Crystal grain needle print detection method and device, electronic equipment and storage medium

Abstract

The invention provides a crystal grain needle print detection method and device, electronic equipment and a storage medium. The detection method comprises the following steps: determining a bonding pad inner area in a target image comprising a target crystal grain; and determining a boundary area and an area outside the bonding pad based on the position of the area inside the bonding pad in the target image. And determining a first pin print detection result of the pad inner area and a second pin print detection result of the pad outer area. And determining a crystal grain needle print detection result of the target crystal grain based on the first needle print detection result and the second needle print detection result. The detection method is beneficial to improving the accuracy of needle print recognition.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and more specifically to a method for detecting grain pin marks, a device for detecting grain pin marks, an electronic device, a storage medium, and a computer program product. Background Technology

[0002] In semiconductor manufacturing, wafers undergo Circuit Probe Testing (CPT) to screen out functional chips. This test is performed by pressing probes on a probe card onto the chip's pads. During this process, the probes leave tiny indentations on or around the pads, known as "pin marks." Pin mark detection is crucial for ensuring test quality and process stability. Qualified pin marks indicate good contact between the probe and the pad, ensuring accurate test signals. Abnormal pin marks (e.g., misalignment, shape distortion, uneven depth) may indicate problems such as probe card misalignment, probe card wear, or uncontrolled pressure. These issues directly lead to unreliable probe test results and may even damage the chip's circuitry. Therefore, rapid and accurate pin mark detection using Automated Optical Inspection (AOI) technology is essential for improving yield and monitoring equipment status.

[0003] In related technologies, the detection of pin marks typically relies on manual visual inspection or simple optical equipment. However, manual visual inspection is inefficient, highly subjective, and prone to missed detections due to inspector fatigue, failing to meet the pin mark detection requirements of large-scale wafer manufacturing. Simple optical inspection equipment, on the other hand, suffers from insufficient resolution and weak anti-interference capabilities, making it unsuitable for complex scenarios. Therefore, how to efficiently and accurately detect pin marks on wafers is a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0004] The present invention is proposed in view of the above-mentioned problems. The present invention provides a method for detecting grain pin marks, a device for detecting grain pin marks, an electronic device, a storage medium, and a computer program product.

[0005] According to one aspect of the present invention, a method for detecting die pin marks is provided. The method includes: determining a pad-in-region in a target image including a target die, wherein the pad-in-region is the region in the target image corresponding to the interior of the pad in the target die; determining a boundary region and an outer region based on the position of the pad-in-region in the target image, wherein the boundary region is the region in the target image corresponding to the edge of the pad of the target die, and the outer region is the region in the target image corresponding to the exterior of the pad in the target die; determining a first pin mark detection result for the pad-in-region and a second pin mark detection result for the pad-out-region, wherein the first pin mark detection result indicates the position of a first pin mark region corresponding to either the pad-in-region not containing pin marks or the pin mark in the pad-in-region, and the second pin mark detection result indicates the position of a second pin mark region corresponding to either the pad-out-region not containing pin marks or the pin mark in the pad-out-region; and determining a die pin mark detection result for the target die based on the first pin mark detection result and the second pin mark detection result.

[0006] For example, the detection method further includes:

[0007] Acquire a target image including the target grain;

[0008] Determine the pixel correspondence between the target image and the reference image including the reference grains;

[0009] Based on the position and pixel correspondence of the reference region within the pad in the reference image, the corresponding region within the pad in the target image is determined. Here, the reference region within the pad is the region inside the pad in the reference die that corresponds to the reference image.

[0010] For example, determining the boundary region and the region outside the pad based on the position of the region within the pad in the target image includes:

[0011] Based on the position of the region within the pad in the target image and the preset outward expansion threshold, the boundary region is determined, wherein the geometric center of the boundary region is at the same position as the geometric center of the region within the pad;

[0012] The area outside the pads is determined based on the location of the boundary region in the target image.

[0013] For example, if the image brightness of the target image is higher than a brightness threshold, the first pin mark detection result for the region within the pad is determined, including:

[0014] Based on the target image and a preset grayscale threshold, at least one first candidate region is determined, wherein the grayscale value of the pixels included in each first candidate region is less than the preset grayscale threshold.

[0015] Select at least one first candidate region whose area is greater than a preset area threshold, and use it as the first needle print region.

[0016] For example, determining the second pin mark detection result in the area outside the pad includes:

[0017] Based on the image brightness of the target image, a target template image is determined among multiple preset template images. Each of the multiple preset template images has a different image brightness. All of the multiple preset template images include the outside of the pads in the reference die and do not include pin marks. The target template image is the preset template image with the smallest absolute value of the difference between the image brightness of the multiple preset template images and the image brightness of the target image.

[0018] Based on the target image and the target template image, the detection result of the second needle mark in the region outside the pad is determined.

[0019] For example, the first pin mark detection result is used to indicate the position of the first pin mark area corresponding to the pin mark in the area within the pad, and the second pin mark detection result is used to indicate the position of the second pin mark area corresponding to the pin mark in the area outside the pad. Based on the first pin mark detection result and the second pin mark detection result, the grain pin mark detection result of the target grain is determined, including:

[0020] Based on the position of each of the at least one first pinprint region in the target image, at least one first target region among the at least one first pinprint region is determined, wherein the first target region is a first pinprint region that intersects with the edge of the region within the pad.

[0021] Based on the position of each of the at least one second pinprint region in the target image, at least one second target region is determined among the at least one second pinprint region, wherein the second target region is a second pinprint region that intersects with the inner edge of the region outside the pad.

[0022] Based on the positions of at least one first target region and at least one second target region in the target image, the first target region and the second target region belonging to the same needle mark are determined as part of the grain needle mark detection result of the target grain.

[0023] According to another aspect of the present invention, a device for detecting grain needle marks is also provided, the device comprising:

[0024] The pad region determination module is used to determine the pad region in the target image including the target die, wherein the pad region is the region in the target image corresponding to the interior of the pad in the target die;

[0025] The boundary region and pad outer region determination module is used to determine the boundary region and pad outer region based on the position of the pad inner region in the target image. The boundary region is the region in the target image corresponding to the edge of the pad of the target die, and the pad outer region is the region in the target image corresponding to the outer part of the pad in the target die.

[0026] The pin mark detection result determination module is used to determine the first pin mark detection result in the area inside the pad and the second pin mark detection result in the area outside the pad. The first pin mark detection result is used to indicate the location of the first pin mark area corresponding to the area inside the pad where the pin mark is not included or where the pin mark is in the area inside the pad. The second pin mark detection result is used to indicate the location of the second pin mark area corresponding to the area outside the pad where the pin mark is not included or where the pin mark is in the area outside the pad.

