Identification detection method and device for semiconductor packaging workpiece and cutting and sorting integrated machine
By extracting the image skeleton and projecting the identification reference line from the surface image of the integrated circuit packaged chip, the problem of low identification detection accuracy is solved, and high-precision and standardized detection of semiconductor packaged workpieces is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHENYANG HEYAN TECH CO LTD
- Filing Date
- 2026-04-28
- Publication Date
- 2026-07-14
Smart Images
Figure CN122115439B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of chip inspection technology, and in particular to a method, apparatus and integrated cutting and sorting machine for identifying semiconductor packaged workpieces. Background Technology
[0002] Before sorting integrated circuit packaged chips, the surface of the integrated circuit packaged chips is usually marked with good material markings (O-shaped markings) or bad material markings (X-shaped markings) based on the inspection results. When marking the surface of integrated circuit packaged chips, the markings are usually formed by handwriting or laser printing. Handwritten markings are prone to uneven thickness and random intersection angles, while laser-printed markings are prone to broken lines.
[0003] Although existing technologies employ connected component analysis or Hough transform to detect markings on the surface of integrated circuit packaged chips, connected component analysis is prone to detection failure when the markings are broken, while Hough transform has limited effectiveness in detecting irregular handwritten markings. Therefore, the above methods tend to result in low detection accuracy of markings on the surface of integrated circuit packaged chips, and the detection results are difficult to meet the industrial precision sorting requirements of integrated circuit packaged chips. Summary of the Invention
[0004] This application provides a method, apparatus, and integrated cutting and sorting machine for detecting markings on semiconductor packaged workpieces, which can solve the problem of low accuracy in detecting markings on workpieces to be inspected in the prior art.
[0005] In a first aspect, a method for identifying semiconductor packaged workpieces is provided, comprising:
[0006] Acquire a surface image of the workpiece to be inspected; the surface image includes an identifier used to mark the workpiece type; the workpiece type is either good material or defective material.
[0007] Image skeleton extraction is performed on the surface image to obtain the skeleton lines corresponding to the patterns in the surface image;
[0008] For each pixel on the skeleton line, if the minimum distance between the pixel and the preset identification reference line is less than the first preset distance, then the pixel is projected onto the identification reference line; the identification reference line is drawn in the surface image according to the identification shape corresponding to the defective material type.
[0009] The type of workpiece corresponding to the marking on the workpiece to be inspected is determined by the proportion of the area projected by the skeleton line in the marking reference line.
[0010] Optionally, it also includes:
[0011] Based on the outline information of the mark shape corresponding to the defective material type, the outline feature points of the mark shape corresponding to the defective material type are obtained.
[0012] Based on the size information of the surface image, determine the position information of the contour feature points in the surface image;
[0013] For each contour feature point, the contour feature point is calibrated in the surface image according to its position information in the surface image;
[0014] Based on the contour feature points marked in the surface image, lines are drawn in the surface image to obtain a matching line with the same contour as the defective material type identifier.
[0015] Optionally, the identification shape corresponding to the defective material type includes a shape with two intersecting straight lines, the two straight lines being a first straight line and a second straight line; correspondingly, based on the contour information of the identification shape corresponding to the defective material type, the contour feature points of the identification shape corresponding to the defective material type are obtained, including:
[0016] Based on the contour information corresponding to the first straight line and the second straight line, determine the intersection point between the first straight line and the second straight line, as well as the direction of the first straight line and the second straight line respectively.
[0017] Based on the intersection point between the first straight line and the second straight line, and the directions of the first straight line and the second straight line, determine at least two first target points in the first straight line and at least two second target points in the second straight line;
[0018] The intersection point, the first target point, and the second target point are respectively used as contour feature points.
[0019] Optionally, based on the size information of the surface image, the position information of the contour feature points in the surface image is determined, including:
[0020] Based on the size information of the surface image, and a first ratio of the distance between the intersection point and the two endpoints of the first straight line, and / or a second ratio of the distance between the intersection point and the two endpoints of the second straight line, the position information of the intersection point in the surface image is determined.
[0021] Based on the positional correspondence between the first target point and the second target point and the intersection point, the positional information of the first target point and the second target point in the surface image is determined.
[0022] Optionally, the workpiece type corresponding to the marking on the workpiece to be inspected is determined based on the proportion of the area projected by the skeleton line in the marking reference line, including:
[0023] The pixel length of the region projected by the skeleton line in the first straight line is calculated to obtain the first pixel length value;
[0024] The first percentage of the area projected by the skeleton line in the first straight line is obtained by calculating the length value of the first pixel and the length value of the first straight line.
[0025] The pixel length of the region projected onto the skeleton line in the second straight line is calculated to obtain the second pixel length value.
