Method for detecting the height of the probe card relative to the test platform

Abstract

The invention discloses a method for detecting the height of a probe card relative to a test platform. The complete detection algorithm of the invention mainly includes confirmation of the position of the probe in an image, evaluation of the clarity of the probe in the image, and confirmation of the height of the probe; mean filtering is performed on an input image, wherein the size of the filter kernel is determined, and a multiplication operation is performed on the filtered image, and the position of the probe in the image can be obtained by threshold segmentation; Laplace transformation is performed on the image, and the mean and standard deviation of the grayscale value in the image are calculated after addition operation is performed on the obtained image, wherein the image with the largest annotation difference is the best focus image, and the distance between the probe and the chip test platform is calculated; the invention is suitable for rapid detection of the probe position height of chip test equipment in the semiconductor industry, and completes automated detection in the chip test production process; the method can be used to rapidly and accurately identify the probe height, and compared with manual detection, the labor intensity of workers is effectively reduced, and the efficiency of card replacement and production change is improved.

Description

Technical Field

[0001] The invention relates to the technical field of optical detection, and in particular to a method for detecting the height of a probe card relative to a test platform. Background Art

[0002] In the automated testing of chip wafers, if the relative distance between the probe on the probe card and the pad on the chip to be tested is inaccurate, it is easy for the pad on the chip to have poor contact with the probe, causing the originally good product to be judged as defective. If the probe is pressed into the pad too heavily, it will cause damage to the pad, which will make it impossible to guarantee the yield of the product.

[0003] The traditional way to confirm the distance between the probe card and the chip test platform is to take a small amount of chip wafers for manual testing, put the tested chips under a microscope, and manually judge whether the size and depth of the test pressure points on the chip pads are qualified. If they are unqualified, a new batch of chips needs to be taken for manual testing and observation. This easily leads to waste of materials, is time-consuming and labor-intensive, and at the same time, the consistency of the size and depth of the chip test pressure points cannot be guaranteed. Summary of the invention

[0004] The object of the present invention is to provide a method for detecting the height of a probe card probe relative to a test platform, so as to solve the problems raised in the above background technology.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: A method for detecting the height of a probe card relative to a test platform comprises the following steps:

[0006] S1. After receiving the focus request, the system drives the camera up and down through the Z-axis motor to collect images, and the algorithm starts to pre-process the images;

[0007] S2. In the preprocessing work, the image is firstly mean filtered, and the filtered image is multiplied, and then the image is binarized to obtain the image position area where the probe is located;

[0008] S3, after image preprocessing, the position area of ​​the probe in the image can be obtained, and then the clarity of the image in the area is evaluated;

[0009] S4, performing a division operation on the image to further enhance the contrast of the high brightness area of ​​the image and further reduce the contrast of the low brightness area, thereby stretching the image contrast, and finally calculating the mean and standard deviation of the grayscale value of the image;

[0010] S5. Record the standard deviation of the current image and the Z-axis height of the current image. The Z-axis drives the camera from the starting position to the ending position, and takes pictures at intervals of 0.002mm to calculate the standard deviation of the recorded image and the corresponding Z height of the image. Finally, extract the Z height corresponding to the maximum value of the standard deviation from the recorded data set. This height is the optimal focus position H of the current camera lens on the probe. Since the optimal focus distance of the lens itself is known, the height position of the current probe can be directly calculated as H+L.

[0011] Furthermore, in step S1, the industrial camera and the lens are first installed horizontally, and the probe is photographed through a 45-degree prism.

[0012] Furthermore, the light source lens is integrated into one through a 45-degree prism, and the annular light source is installed at one end of the 45-degree prism away from the lens. The annular light source is located below the probe, so that the camera lens is installed horizontally.

[0013] Furthermore, the ring light source illuminates the probe tip and then performs image acquisition processing.

[0014] Furthermore, in step S2, the grayscale value greater than 100 is set to 255, and the grayscale value less than 100 is set to 0, thereby obtaining the image position area where the probe is located.

[0015] Furthermore, in step S2, the image processing used includes mean filtering and multiplication operations; the average filtering value of a point (x, y) on the image is calculated as follows:

[0016]

[0017] n is the size of the filter, P(i,j) is the original image, P(x,y) is the filtered image, and the average grayscale value of all pixels in the pixel point (x,y) window size n*n is calculated.