[0027] The grain imprint detection result determination module is used to determine the grain imprint detection result of the target grain based on the first imprint detection result and the second imprint detection result.

[0028] According to another aspect of the present invention, an electronic device is also provided. The electronic device includes a processor and a memory. The memory stores computer program instructions, which, when executed by the processor, are used to perform the aforementioned detection method.

[0029] According to another aspect of the present invention, a storage medium is also provided. Program instructions are stored on this storage medium, which, when executed, are used to perform the detection method described above.

[0030] According to another aspect of the present invention, a computer program product is also provided. This computer program product includes computer program instructions that, when executed by a processor, are used to perform the detection method described above.

[0031] According to the above-described scheme of the present invention, the pad region in the target image including the target die can be determined. Then, based on the position of the pad region in the target image, the boundary region and the pad region are determined. Next, the first pin mark detection result of the pad region and the second pin mark detection result of the pad region are determined. Finally, based on the first and second pin mark detection results, the die pin mark detection result of the target die is determined. The above scheme divides the target image of the target die into the pad region, the boundary region, and the pad region, focusing only on the two key regions of the pad region and the pad region for pin mark detection. On the one hand, the above scheme can reduce the redundant computational burden of complex processing of the entire target image, which is beneficial to improving the detection speed and meeting the real-time detection requirements of high-speed production lines. On the other hand, the pad edge itself is an edge with a strong image gradient, which is easily confused with the edge features of adjacent pin marks. The above scheme can avoid the boundary region for pin mark detection, which is beneficial to improving the accuracy and reliability of pin mark recognition. Attached Figure Description

[0032] The above and other objects, features, and advantages of the present invention will become more apparent from the more detailed description of the embodiments of the invention in conjunction with the accompanying drawings. The drawings are provided to further illustrate the embodiments of the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings, the same reference numerals generally represent the same parts or steps.

[0033] Figure 1 A schematic flowchart of a method for detecting grain needle marks according to an embodiment of the present invention is shown;

[0034] Figure 2 A schematic diagram of a target image according to an embodiment of the present invention is shown;

[0035] Figure 3 A schematic diagram of a first needleprint area according to an embodiment of the present invention is shown;

[0036] Figure 4 A schematic diagram of the second needleprint area according to an embodiment of the present invention is shown;

[0037] Figure 5 A schematic diagram showing the first and second needle mark detection results of the target grain according to an embodiment of the present invention is illustrated.

[0038] Figure 6 A schematic block diagram of a device for detecting grain needle marks according to an embodiment of the present invention is shown;

[0039] Figure 7A schematic block diagram of an electronic device according to an embodiment of the present invention is shown. Detailed Implementation

[0040] To make the objectives, technical solutions, and advantages of the present invention more apparent, exemplary embodiments according to the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are merely a part of the embodiments of the present invention, and not all of the embodiments of the present invention. It should be understood that the present invention is not limited to the exemplary embodiments described herein. Based on the embodiments of the present invention described herein, all other embodiments obtained by those skilled in the art without inventive effort should fall within the protection scope of the present invention.

[0041] High-precision automated detection of pin marks faces numerous technical challenges. On one hand, the solder pad area itself is highly reflective and composed of diverse materials (e.g., aluminum, copper, gold, nickel-palladium-gold), and the shape and depth of pin marks vary significantly. Some shallow or tiny pin marks have weak optical signals that are easily masked by background noise. On the other hand, the surrounding area of ​​the solder pads is prone to scratches, residue, and other complex background interference, further increasing the difficulty of pin mark recognition.

[0042] In related technologies, the detection of pin marks typically relies on manual visual inspection or simple optical equipment. However, due to the strong image gradient edges of the pads themselves, their features are easily confused with the features of adjacent pin mark edges. When detecting near the pad edges, the detection methods in related technologies are prone to misclassifying the pad edges as part of the pin mark, or failing to accurately segment the actual pin mark due to the strong interference from the pad edges, further exacerbating the problems of false detection and missed detection of pin marks.

[0043] Therefore, how to efficiently and accurately detect pin marks on wafers is a technical problem that urgently needs to be solved by those skilled in the art.

[0044] To at least partially solve the above problems, embodiments of the present invention provide a method for detecting grain needle marks. Figure 1 A schematic flowchart of a method for detecting grain pin marks according to an embodiment of the present invention is shown. Figure 1 As shown, the method may include steps S110 to S140.

[0045] In step S110, the pad region in the target image including the target die is determined.

[0046] The target die can be a single chip unit that needs to be inspected for pin marks after electrical testing is completed.

[0047] The target image for the target die can be a static image or any video frame from a dynamic video. The target image for the target die can be a raw image captured by an image acquisition device (e.g., a raw image captured by the image sensor in a camera), or an image obtained after preprocessing the raw image (e.g., digitization, normalization, smoothing, etc.). It is understood that... Figure 2 A schematic diagram of a target image according to an embodiment of the present invention is shown. Figure 2 As shown, each target image may include an image of a single die on the wafer.

[0048] The area within a pad refers to the region in the target image corresponding to the interior of the pad in the target die. For example, the pad can be a metal contact area on the target die used for electrical connection, which can contact the probe during electrical testing. The area within the pad can be the set of pixels occupied by the portion within the pad's boundary line in the target image (excluding the pixels corresponding to the pad's boundary line in the target image). The pixels corresponding to the pad's boundary line in the target image can be separately classified into the boundary region in step S120 below to distinguish the area within the pad from the boundary region.

[0049] This invention provides an example for reference. The region within a solder pad in a target image can be determined using template matching. For example, the target image can be matched with an image of a solder pad without defects to determine the region within the solder pad in the target image. Specifically, a reference image of the solder pad region (i.e., the image of the aforementioned solder pad without defects) is placed at the upper left corner of the target image. Then, the image is slid across the target image, comparing each possible position. For each position in the target image, the similarity (e.g., normalized cross-correlation coefficient) between the reference image of the solder pad region and the local region of the target image is calculated, resulting in a result matrix. Each element in the result matrix represents the degree of matching between the reference image of the solder pad region and the corresponding position in the target image. With the result matrix obtained, the `cv2.minMaxLoc()` function can be used to analyze the result matrix and find the position in the target image with the highest matching degree to the reference image of the solder pad region. The region corresponding to this position with the highest matching degree can be considered as the solder pad region in the target image. Alternatively, a trained region detection model can be used to detect the solder pad region in the target image. The set of pixels within the boundary line of the pad region in the target image can be considered as the pad-in-region. The pixels on the boundary line of the pad region correspond to the pixels on the pad boundary line in the target image; these pixels may not belong to the pad-in-region but rather to the boundary region in step S120 below. In other words, the boundary line of the boundary region may be adjacent to the boundary line of the pad-in-region, but they do not overlap.

