the present invention proposes a wafer height detection method based on arc feature extraction based on the image features required for wafer bump height calculation. This method obliquely projects line structured light onto the wafer surface, and the reflected light information modulated by the shape of the bumps on the wafer surface is collected by a microscopic objective lens and a area array camera. For the geometric feature information collected by the camera, through image preprocessing, Hough circle detection, non-maximum suppression and mask removal, the information on the distance between the center of the top bright spot and the lowest point of the depression of the rectangular bright spot and the radius of the arc is accurately extracted, thereby realizing the accurate measurement of the height value of the bump top relative to the base surface. Compared with the traditional method, the present invention avoids the problems of complex parameter adjustment, easy generation of multiple solutions, unstable solutions, etc. existing in Hough arc detection. The method is simple, has a small amount of calculation, strong operability, high implementation efficiency, and effectively solves the applicability problem of the traditional triangulation method in microscopic measurement applications.

[0086] Embodiment 2: The above Embodiment 1 provides a method for detecting the height of wafer bumps based on image arc feature extraction. Correspondingly, this embodiment provides a device for detecting the height of wafer bumps based on image arc feature extraction. The device provided in this embodiment can implement the method for detecting the height of wafer bumps based on image arc feature extraction in Embodiment 1, and the device can be implemented in a software, hardware, or a combination of software and hardware manner. For the convenience of description, when describing this embodiment, various units are described separately according to their functions. Of course, in implementation, the functions of each unit can be implemented in the same or multiple software and / or hardware. For example, the device can include integrated or separate functional modules or functional units to execute the corresponding steps in each method of Embodiment 1. Since the device in this embodiment is basically similar to the method embodiment, the description process in this embodiment is relatively simple, and the relevant parts can refer to the partial description in Embodiment 1. The embodiment of the device for detecting the height of wafer bumps based on image arc feature extraction provided by the present invention is only illustrative.

[0087] Specifically, the device for detecting the height of wafer bumps based on image arc feature extraction provided in this embodiment includes:

[0088] An image acquisition unit, configured to acquire a light stripe image formed by the bumps and the substrate plane of the wafer to be measured;

[0089] An edge extraction unit, configured to perform edge extraction on the light stripe image to obtain a light stripe edge;

[0090] A Hough transform unit, configured to perform arc feature detection on the light stripe edge based on the Hough transform to obtain a plurality of circles;

[0091] A redundant circle removal unit, configured to remove redundant solutions of the detected plurality of circles;

[0092] A mask unit, generating a mask region according to the geometric features of the light stripe image, and removing the remaining circles after removing the redundant solutions based on the generated mask region to obtain image arc features;

[0093] A height calculation unit, calculating the height of the wafer bumps based on the image arc features.

[0094] Embodiment 3: This embodiment provides an electronic device corresponding to the method for detecting the height of wafer bumps based on image arc feature extraction provided in Embodiment 1. The electronic device can be an electronic device for a client, such as a mobile phone, a laptop computer, a tablet computer, a desktop computer, etc., to execute the method in Embodiment 1.

[0095] Such as Figure 7As shown, the electronic device includes a processor, a memory, a communication interface, and a bus. The processor, the memory, and the communication interface are connected through the bus to complete communication with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Component (EISA) bus, etc. The memory stores a computer program that can run on the processor. When the processor runs the computer program, it executes the method of Embodiment 1. The implementation principle and technical effects are similar to those of Embodiment 1 and will not be elaborated here. Those skilled in the art can understand that Figure 7 The structure shown in is only a block diagram of some structures related to the solution of this application, and does not constitute a limitation on the computing device to which the solution of this application is applied. The specific computing device may include more or fewer components than those shown in the figure, or combine some components, or have a different component layout.

[0096] In a preferred embodiment, when the logical instructions in the above-mentioned memory can be implemented in the form of software functional units and sold or used as an independent product, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), and optical discs that can store program codes.

[0097] In a preferred embodiment, the processor can be various types of general-purpose processors such as a central processing unit (CPU) and a digital signal processor (DSP), which are not limited here.

[0098] Embodiment 4: This embodiment provides a computer-readable storage medium storing one or more programs. The one or more programs include computer instructions that, when executed by a computer, cause the computer to execute the method provided in Embodiment 1 above.

[0099] Embodiment 5: This embodiment provides a computer program product. The computer program product may be a computer program stored on a computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the method provided in the above Embodiment 1. Its implementation principle and technical effects are similar to those of Embodiment 1 and will not be elaborated here.

[0100] In a preferred embodiment, the computer-readable storage medium may be a tangible device that holds and stores instructions used by an instruction execution device. For example, it may be, but is not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any combination of the above. The computer-readable storage medium stores computer program instructions that cause the computer to execute the method provided in the above Embodiment 1.