[0018] Furthermore, in step S2, the image multiplication operation is performed to further enhance the contrast of the high brightness area of ​​the image and further reduce the contrast of the low brightness area. The formula is as follows:

[0019] Z(x,y)=X1(x,y)×X2(x,y) (2)

[0020] X1(x,y) is the grayscale value of the point with coordinates (x,y) in the image, X2(x,y) is the grayscale value of the point with coordinates (x,y) in the image, and Z(x,y) is the product of the two grayscale values.

[0021] Further, in step S3, the method is to take two transformation masks to perform Laplace transform on the image, wherein the masks are:

[0022] Mask 1 is:

[0023] Mask 2 is:

[0024] Furthermore, the image obtained by Laplace transformation is added:

[0025] Z(x,y)=a1X1(x,y)+a2X2(x,y)+…anX n (x,y)+c (3)

[0026] an represents the values ​​around the mask used in the transformation, such as mask 1, mask 2, X n (x, y) represents the grayscale value of the pixel that coincides with the mask value, c represents the value in the middle of the mask, and Z(x, y) represents the position of the mask center at (x, y) in the image. The values ​​in the mask are multiplied by the pixels that coincide with it and then summed.

[0027] Furthermore, in step S4, the mean and standard deviation formulas are as follows:

[0028]

[0029]

[0030] mean is the average grayscale value of all pixels in the image, F is the number of pixels, g(p) is the grayscale value of point p in the image, and Deviation is the standard deviation of the image grayscale value.

[0031] Compared with the prior art, the beneficial effects achieved by the present invention are:

[0032] 1. The present invention integrates the light source lens through a 45-degree prism, so that the camera lens is installed horizontally, reducing the space occupied in the Z direction; the complete detection algorithm mainly includes the confirmation of the position of the probe in the image, the evaluation of the clarity of the probe in the image, and the confirmation of the probe height; the input image is mean filtered with the size of the filter kernel, and the filtered image is multiplied, and the position of the probe in the image can be obtained by threshold segmentation; after the position of the probe in the image is extracted in the previous step, the image of the area where the probe is located is Laplace transformed using a predefined mask, and the mean and standard deviation of the grayscale value in the image are calculated after the obtained image is added, wherein the image with the largest annotation difference is the best focus image, and the Z height H of the position can be obtained according to the focus image index, the best focus distance of the camera lens is fixed L, the Z height position of the probe = H + L, and the distance between the probe and the chip test platform = H + L - H1;

[0033] 2. The present invention collects images through images and motion axes, and confirms the optimal focus position of the probe by algorithm analysis, thereby calculating the distance between the probe and the chip test platform, improving detection efficiency, reducing chip consumables and avoiding damage to the probe during manual confirmation; the present invention is mainly aimed at the probe card installation, disassembly and replacement process in the chip automated test production process. The present invention is suitable for the rapid detection of the probe position height of chip testing equipment in the semiconductor industry, and completes the automated detection in the chip test production process; the method can be used to quickly and accurately identify the probe height. Compared with manual detection, it effectively reduces the labor intensity of workers and improves the efficiency of card replacement and production change. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0035] Figure 1 It is a schematic diagram of the optical scheme for detecting the height position of the probe in the present invention;

[0036] Figure 2 This is a probe defocus optical effect diagram in the present invention;

[0037] Figure 3 This is the optical effect diagram of the best focusing of the probe in the present invention;

[0038] Figure 4 It is the core algorithm flow chart of the present invention;

[0039] Figure 5 It is a schematic diagram of addition operation in the present invention;

[0040] Figure 6 It is a flow chart of the algorithm for evaluating the probe image position and the probe clarity in the image in the present invention;

[0041] Figure 7 It is a schematic diagram of the detection of the probe and chip test platform in the present invention. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] like Figure 1-Figure 7 The method for detecting the height of the probe card relative to the test platform shown includes the following steps:

[0044] S1. After receiving the focus request, the system drives the camera up and down through the Z-axis motor to collect images, and the algorithm starts to pre-process the images; first install the industrial camera and lens horizontally, and take pictures of the probe through a 45-degree prism; integrate the light source lens through a 45-degree prism, and install the annular light source at the end of the 45-degree prism away from the lens. The annular light source is located below the probe, so that the camera lens is installed horizontally; the annular light source illuminates the probe tip, and then the image is collected and processed, and the camera is driven up and down in a small range through the Z axis, and a picture is taken every time a fixed distance is moved, and each height probe image is recorded;

[0045] S2. In the preprocessing work, the image is first mean filtered, and the filtered image is multiplied, and then the image is binarized: the gray value greater than 100 is set to 255, and the gray value less than 100 is set to 0, and the image position area where the probe is located is obtained; the image processing used includes mean filtering and multiplication; mean filtering can suppress the noise in the image (image noise usually refers to other information in the image besides the imaging object, such as spots and particles. These additional erroneous information interferes with the display of the imaging object and affects the imaging quality, so it is often necessary to eliminate these noise points through image filtering). The average filter value calculation formula of the point (x, y) on the image is as follows:

[0046]

[0047] n is the size of the filter, P(i,j) is the original image, P(x,y) is the filtered image, and the average grayscale value of all pixels in the pixel point (x,y) window size n*n is calculated.