[0050] In step S120, the boundary region and the region outside the pad are determined based on the position of the region inside the pad in the target image.

[0051] The boundary region is the area in the target image corresponding to the edge of the pad of the target die.

[0052] In one example, it can be like Figure 2 As shown, based on the determined pad region in the target image, the corresponding region of the pad on the target die in the target image (hereinafter referred to as the pad region) can be determined by edge detection methods (e.g., Canny edge detection algorithm, etc.). Using the boundary line of the pad region as a reference, a first preset pixel width (e.g., 2 pixels, which can be determined according to the actual size of the pad) is extended away from the pad region. The region obtained after the above extension operation, as well as the boundary line of the pad region, are both considered as the boundary region in the target image.

[0053] In yet another example, it can be like Figure 2 As shown, based on the determined pad region in the target image, the width of the boundary region can be set to a first preset pixel width (the value of the first preset pixel width can be the same as above). Here, we take an example where the first preset pixel width is 2 pixels. By expanding the pad region according to the first preset pixel width, we can extend it outwards by 2 pixels from the boundary line of the pad region, obtaining the expanded region (i.e., the expanded region includes the pad inner region and the boundary region). After removing the pad inner region from the expanded region, we obtain the remaining region. The pixels in this remaining region can include the pixels on the pad boundary line. This remaining region is used as the boundary region. The pad outer region is the region in the target image corresponding to the outside of the pad in the target die. For example, as... Figure 2 As shown, based on the determination of the area within the pads and the boundary area, the area in the target image other than the area within the pads and the boundary area can be regarded as the aforementioned area outside the pads. It can be understood that the boundary line of the area outside the pads near the boundary area can be adjacent to the boundary line of the boundary area near the area outside the pads, but the two do not overlap.

[0054] In step S130, the first pin mark detection result in the area inside the pad and the second pin mark detection result in the area outside the pad are determined.

[0055] The first pin mark detection result is used to indicate whether the area within the pad does not contain pins or the location of the first pin mark area corresponding to a pin mark within the pad area. For example, if pins exist within the pad area, each pin within the pad area can be considered a first pin mark area. The location of each first pin mark area (which can be represented by pixel coordinates, etc.) is then used as the first pin mark detection result for that pad area. If no pins exist within the pad area, the first pin mark detection result indicates that the pad area does not contain pins. For example... Figure 3 A schematic diagram of a first needleprint area according to an embodiment of the present invention is shown. Figure 3 As shown, there are first pin marks A1, first pin marks A2, and first pin marks A3 within the pad region. A trained first pin mark detection model can determine whether a pin mark exists within the pad region and the location of the corresponding first pin mark region within the pad region. This first pin mark detection model can be used to output whether a first pin mark region exists within the pad region and the location of each first pin mark region based on the input image of the pad region. This embodiment of the invention provides a training method for the first pin mark detection model for reference. The training process of the first pin mark detection model may include: obtaining a first training dataset; iteratively training the first pin mark detection model using the first training dataset and adjusting the model parameters of the detection model using a loss function until training is complete. The first training dataset includes multiple images of training pad regions, each image labeled with whether a pin mark exists within the training pad region. If a pin mark exists in the training pad region of an image, the location of the pin mark within the training pad region needs to be simultaneously labeled. The conditions for completing the training may include the loss value calculated by the loss function stabilizing and the number of iterations reaching a preset number, etc. Based on the above training process, a first pinprint detection model for detecting areas within solder pads can be obtained. It can be understood that the images of the areas within solder pads used for training the first pinprint detection model can be used as examples of such images.

[0056] The second pin mark detection result is used to indicate the location of the second pin mark area corresponding to either the area outside the pad not containing pin marks or the area outside the pad containing pin marks. For example, Figure 4 A schematic diagram of a second needleprint area according to an embodiment of the present invention is shown. Figure 4As shown, there are second pin marks B1 and B2 outside the pad area. Similarly, the trained second pin mark detection model can determine whether pin marks exist outside the pad area and the location of the corresponding second pin mark area. This second pin mark detection model can be used to output whether a second pin mark area exists outside the pad area and the location of each second pin mark area based on the input image of the pad area outside the model. The training process of the second pin mark detection model is the same as that of the first pin mark detection model. When training the second pin mark detection model, a second training dataset can be used. This second training dataset includes multiple images of the pad area outside the training area. Each image is labeled with whether a pin mark exists outside the pad area. If a pin mark exists in the training pad area outside the image, the location of the pin mark outside the training pad area needs to be labeled simultaneously. It can be understood that the images of the pad area outside the training area can be the images of the pad area outside the training area used to train the second pin mark detection model.

[0057] It is understandable that the first and second pinprint detection models mentioned above can be the same model. This means that a single model (hereinafter referred to as the target pinprint detection model) can simultaneously detect pinprints within the pad region (i.e., the first pinprint region) and pinprints outside the pad region (i.e., the second pinprint region) in the target image. The training process for the target pinprint detection model is the same as that for the first pinprint detection model. A third training dataset can be used when training the target pinprint detection model. This third training dataset includes multiple training die images (e.g., complete die images). Each training die image is labeled with the presence or absence of pinprints in the pad region and the pad region outside the training die image. If pinprints exist in the pad region and / or the pad region outside the training die image, it is necessary to simultaneously label the position of the pinprint in the training die image and whether the pinprint belongs to the first or second pinprint region. Understandably, the boundary region in the target image input to the target needle print detection model can be masked (i.e., a mask can be created to set the boundary region to 0), and the masked image can be input into the target needle print detection model.

[0058] In step S140, based on the first needle mark detection result and the second needle mark detection result, the grain needle mark detection result of the target grain is determined.