[0026] The second pixel length value and the second line length value are calculated to obtain the second proportion of the area in the second line that is projected by the skeleton line;
[0027] Compare the first proportion with the first preset proportion, and compare the second proportion with the second preset proportion;
[0028] If the first proportion is greater than the first preset proportion and the second proportion is greater than the second preset proportion, then the workpiece type corresponding to the mark on the workpiece to be inspected is determined to be a defective material type.
[0029] Optionally, for each pixel on the skeleton line, if the minimum distance between the pixel and the preset identification reference line is less than a first preset distance, then before projecting the pixel onto the identification reference line, the method further includes:
[0030] For each contour feature point, determine whether there is a segment of the skeleton line in the region centered on the contour feature point in the surface image.
[0031] If there is no part of the skeleton line in any region centered on the contour feature point, then the workpiece type corresponding to the mark on the workpiece to be inspected is determined to be a good material type.
[0032] If all regions centered on the contour feature point contain a portion of the skeleton line, then for each region centered on the contour feature point, calculate the deviation between the portion of the skeleton line located within that region and the portion of the marker line located within that region.
[0033] If the deviation value is greater than the preset deviation value, the position of the marker line in the surface image is adjusted until the deviation value between the corresponding part of the skeleton line and the marker line in each region centered on the contour feature point is less than the preset deviation value.
[0034] Optionally, for each pixel on the skeleton line, if the minimum distance between the pixel and the preset identification reference line is less than a first preset distance, before projecting the pixel onto the identification reference line, the method further includes:
[0035] Pixels whose minimum distance from the skeleton line to the identification line is greater than the second preset distance are removed from the skeleton line to obtain the set of pixels to be compared.
[0036] The minimum distance between each pixel in the set of pixels to be compared and the identification line is statistically analyzed.
[0037] Based on the value corresponding to the minimum distance between each pixel and the identification line in the statistical set of pixels to be compared, calculate the average value corresponding to the minimum distance between the pixel and the identification line.
[0038] The variance value is obtained by calculating the variance between the minimum distance and the average distance between each pixel in the set of pixels to be compared and the marker line.
[0039] The mean and variance values are calculated, and the results are used as the first preset distance.
[0040] Optionally, image skeleton extraction is performed on the surface image to obtain the skeleton lines corresponding to the patterns in the surface image, including:
[0041] The surface image is binarized to obtain a binary image;
[0042] Extract the contour map corresponding to the pattern from the binarized image;
[0043] The skeleton of the outline corresponding to the pattern is extracted to obtain the skeleton line of the pattern; the skeleton line is a single-layer pixel width outline line.
[0044] Secondly, a marking and detection device for semiconductor packaged workpieces is provided, comprising:
[0045] The image acquisition module is used to acquire a surface image of the workpiece to be inspected; the surface image includes an identifier for marking the workpiece type; the workpiece type is either good material or defective material.
[0046] The image skeleton extraction module is used to extract the image skeleton from the surface image to obtain the skeleton lines corresponding to the patterns in the surface image.
[0047] The projection module is used to project each pixel on the skeleton line onto the labeling reference line if the distance between the pixel and the preset labeling reference line is less than a first preset distance. The labeling reference line is drawn in the surface image according to the labeling shape corresponding to the defective material type.
[0048] The marking determination module is used to determine the workpiece type corresponding to the marking on the workpiece to be inspected based on the proportion of the area projected by the skeleton line in the marking reference line.
[0049] Thirdly, an electronic device is provided, comprising: a processor and a memory for storing a computer program, the processor for calling and running the computer program stored in the memory, and performing the methods as described in the first aspect or its various implementations.
[0050] Fourthly, a computer-readable storage medium is provided for storing a computer program that causes a computer to perform the methods described in the first aspect or its various implementations.
[0051] Fifthly, a cutting and sorting integrated machine is provided, including a marking and detection device for semiconductor packaged workpieces as in the second aspect, or an electronic device as in the third aspect.
[0052] The technical solution provided in this application projects the skeleton line corresponding to the pattern in the surface image of the workpiece to be inspected onto the mark reference line drawn in the surface image according to the mark shape corresponding to the defect type. Then, based on the proportion of the area projected by the skeleton line in the mark reference line, the workpiece type corresponding to the mark on the workpiece to be inspected is determined. This solves the problems of low mark detection accuracy and weak anti-interference ability in the prior art, and improves the accuracy of mark detection for semiconductor packaged workpieces.
[0053] Other features and advantages of this disclosure will be described in detail in the following detailed description section. Attached Figure Description
[0054] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0055] Figure 1 An application scenario diagram provided for an embodiment of this application;
[0056] Figure 2 A flowchart illustrating a method for identifying semiconductor packaged workpieces, provided as an embodiment of this application;
[0057] Figure 3 A schematic diagram illustrating an example of a surface image provided in an embodiment of this application;
[0058] Figure 4 This is a schematic diagram illustrating an example of a binarized image corresponding to a surface image provided in an embodiment of this application.