[0048] Image multiplication operation further enhances the contrast of high brightness area and further weakens the contrast of low brightness area. The formula is as follows:

[0049] Z(x,y)=X1(x,y)×X2(x,y) (2)

[0050] X1(x,y) is the grayscale value of the point with coordinates (x,y) in the image, X2(x,y) is the grayscale value of the point with coordinates (x,y) in the image, and Z(x,y) is the product of the two grayscale values;

[0051] S3. After image preprocessing, the position area of ​​the probe in the image can be obtained, and then the clarity of the image in the area is evaluated; the method is to take two transformation masks to perform Laplace transform on the image (Laplace transform is a linear space transformation operator independent of direction), where the masks are:

[0052] Mask 1 is:

[0053] Mask 2 is:

[0054] Add the images obtained by Laplace transformation:

[0055] Z(x,y)=a1X1(x,y)+a2X2(x,y)+…anX n (x,y)+c (3)

[0056] an represents the values ​​around the mask used in the transformation, such as mask 1, mask 2, X n (x, y) represents the grayscale value of the pixel that coincides with the mask value, c represents the value in the middle of the mask, Z(x, y) represents the position of the mask center in the image (x, y), and the pixel values ​​in the mask that coincide with it are multiplied and summed;

[0057] S4, performing a division operation on the image obtained after the addition operation, so that the contrast of the high brightness area of ​​the image is further enhanced and the contrast of the low brightness area is further weakened, thereby stretching the image contrast, and finally calculating the mean and standard deviation of the gray value of the image; the formulas for the mean and standard deviation are as follows:

[0058]

[0059]

[0060] Mean is the average grayscale value of all pixels in the image, F is the number of pixels, g(p) is the grayscale value of point p in the image, and Deviation is the standard deviation of the image grayscale value;

[0061] S5. Record the standard deviation of the current image and the Z-axis height of the current image. The Z-axis drives the camera from the starting position to the ending position, and takes pictures at intervals of 0.002mm to calculate the standard deviation of the recorded image and the corresponding Z height of the image. Finally, extract the Z height corresponding to the maximum value of the standard deviation from the recorded data set. This height is the optimal focus position H of the current camera lens on the probe. Since the optimal focus distance of the lens itself is known, the height position of the current probe can be directly calculated as H+L.

[0062] Optical solution for detecting the height position of the probe Figure 1 As shown, the industrial camera and lens are installed horizontally, and the probe is photographed through a 45-degree prism. The ring light source is installed at the other end of the prism. The light source illuminates the probe tip, and the camera is driven up and down in a small range through the Z axis. A photo is taken every time a fixed distance (such as 0.005mm) is moved, and the probe image at each height is recorded. Since the optimal shooting height of the camera lens is fixed (the depth of field of the telecentric lens is within a small range), we only need to control the Z axis to move up and down within a small range. The effect of the optical solution is as follows Figure 2 , 3 As shown, Figure 2 is the defocus position, Figure 3 is the optimal focus position; the core algorithm flow of the present invention is as follows Figure 4 shown.

[0063] The beneficial effects of the present invention are:

[0064] The present invention integrates the light source lens into one by a 45-degree prism, so that the camera lens is installed horizontally, reducing the space occupied in the Z direction; as mentioned above, the complete detection algorithm mainly includes the confirmation of the position of the probe in the image, the evaluation of the probe clarity in the image, and the confirmation of the probe height; Figure 6 , perform mean filtering on the input image, where the size of the filter kernel is (Image Width, 7), perform multiplication on the filtered image, and the grayscale value of the probe area can be stretched to 255, and the position of the probe in the image can be obtained by threshold segmentation; after the probe position in the image is extracted in the previous step, the predefined mask is used to perform Laplace transform on the image of the probe area, and the obtained image is added to calculate the mean and standard deviation of the grayscale value in the image, where the image with the largest annotation difference is the best focus image. According to the focus image index, the Z height H of the position can be obtained. The best focus distance of the camera lens is fixed at L. The Z height position of the probe = H+L, and the distance between the probe and the chip test platform = H+L-H1, such as Figure 7 ;