[0059] This invention provides an example for reference. See also: Figure 3 as well as Figure 4 As shown, in Figure 3 as well as Figure 4Given images of the area within the pads and the area outside the pads of the same target die, the first pin mark detection result for the target die can be determined as the positions of the first pin mark area A1, the first pin mark area A2, and the first pin mark area A3. The second pin mark detection result for the target die is the positions of the second pin mark area B1 and the second pin mark area B2. Since in normal electrical testing, probes only leave pin marks within the pads and not outside the pads, an abnormal pin mark position exists in this target die, and this "abnormal pin mark position" can be considered the die pin mark detection result for this target die. Furthermore, the die pinning test result of the target die can be determined based on factors such as whether the area of ​​each first pinning region of the target die is greater than a preset threshold (which can be used to determine whether the size of the first pinning region is abnormal), whether there is a first pinning region in the pad area of ​​the first pinning test result (which can be used to determine whether there is a missing first pinning region), whether the number of first pinning regions is within a preset range (which can be used to determine whether there is an excessive number of first pinning regions), and whether the area of ​​the second pinning region in the second pinning test result is greater than a preset threshold (which can be used to determine whether the size of the second pinning region is abnormal). Taking the determination of the die pinning test result of the target die based on whether the area of ​​each first pinning region of the target die is greater than a preset threshold as an example, when the area of ​​a first pinning region is greater than the preset threshold, it can be determined that there is excessive probe pressure during electrical testing in the target die, and "abnormal pinning size" can be used as the die pinning test result of the target die. It should be understood that if both "abnormal needle position" and "abnormal needle size" exist simultaneously, then both "abnormal needle position" and "abnormal needle size" can be used as the results of the grain needle detection.

[0060] According to the above-described scheme of the present invention, the pad region in the target image including the target die can be determined. Then, based on the position of the pad region in the target image, the boundary region and the pad region are determined. Next, the first pin mark detection result of the pad region and the second pin mark detection result of the pad region are determined. Finally, based on the first and second pin mark detection results, the die pin mark detection result of the target die is determined. The above scheme divides the target image of the target die into the pad region, the boundary region, and the pad region, focusing only on the two key regions of the pad region and the pad region for pin mark detection. On the one hand, the above scheme can reduce the redundant computational burden of complex processing of the entire target image, which is beneficial to improving the detection speed and meeting the real-time detection requirements of high-speed production lines. On the other hand, the pad edge itself is an edge with a strong image gradient, which is easily confused with the edge features of adjacent pin marks. The above scheme can avoid the boundary region for pin mark detection, which is beneficial to improving the accuracy and reliability of pin mark recognition.

[0061] For example, the above-mentioned method for detecting grain needle marks may further include steps S210 to S230.

[0062] In step S210, a target image including the target grain is obtained.

[0063] The acquisition of the target image can be found in step S110, and will not be described in detail here.

[0064] In step S220, the pixel correspondence between the target image and the reference image including the reference grain is determined.

[0065] The reference grain can be the grain before the electrical test, which is free of defects such as pin marks, foreign matter, and scratches.

[0066] This invention provides an example for reference. At least one feature point can be extracted from both a reference image and a target image based on a feature extraction algorithm, and a descriptor for each feature point can be generated. Then, based on the descriptor of each feature point and a feature point matching algorithm (e.g., nearest neighbor search algorithm, brute-force matching algorithm, etc.), at least one pair of feature points is determined. Each pair of feature points consists of a feature point on the target image and a feature point on the reference image; the two feature points in each pair can be used to represent points at the same physical location in the target image and the reference image. Based on each pair of feature points, the geometric transformation relationship between the reference image and the target image is determined. With the objective of minimizing reprojection error, a random sampling consensus algorithm is used to eliminate abnormal feature point pairs (e.g., feature point pairs incorrectly matched due to pad reflection, etc.), determining the optimal transformation matrix corresponding to the geometric transformation relationship between the reference image and the target image. This optimal transformation matrix represents the pixel mapping relationship from the reference image to the target image (i.e., the pixel correspondence between the target image and the reference image). Any pixel coordinate in the reference image can be mapped to the corresponding pixel coordinate in the target image through this optimal transformation matrix. It is understood that the aforementioned feature extraction algorithms may include Scale-Invariant Feature Transform (SIFT), Speeded-Up Robust Features (SURF), etc. An appropriate feature extraction algorithm can be selected to extract at least one of the aforementioned feature points based on the actual situation. It is understood that the aforementioned reprojection error refers to the pixel distance between the theoretical pixel coordinates of the feature point in the reference image in the target image (i.e., the pixel coordinates obtained through the aforementioned optimal transformation matrix mapping) and the pixel coordinates of the actual feature point matched by that feature point in the target image.

[0067] In step S230, based on the position of the reference region within the pad in the reference image and the pixel correspondence, the pad region corresponding to the reference region within the pad in the target image is determined.

[0068] The reference region within a pad is the area in the reference image corresponding to the interior of the pad in the reference die. The pad can be a metal contact area on the reference die used for electrical connection, which can contact the probe during electrical testing. The reference region within a pad can be the set of pixels occupied by the portion within the pad's boundary line in the reference image, excluding the pixels corresponding to the pad's boundary line in the reference image.

[0069] This invention provides an example for reference. Taking the above pixel correspondence as an example to represent the pixel coordinates (x, y) in the reference image corresponding to the pixel coordinates (x+m, y+n) in the target image, when the reference area within the pad is determined to be a rectangular area, and the vertex pixel coordinates of the rectangular area are (1, 1), (1, 5), (5, 1), (5, 5), it can be determined that the area within the pad corresponding to the reference area in the target image is a rectangular area with vertex pixel coordinates (1+m, 1+n), (1+m, 5+n), (5+m, 1+n), (5+m, 5+n).

[0070] According to the above-described scheme of the present invention, a target image including the target die can be acquired. Then, the pixel correspondence between the target image and a reference image including the reference die is determined. Subsequently, based on the position of the reference region within the pad in the reference image and the pixel correspondence, the corresponding pad region in the target image is determined. This scheme improves the accuracy of pad region determination, thereby ensuring the reliability of the first pinprint detection result and improving pinprint recognition accuracy. Furthermore, this scheme determines the pad region through a lightweight image registration process, is simple to operate, and can be applied to large-scale wafer inspection, improving wafer testing efficiency and product quality, and providing reliable technical support for improving yield in the semiconductor manufacturing process.

[0071] For example, step S120 above, which determines the boundary region and the region outside the pad based on the position of the region inside the pad in the target image, may include steps S121 and S122.

[0072] In step S121, the boundary region is determined based on the position of the region within the pad in the target image and a preset outward expansion threshold.

[0073] This invention provides an example for reference. Based on the position of the region within the pad in the target image, the region within the pad can be expanded outward according to a preset expansion threshold. The expanded region is taken as the boundary region and the total region of the region within the pad. Pixels corresponding to the region within the pad are removed from all pixels corresponding to the total region, and the region corresponding to the remaining pixels is taken as the boundary region. It is understood that the preset expansion threshold can be determined according to actual conditions. It is understood that the geometric center of the boundary region is at the same position as the geometric center of the region within the pad. For example, Figure 2 The pixel coordinates of the geometric center of the boundary region are (4, 4), so the pixel coordinates of the geometric center of the region inside the pad are (4, 4). In step S122, the region outside the pad is determined based on the position of the boundary region in the target image.