[0101] This application is described with reference to the flowcharts and / or block diagrams of methods, devices (apparatuses), and computer program products according to the embodiments of this application. It should be understood that each process and / or block in the flowchart and / or block diagram, and the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0102] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing devices to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device implements the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0103] These computer program instructions can also be loaded onto a computer or other programmable data processing devices, so that a series of operation steps are executed on the computer or other programmable devices to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable devices provide steps for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0104] The various embodiments in this specification are described in a progressive manner. For the same or similar parts among the various embodiments, reference can be made to each other, and the key points of each embodiment are the differences from other embodiments. In the description of this specification, the descriptions with reference to terms such as "a preferred embodiment", "furthermore", "specifically", "in this embodiment", etc. mean that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the embodiments of this specification. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in a suitable manner in any one or more embodiments or examples. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0105] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A wafer bump height detection method based on image arc feature extraction, characterized in that: The method includes: Acquire a light stripe image formed by the bumps and substrate plane of the wafer to be tested; Perform edge extraction on the light strip image to obtain the light strip edge; Based on Hough transform, arc feature detection is performed on the edge of the light strip to obtain several circles; Remove redundant solutions of several detected circles; Generate a mask area according to the geometric features of the light strip image, and remove the remaining circles after removing the redundant solutions based on the generated mask area to obtain the image arc features; The height of the wafer bump is calculated based on the arc features of the image.

2. The wafer bump height detection method based on image arc feature extraction according to claim 1 is characterized in that: The Canny operator is used to extract the edge of the light strip image.

3. The wafer bump height detection method based on image arc feature extraction according to claim 1 is characterized in that: Based on Hough transform, arc feature detection is performed on the edge of the light strip to obtain several circles, including two types: positive examples that are correctly detected and negative examples that are incorrectly detected. Among them, the positive example is a circle that can be well matched with the light strip concave in the Hough circle detection.

4. The wafer bump height detection method based on image arc feature extraction according to claim 1 is characterized in that: The redundant solutions of several detected circles are removed, including: The circle detected by Hough transform is generated into a circumscribed square, and whether to suppress it is determined according to the degree of overlap between the circumscribed squares of adjacent circles. Specifically, the degree of overlap is described based on the image intersection-over-union (IOU) of the circumscribed squares of adjacent circles. The non-maximum suppression algorithm is based on comparing the image intersection-over-union (IOU) of the circumscribed squares of the circle with the set threshold to eliminate redundant circles.

5. The wafer bump height detection method based on image arc feature extraction according to claim 1 is characterized in that: Generate a mask area based on the geometric features of the light strip image, and remove the remaining circles after removing the redundant solutions based on the generated mask area to obtain the image arc features, including: The grayscale centroid pixel points of the light strip image are calculated row by row to obtain the centroid line segment of the light strip area; A parallelogram region is generated with the centroid line segment as the center as the mask region of the image. If the remaining circles after removing the redundant solutions are not completely within the mask region, they are removed.

6. The wafer bump height detection method based on image arc feature extraction according to claim 1 is characterized in that: The height of the wafer bump is calculated based on the image arc feature. The distance between the center of the bright spot at the top of the bump and the lowest point of the rectangular bright spot depression and the radius of the arc is used to measure the height value H of the top of the bump from the base surface: Where x is the distance between the lowest point of the rectangular bright spot depression and the center of the circular bright spot on the light strip image, β is the magnification of the imaging module, θ is the incident angle, and r′ represents the radius of the arc in the light strip image after the magnification is introduced.

7. The wafer bump height detection method based on image arc feature extraction according to claim 1 is characterized in that: Before edge extraction of the light stripe image is performed, the method further includes the steps of median filtering and binarization processing on the collected light stripe image and / or image expansion and erosion on the light stripe image.

8. A wafer bump height detection device based on image arc feature extraction, characterized in that: The device includes: An image acquisition unit is configured to acquire a light stripe image formed by the bumps and the substrate plane of the wafer to be tested; An edge extraction unit is configured to extract the edge of the light strip image to obtain the edge of the light strip; A Hough transform unit is configured to perform arc feature detection on the edge of the light strip based on the Hough transform to obtain a plurality of circles; A redundant circle removal unit is configured to remove redundant solutions of the detected circles; The mask unit generates a mask area according to the geometric features of the light strip image, and removes the remaining circles after removing the redundant solutions based on the generated mask area to obtain the image arc features; The height calculation unit calculates the height of the wafer bump based on the arc features of the image.

9. An electronic device, characterized in that: include: at least one processor; as well as A memory communicatively connected to the processor; wherein, The memory stores instructions executable by the processor, and the instructions are executed by the processor to enable the processor to perform the method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing one or more programs, characterized in that: The one or more programs include computer instructions for causing a computer to execute the method according to any one of claims 1-7.

Images