[0065] The disadvantage of the existing method for confirming the distance between the probe card probe and the chip test platform is that it is inefficient; in the process of adjusting and confirming the height distance between the test platform and the probe, it is necessary to use a test chip for pre-pressure testing and observe the shape and size of the chip test pressure point. This process is not only time-consuming but also easy to cause damage to the probe card probe; the present invention collects images through images and motion axes, and confirms the optimal focus position of the probe by algorithm analysis, thereby calculating the distance between the probe and the chip test platform, improving detection efficiency, reducing chip consumables and avoiding damage to the probe during manual confirmation; the present invention is mainly aimed at the probe card installation, disassembly and replacement process in the chip automated test production process. The present invention is suitable for the rapid detection of the probe position height of chip test equipment in the semiconductor industry, and completes the automated detection in the chip test production process; the method can quickly and accurately identify the probe height, and compared with manual detection, it effectively reduces the labor intensity of workers and improves the efficiency of card replacement and production change.

[0066] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for detecting the height of a probe card relative to a test platform, characterized in that: The following steps are involved: S1. After receiving the focus request, the system drives the camera up and down through the Z-axis motor to collect images, and the algorithm starts to pre-process the images; In step S1, the industrial camera and the lens are first installed horizontally, and the probe is photographed through a 45-degree prism; the light source lens is integrated through the 45-degree prism, and the annular light source is installed at the end of the 45-degree prism away from the lens. The annular light source is located below the probe, so that the camera lens is installed horizontally; the annular light source illuminates the probe tip, and then the image is collected; S2. In the preprocessing work, the image is firstly mean filtered, and the filtered image is multiplied, and then the image is binarized to obtain the image position area where the probe is located; S3. After image preprocessing, the position area of ​​the probe in the image can be obtained, and then the clarity of the image in the area is evaluated; the method is to take two transformation masks to perform Laplace transform on the image, where the masks are: Mask 1 is: ; Mask 2 is: ; Add the images obtained by Laplace transformation: （3）； Represents the values ​​around the mask used for the transformation, such as mask 1 and mask 2. Represents the grayscale value of the pixel that coincides with the mask value, Represents the value in the middle of the mask, Represents the center of the mask in the image Position, the value in the mask is multiplied by the pixels that coincide with it and then summed; S4, performing a division operation on the image to further enhance the contrast of the high brightness area of ​​the image and further reduce the contrast of the low brightness area, thereby stretching the image contrast, and finally calculating the mean and standard deviation of the grayscale value of the image; S5. Record the standard deviation of the current image and the Z-axis height of the current image. The Z-axis drives the camera from the starting position to the ending position, and takes pictures at intervals of 0.002mm to calculate the standard deviation of the recorded image and the corresponding Z height of the image. Finally, extract the Z height corresponding to the maximum standard deviation from the recorded data set. This height is the optimal focus position H of the current camera lens on the probe. Since the optimal focus distance of the lens itself is known, the height position of the current probe can be directly calculated as H+L.

2. The method for detecting the height of the probe card relative to the test platform according to claim 1, characterized in that: In step S2, the grayscale value greater than 100 is set to 255, and the grayscale value less than 100 is set to 0, and the image position area where the probe is located is obtained.

3. The method for detecting the height of the probe card relative to the test platform according to claim 2, characterized in that: In step S2, the image processing used includes mean filtering and multiplication operation; The average filter value is calculated as follows: （1）； is the size of the filter, is the original image, For the filtered image, find the pixel Window size * The average grayscale value of all pixels in .

4. The method for detecting the height of the probe card relative to the test platform according to claim 3, characterized in that: In step S2, the image multiplication operation is performed to further enhance the contrast of the high brightness area of ​​the image and further reduce the contrast of the low brightness area. The formula is as follows: （2）； The coordinates in the image are The gray value of the point, The coordinates in the image are The gray value of the point, is the product of two grayscale values.

5. The method for detecting the height of the probe card relative to the test platform according to claim 1, characterized in that: In step S4, the mean and standard deviation formulas are as follows: （4）； （5）； is the average grayscale value of all pixels in the image. is the number of pixels, For the image The gray value of the point, is the standard deviation of the image grayscale value.

Images