[0074] For details, please refer to the section on determining the area outside the pads in step S120 above; the present invention will not elaborate further here.

[0075] According to the above-described scheme of the present invention, the boundary region can be determined based on the position of the region within the pad in the target image and a preset outward expansion threshold. Then, the region outside the pad is determined based on the position of the boundary region in the target image. The above scheme helps to improve the accuracy of determining the region within the pad, the boundary region, and the region outside the pad, thereby ensuring the reliability of the first and second pinprint detection results, and thus improving the accuracy of pinprint recognition.

[0076] For example, the image brightness of the target image is higher than the brightness threshold. The determination of the first pin mark detection result in the region within the pad in step S130 above may include steps S131a and S132a.

[0077] To reduce the impact of textures on the target grain background (e.g., lattice patterns of the grain substrate, micro-scratches on the surface of the grain substrate, light and dark stripes formed by material inhomogeneity of the grain substrate, etc.) on the first imprint detection result, the target grain can be illuminated by diffuse lighting (e.g., integrating sphere lighting), uniform surface light, or low-angle soft light when acquiring the target image, so as to reduce the contrast and detail clarity of the aforementioned textures and reduce their interference with the first imprint detection result.

[0078] In step S131a, at least one first candidate region is determined based on the target image and a preset grayscale threshold.

[0079] At least one first candidate region includes pixels whose corresponding grayscale values ​​are all less than a preset grayscale threshold. For example, a color target image can be converted to a grayscale image using a grayscale conversion algorithm, thereby obtaining a target grayscale image corresponding to the target image. Then, based on the grayscale value of each pixel in the target grayscale image, pixels in the target grayscale image whose grayscale values ​​are less than the preset grayscale threshold can be determined. According to the pixel correspondence between the target image and the target grayscale image (e.g., a one-to-one correspondence), the pixel coordinates of the pixels in the target grayscale image whose grayscale values ​​are less than the preset grayscale threshold are determined, thereby identifying the pixels in the target image whose grayscale values ​​are less than the preset grayscale threshold. Then, connected component analysis can be performed on the pixels in the target image whose grayscale values ​​are less than the preset grayscale threshold, aggregating adjacent pixels into several connected regions, each of which is a first candidate region. Each pixel located in a first candidate region has a grayscale value less than the preset grayscale threshold. It is understood that the preset grayscale threshold can be determined according to actual conditions.

[0080] In step S132a, at least one first candidate region with an area greater than a preset area threshold is selected as the first needle imprint region.

[0081] This invention provides an example for reference. Taking pixels as the unit of a preset area threshold, if the total number of pixels in the first candidate region is greater than the preset area threshold, the first candidate region can be determined as the first pin mark region. The preset area threshold can be determined according to the actual situation. It is understood that if the unit of the preset area threshold is a unit other than pixels, the area of ​​the first candidate region can be determined according to the correspondence between pixels and that unit, and the total number of pixels in the first candidate region. Specifically, for example, if the unit of the preset area threshold is square micrometers, the area of ​​the first candidate region can be determined according to the physical area corresponding to each pixel and the total number of pixels in the first candidate region (i.e., the product of the total number of pixels in the first candidate region and the physical area corresponding to each pixel). It is understood that when a first pin mark region exists, the position of each first pin mark region is the first pin mark detection result. When no first pin mark region exists, the area within the pad excluding pin marks is the first pin mark detection result.

[0082] According to the above-described scheme of the present invention, when the image brightness of the target image is higher than a brightness threshold, at least one first candidate region can be determined based on the target image and a preset grayscale threshold. Then, first candidate regions with an area greater than a preset area threshold are selected from the at least one first candidate region as the first imprint region. This scheme can reduce the influence of texture on the target grain background on the first imprint detection result, thereby ensuring the accuracy of the first imprint detection result and improving the accuracy of imprint recognition.

[0083] For example, determining the second pin mark detection result in the area outside the pad in step S130 above may include steps S131b and S132b.

[0084] In step S131b, based on the image brightness of the target image, a target template image is determined from among multiple preset template images.

[0085] In one example, the arithmetic mean of the grayscale values ​​of all pixels in the target grayscale image can be calculated based on the target image. The image brightness of the target image can then be determined based on this arithmetic mean. For example, the arithmetic mean can be used as the image brightness.

[0086] In another example, the perceived brightness of each pixel in the target image can be determined based on the color value (i.e., RGB value) corresponding to each pixel in the target image and the standard grayscale conversion formula (i.e., perceived brightness equals 0.299 * red component + 0.587 * green component + 0.114 * blue component). The image brightness of the target image can then be determined based on the arithmetic mean of the perceived brightness of all pixels in the target image.

[0087] Multiple preset template images each have different image brightness. All preset template images include the exterior of the pads in the reference die, but none include pin marks. For example, the image brightness determination process for each preset template image can be the same as the image brightness determination process for the target image described above, and will not be elaborated upon here. Preset template images with different image brightness can be obtained by illuminating the reference die with different brightness levels.

[0088] The target template image is the preset template image with the smallest absolute value of the difference between the brightness of the target image and the brightness of the image in a plurality of preset template images. For example, the absolute value of the difference between the brightness of each preset template image and the brightness of the target image can be determined. The preset template image with the smallest absolute value of the difference between the brightness of the target image and the image brightness of the image in the preset template image is taken as the target template image corresponding to that target image.

[0089] In step S132b, based on the target image and the target template image, the detection result of the second pin mark in the area outside the pad is determined.

[0090] This invention provides an example for reference. The second pin mark detection result in the outer region of the pads can be determined based on the pixel value of each pixel in the outer region of the target image and the pixel value of each pixel in the outer reference region (i.e., the region outside the pads in the reference die corresponding to the corresponding pixel in the target template image). Specifically, for each pixel in the outer region of the pads of the target image, based on the pixel correspondence between the target image and the target template image (the process for determining this pixel correspondence can refer to the process for determining the pixel correspondence between the target image and the reference image in step S220 above), the pixel corresponding to that pixel in the outer reference region (hereinafter referred to as the reference pixel) is determined. The difference between the pixel value of the reference pixel and the pixel value of the target pixel is calculated. If the absolute value of this difference is greater than a preset difference, the pixel is marked as a second pin mark pixel. All second pin mark pixels in the outer region of the pads can be determined through the above process. Connectivity analysis can be performed on all second pin mark pixels in the outer region of the pads to aggregate adjacent second pin mark pixels into several connected regions, each of which is a second candidate region. The second pin mark detection result in the area outside the pad can be determined based on all the second candidate areas. For example, a second candidate area larger than a preset area threshold can be used as a second pin mark area. Alternatively, each second candidate area can be considered as a second pin mark area. It is understood that when a second pin mark area exists, the position of each second pin mark area is the second pin mark detection result. When no second pin mark area exists, the area outside the pad excluding pin marks is the second pin mark detection result. It is also understood that the process for determining the second candidate area can be the same as the process for determining the first candidate area in step S131a.

[0091] According to the above-described scheme of the present invention, a target template image can be determined from multiple preset template images based on the image brightness of the target image. Then, based on the target image and the target template image, the second pin mark detection result in the area outside the pad is determined. The above scheme adaptively matches the preset template image based on the brightness of the target image, effectively reducing the interference of illumination changes on pin mark detection, which is beneficial to improving detection accuracy and stability, and thus can improve chip yield and process stability.

[0092] For example, the first pin mark detection result is used to indicate the position of the first pin mark area corresponding to the pin mark in the area inside the pad, and the second pin mark detection result is used to indicate the position of the second pin mark area corresponding to the pin mark in the area outside the pad. The above step S140, based on the first pin mark detection result and the second pin mark detection result, determines the grain pin mark detection result of the target grain, and may include steps S141 to S143.

[0093] In step S141, based on the position of each of the at least one first needle imprint region in the target image, at least one first target region among the at least one first needle imprint region is determined.

[0094] The first target area is the first pin mark area that intersects with the edge of the area within the solder pad. For example, Figure 5 A schematic diagram illustrating the first and second needle mark detection results of a target grain according to an embodiment of the present invention is shown. Figure 5 As shown, the edges of the first pin-printed area A1 and the first pin-printed area A2 with the area inside the pad (refer to...) Figure 3 The boundary lines S1 and A2 intersect. Therefore, the first needle print area A1 and the first needle print area A2 can be defined as a first target area.

[0095] This invention provides an example for reference. Taking the position of the first pinprint region in the target image as represented by the pixel coordinates of each pixel within the first pinprint region as an example, for each first pinprint region within the pad area, the presence of pixels located at the edge of the pad area within the first pinprint region can be determined based on the pixel coordinates of each pixel within the first pinprint region. If pixels located at the edge of the pad area exist within the first pinprint region, the first pinprint region can be defined as a first target region.

[0096] In step S142, based on the position of each of the at least one second needle imprint region in the target image, at least one second target region among the at least one second needle imprint region is determined.

[0097] The second target area is the second pin mark area that intersects with the inner edge of the area outside the solder pad. For example, such as Figure 5 As shown, the second pin mark area B2 and the inner edge of the area outside the pad (refer to...) Figure 3 The boundary lines S2 and B2 intersect. Therefore, the second needle mark area B2 can be identified as a second target area.

[0098] This invention provides an example for reference. Taking the position of the second pinprint region in the target image as represented by the pixel coordinates of each pixel in the second pinprint region as an example, for each second pinprint region outside the pad area, it can be determined whether there are pixels located at the inner edge of the outer pad area in the second pinprint region based on the pixel coordinates of each pixel in the second pinprint region. If there are pixels at the inner edge of the outer pad area in the second pinprint region, the second pinprint region can be defined as a second target region.

[0099] In step S143, based on the positions of at least one first target region and at least one second target region in the target image, the first target region and the second target region belonging to the same needle mark are determined as part of the grain needle mark detection result of the target grain.

[0100] In one example, for each first target region, pixels located on the boundary line of that first target region (hereinafter referred to as first boundary pixels) can be determined. The Euclidean distance between each first boundary pixel and each second boundary pixel (i.e., pixels located on the boundary line of a second target region) is calculated. The second target region corresponding to a second boundary pixel whose Euclidean distance to the first boundary pixel is less than a preset distance is considered as a second target region belonging to the same needle mark as the first target region.

[0101] In another example, each first target region and each second target region can be extended outward by a second preset pixel width (for example, the second preset pixel width can be the same as or greater than the pixel width of the boundary region). A first target expansion region obtained by extending each first target region outward by the second preset pixel width, and a second target expansion region obtained by extending each second target region outward by the second preset pixel width, are determined. For each first target expansion region, it can be determined whether there exists a second target expansion region whose overlapping area with the first target expansion region is greater than a preset overlap threshold. If so, the first target region corresponding to the first target expansion region and the second target region corresponding to the second target expansion region are determined as belonging to the same needle imprint's first target region and second target region. This can be done as follows: Figure 5 As shown, Figure 5 The first target area corresponding to the first needle mark area A2 and the second target area corresponding to the second needle mark area B2 belong to the same needle mark.

[0102] This invention provides an example for reference, in which the first and second fingerprint detection results are determined as follows: Figure 5 As shown, and based on the above process, if the first target area corresponding to the first needle mark area A2 and the second target area corresponding to the second needle mark area B2 belong to the same needle mark, then according to... Figure 5 The first pin mark area A1 in the image determines the pin mark detection results for the target die, which can include "pin mark distance from the edge," indicating that the pin mark is close to the boundary of the pad area but does not exceed the boundary of the pad area. This can be based on... Figure 5 The first pinprint area A2 and the second pinprint area B2 in the image determine the pinprint detection results for the target die. These results can include "pinprint out of bounds," meaning that part of the pinprint is located within the pad and another part is outside the pad. This can be determined based on... Figure 5The second pinprint area B1 in the model can determine the pinprint detection result of the target die, which may include "pinprint is completely located outside", that is, the pinprint is completely located outside the pad.

[0103] According to the above-described scheme of the present invention, at least one first target region among at least one first needle mark regions can be determined based on the respective positions of at least one first needle mark region in the target image. Then, based on the respective positions of at least one second needle mark region in the target image, at least one second target region among at least one second needle mark region can be determined. Finally, based on the respective positions of at least one first target region and at least one second target region in the target image, the first target region and the second target region belonging to the same needle mark are determined as part of the die needle mark detection result for the target die. The above scheme uses the boundary region as a reference point to correlate and match the dispersed first and second needle mark regions, which can, to a certain extent, solve the misjudgment problem caused by the dispersion of regions in needle mark detection. This is beneficial to improving the accuracy and reliability of needle mark detection, reducing the occurrence of misjudging multiple unrelated needle mark regions as the same needle mark or dividing the same needle mark into multiple regions, thereby providing more accurate needle mark defect detection results for wafer production and effectively improving product quality and production efficiency.

[0104] This invention also provides a device for detecting grain needle marks. Figure 6 A schematic block diagram of a detection device 300 according to an embodiment of the present invention is shown. (In conjunction with...) Figure 6 As shown, the detection device 300 may include: a pad area determination module 310, a boundary area and pad area determination module 320, a pinprint detection result determination module 330, and a die pinprint detection result determination module 340.

[0105] The pad region determination module 310 is used to determine the pad region in the target image including the target die, wherein the pad region is the region in the target image corresponding to the interior of the pad in the target die.

[0106] The boundary region and pad outer region determination module 320 is used to determine the boundary region and pad outer region based on the position of the pad inner region in the target image. The boundary region is the region in the target image corresponding to the edge of the pad of the target die, and the pad outer region is the region in the target image corresponding to the outside of the pad in the target die.

[0107] The pin mark detection result determination module 330 is used to determine the first pin mark detection result in the area inside the pad and the second pin mark detection result in the area outside the pad. The first pin mark detection result is used to indicate the location of the first pin mark area corresponding to the pin mark in the area inside the pad or the pin mark in the area inside the pad. The second pin mark detection result is used to indicate the location of the second pin mark area corresponding to the pin mark in the area outside the pad or the pin mark in the area outside the pad.

[0108] The grain imprint detection result determination module 340 is used to determine the grain imprint detection result of the target grain based on the first imprint detection result and the second imprint detection result.

[0109] For example, the detection device 300 may further include: a target image acquisition module, a pixel correspondence determination module, and a first determination module.

[0110] The target image acquisition module can be used to acquire target images including target grains.

[0111] The pixel correspondence determination module can be used to determine the pixel correspondence between the target image and the reference image including the reference grains.

[0112] The first determining module can be used to determine the region within the pad in the target image corresponding to the reference region within the pad in the reference image based on the position and pixel correspondence of the reference region within the pad in the reference image, wherein the reference region within the pad is the region inside the pad in the reference die that corresponds to the reference image.

[0113] For example, the boundary area and pad-outside area determination module 320 may include: a second determination module and a third determination module.

[0114] The second determining module can be used to determine the boundary region based on the position of the region within the pad in the target image and a preset outward expansion threshold, wherein the geometric center of the boundary region is at the same position as the geometric center of the region within the pad.

[0115] The third determination module can be used to determine the area outside the pads based on the position of the boundary region in the target image.

[0116] For example, if the image brightness of the target image is higher than the brightness threshold, the needle mark detection result determination module 330 may include: a first candidate region determination module and a fourth determination module.

[0117] The first candidate region determination module can be used to determine at least one first candidate region based on the target image and a preset grayscale threshold, wherein the grayscale value of the pixels included in each first candidate region is less than the preset grayscale threshold.

[0118] The fourth determining module can be used to filter at least one first candidate region whose area is greater than a preset area threshold, as the first needle imprint region.

[0119] For example, the needle mark detection result determination module 330 may include: a target template image determination module and a fifth determination module.

[0120] The target template image determination module can be used to determine the target template image among multiple preset template images based on the image brightness of the target image. The multiple preset template images have different image brightness, and each preset template image includes the outside of the pads in the reference die, but none of them include pin marks. The target template image is the preset template image with the smallest absolute value of the difference between the image brightness of the multiple preset template images and the image brightness of the target image.

[0121] The fifth determination module can be used to determine the second pin mark detection result in the area outside the pad based on the target image and the target template image.

[0122] For example, the first pin mark detection result is used to indicate the position of the first pin mark area corresponding to the pin mark in the pad area, and the second pin mark detection result is used to indicate the position of the second pin mark area corresponding to the pin mark in the pad area. The die pin mark detection result determination module 340 may include: a first target area determination module, a second target area determination module, and a sixth determination module.

[0123] The first target region determination module can be used to determine at least one first target region among at least one first needle imprint regions based on the position of each of the at least one first needle imprint regions in the target image, wherein the first target region is a first needle imprint region that intersects with the boundary region.

[0124] The second target region determination module can be used to determine at least one second target region among at least one second needle imprint region based on the position of each of the at least one second needle imprint region in the target image, wherein the second target region is a second needle imprint region that intersects with the boundary region.

[0125] The sixth determining module can be used to determine the first target region and the second target region belonging to the same needle mark based on the position of at least one first target region in the target image and the position of at least one second target region in the target image, so as to be part of the grain needle mark detection result of the target grain.

[0126] According to another aspect of the present invention, an electronic device is also provided. Figure 7 A schematic block diagram of an electronic device 400 according to an embodiment of the present invention is shown. Figure 7As shown, the electronic device 400 includes a processor 410 and a memory 420. The memory 420 stores a computer program, and the computer program instructions are executed by the processor 410 to perform the above-described method for detecting die pin marks.

[0127] Furthermore, according to another aspect of the present invention, a storage medium is also provided, on which program instructions are stored. When the program instructions are executed by a computer or processor, the computer or processor performs corresponding steps of the die pinprint detection method described above in the embodiments of the present invention, and is used to implement corresponding modules in the die pinprint detection device or the electronic device described above in the embodiments of the present invention. The storage medium may, for example, include a memory card of a smartphone, a storage component of a tablet computer, a hard disk of a personal computer, a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a portable compact disc read-only memory (CD-ROM), a USB memory, or any combination of the above storage media.

[0128] According to another aspect of the present invention, a computer program product is also provided, comprising computer program instructions, which, when executed by a computer or processor, cause the computer or processor to perform the corresponding steps of the above-described method for detecting die pin marks.

[0129] Those skilled in the art can understand the specific implementation schemes of the above-mentioned electronic devices and storage media by reading the relevant descriptions of the detection methods for grain pin marks. For the sake of brevity, they will not be described in detail here.

[0130] Although exemplary embodiments have been described herein with reference to the accompanying drawings, it should be understood that the above exemplary embodiments are merely illustrative and are not intended to limit the scope of the invention thereto. Various changes and modifications can be made therein by those skilled in the art without departing from the scope and spirit of the invention. All such changes and modifications are intended to be included within the scope of the invention as claimed in the appended claims.

[0131] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0132] In the several embodiments provided by this invention, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed.

[0133] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0134] Similarly, it should be understood that, in order to streamline the invention and aid in understanding one or more of the various aspects of the invention, features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof in the description of exemplary embodiments of the invention. However, this approach should not be construed as reflecting an intention that the claimed invention requires more features than are expressly recited in each claim. Rather, as reflected in the corresponding claims, its inventive point lies in solving the corresponding technical problem with fewer features than all of those in a single disclosed embodiment. Therefore, the claims following the detailed description are hereby expressly incorporated into that detailed description, wherein each claim itself is a separate embodiment of the invention.

[0135] Those skilled in the art will understand that, apart from the mutual exclusion of features, all features disclosed in this specification (including the accompanying claims, abstract, and drawings) and all processes or units of any method or apparatus so disclosed can be combined in any combination. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstract, and drawings) may be replaced by an alternative feature that serves the same, equivalent, or similar purpose.

[0136] Furthermore, those skilled in the art will understand that although some embodiments described herein include certain features but not others included in other embodiments, combinations of features from different embodiments are intended to be within the scope of the invention and form different embodiments. For example, in the claims, any of the claimed embodiments can be used in any combination.

[0137] The various component embodiments of the present invention can be implemented in hardware, or as software modules running on one or more processors, or a combination thereof. Those skilled in the art will understand that microprocessors or digital signal processors (DSPs) can be used in practice to implement some or all of the functions of some modules in the die pinprint detection apparatus according to embodiments of the present invention. The present invention can also be implemented as an apparatus program (e.g., a computer program and computer program product) for performing part or all of the methods described herein. Such programs implementing the present invention can be stored on a computer-readable medium or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, provided on a carrier signal, or provided in any other form.

[0138] It should be noted that the above embodiments are illustrative of the invention and not restrictive, and that those skilled in the art can devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names.

[0139] The above description is merely a specific embodiment of the present invention or an explanation of that embodiment. The scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. The scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for detecting grain needle marks, characterized in that, The method includes: Determine the pad region in the target image including the target die, wherein the pad region is the region in the target image corresponding to the interior of the pad in the target die; Based on the position of the pad region in the target image, the boundary region and the pad region are determined, wherein the boundary region is the region in the target image corresponding to the edge of the pad of the target die, and the pad region is the region in the target image corresponding to the outside of the pad in the target die; Determine the first pin mark detection result of the area inside the pad and the second pin mark detection result of the area outside the pad, wherein the first pin mark detection result is used to indicate the location of the first pin mark area corresponding to the pin mark in the area inside the pad, or the location of the pin mark in the area inside the pad; and the second pin mark detection result is used to indicate the location of the second pin mark area corresponding to the pin mark in the area outside the pad. Based on the first needle mark detection result and the second needle mark detection result, the grain needle mark detection result of the target grain is determined.

2. The method as described in claim 1, characterized in that, The method further includes: Obtain a target image including the target grain; Determine the pixel correspondence between the target image and a reference image including reference grains; Based on the position of the reference region within the pad in the reference image and the pixel correspondence, the pad region within the target image corresponding to the reference region within the pad is determined, wherein the reference region within the pad is the region inside the pad in the reference die that corresponds to the reference image.

3. The method as described in claim 1, characterized in that, The step of determining the boundary region and the region outside the pad based on the position of the region within the pad in the target image includes: Based on the position of the region within the pad in the target image and a preset outward expansion threshold, the boundary region is determined, wherein the geometric center of the boundary region is at the same position as the geometric center of the region within the pad; The area outside the pad is determined based on the position of the boundary region in the target image.

4. The method as described in claim 1, characterized in that, The image brightness of the target image is higher than a brightness threshold, and the determination of the first pin mark detection result in the region within the pad includes: Based on the target image and a preset grayscale threshold, at least one first candidate region is determined, wherein the grayscale value of the pixels included in each of the at least one first candidate region is less than the preset grayscale threshold. Select the first candidate regions whose area is greater than a preset area threshold from the at least one first candidate regions, and use them as the first needle imprint regions.

5. The method as described in claim 1, characterized in that, The determination of the second pin mark detection result in the area outside the pad includes: Based on the image brightness of the target image, a target template image is determined among multiple preset template images, wherein the image brightness of each of the multiple preset template images is different, each of the multiple preset template images includes the outside of the pads in the reference die, and none of them include pin marks, and the target template image is the preset template image with the smallest absolute value of the difference between the image brightness of the multiple preset template images and the image brightness of the target image; Based on the target image and the target template image, the second pin mark detection result in the area outside the pad is determined.

6. The method as described in claim 1, characterized in that, The first pin imprint detection result is used to indicate the position of the first pin imprint area corresponding to the pin imprint in the area within the pad, and the second pin imprint detection result is used to indicate the position of the second pin imprint area corresponding to the pin imprint in the area outside the pad. Determining the grain pin imprint detection result of the target grain based on the first pin imprint detection result and the second pin imprint detection result includes: Based on the position of each of the at least one first pinprint region in the target image, at least one first target region among the at least one first pinprint region is determined, wherein the first target region is a first pinprint region that intersects with the edge of the region within the pad; Based on the position of each of the at least one second pinprint region in the target image, at least one second target region among the at least one second pinprint region is determined, wherein the second target region is a second pinprint region that intersects with the inner edge of the outer region of the pad; Based on the positions of the at least one first target region and the at least one second target region in the target image, the first target region and the second target region belonging to the same needle mark are determined as part of the grain needle mark detection result of the target grain.

7. A device for detecting grain needle marks, characterized in that, The device includes: The pad region determination module is used to determine the pad region in a target image including a target die, wherein the pad region is the region inside the pad in the target die that corresponds to the inside of the pad in the target image; The boundary region and pad outer region determination module is used to determine the boundary region and pad outer region based on the position of the pad inner region in the target image. The boundary region is the region in the target image corresponding to the edge of the pad of the target die, and the pad outer region is the region in the target image corresponding to the outer part of the pad in the target die. The pin mark detection result determination module is used to determine the first pin mark detection result of the area inside the pad and the second pin mark detection result of the area outside the pad. The first pin mark detection result is used to indicate the location of the first pin mark area corresponding to the pin mark in the area inside the pad, and the second pin mark detection result is used to indicate the location of the second pin mark area corresponding to the pin mark in the area outside the pad. The grain needle mark detection result determination module is used to determine the grain needle mark detection result of the target grain based on the first needle mark detection result and the second needle mark detection result.

8. An electronic device, characterized in that, The device includes a memory and a processor, wherein the memory is used to store a computer program; and the processor is used to execute the computer program to implement the method for detecting grain needle marks as described in any one of claims 1-7.

9. A storage medium storing computer program instructions, characterized in that, The computer program instructions, when executed, are used to perform the grain imprint detection method as described in any one of claims 1-7.

10. A computer program product comprising computer program instructions, characterized in that, The computer program instructions, when executed by a processor, are used to perform the method for detecting grain pin marks as described in any one of claims 1-7.

Images