[0059] Figure 5 A schematic diagram illustrating an example of skeleton lines corresponding to a pattern in a surface image provided in an embodiment of this application;
[0060] Figure 6 A schematic diagram illustrating an example of a surface image with marking lines provided in an embodiment of this application;
[0061] Figure 7 A schematic diagram illustrating an example of the projection of the skeleton line onto the identification reference line, provided in an embodiment of this application;
[0062] Figure 8 A schematic diagram illustrating an example of the area projected by the skeleton line in the identification reference line provided in an embodiment of this application;
[0063] Figure 9 A schematic diagram of a module for a marking and detection device for a semiconductor packaged workpiece provided in an embodiment of this application;
[0064] Figure 10 This is a schematic block diagram of the electronic device provided in the embodiments of this application. Detailed Implementation
[0065] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.
[0066] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or server that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.
[0067] As mentioned above, in the production process of integrated circuit packaged chips, to avoid confusion between different batches of products and errors in product orientation identification, text is usually printed on the chip surface. The text includes various forms such as characters and non-characters. Therefore, as a workpiece to be inspected, the surface image of an integrated circuit packaged chip not only contains markings for indicating the workpiece type (good material type, defective material type), but also includes the aforementioned text.
[0068] Before sorting the workpieces to be inspected, it is necessary to distinguish between good materials (O mark) and bad materials (X mark) through marking so that the sorting equipment can complete the sorting accurately. Among them, the bad material marking is usually formed by handwriting or laser printing to form an "X" stroke. However, handwriting is prone to problems such as uneven thickness and random intersection angles, while laser printing is prone to line breakage, which brings difficulties to the marking inspection.
[0069] Existing detection methods primarily identify defective parts by detecting straight lines on the workpiece surface using connected component analysis or Hough transform. However, connected component analysis fails to detect broken X-marks, and Hough transform has limited effectiveness in detecting irregularly shaped handwritten X-marks. Furthermore, neither of these methods can quantify the detection results into normalized scores, resulting in a lack of clear and unified criteria for judging the accuracy of defective part identification, which makes it difficult to meet the needs of precise detection and standardized sorting in industrial production.
[0070] To address at least one of the technical problems existing in the prior art or related technologies, this invention provides a method, apparatus, and integrated cutting and sorting machine for identifying semiconductor packaged workpieces. The method includes: acquiring a surface image of the workpiece to be inspected; the surface image includes an identifier for marking the workpiece type; the workpiece type is either a good material type or a defective material type; extracting an image skeleton from the surface image to obtain skeleton lines corresponding to patterns in the surface image; for each pixel on the skeleton line, if the minimum distance between the pixel and a preset identifier reference line is less than a first preset distance, then projecting the pixel onto the identifier reference line; the identifier reference line is drawn in the surface image according to the identifier shape corresponding to the defective material type; determining the workpiece type corresponding to the identifier on the workpiece to be inspected based on the proportion of the area projected by the skeleton line in the identifier reference line. This application can solve the problems of low accuracy in identifying identifiers on workpieces to be inspected, weak anti-interference ability, and inability to quantify detection results in the prior art, thereby improving the accuracy and standardization level of identifying semiconductor packaged workpieces.
[0071] It should be understood that the technical solution of this application can be applied to the following scenarios, but is not limited to:
[0072] In some possible ways, Figure 1 An application scenario diagram provided for an embodiment of this application, such as... Figure 1 As shown, this application scenario may include electronic device 110 and network device 120. Electronic device 110 can establish a connection with network device 120 through a wired network or a wireless network.
[0073] For example, electronic device 110 may be a desktop computer, laptop computer, tablet computer, etc., but is not limited thereto. Network device 120 may be a terminal device or a server, but is not limited thereto. In one embodiment of this application, electronic device 110 may send a request message to network device 120, which may be used to request motor operation data corresponding to a motor in operation. Further, electronic device 110 may receive a response message sent by network device 120, which includes obtaining motor operation data corresponding to a motor in operation.
[0074] also, Figure 1 An electronic device 110 and a network device 120 are provided as examples, but other numbers of electronic devices and network devices may be included in practice, and this application does not limit this.
[0075] In other possible implementations, the technical solution of this application may also be executed by the aforementioned electronic device 110, or by the aforementioned network device 120, and this application does not impose any restrictions on this.
[0076] After introducing the application scenarios of the embodiments of this application, the technical solution of this application will be described in detail below:
[0077] Figure 2 The flowchart illustrates a method for identifying semiconductor packaged workpieces according to an embodiment of this application, and can be implemented by, for example... Figure 1 The electronic device 110 shown performs, but is not limited to, this method. Applied to a sorting device, the method may include the following steps:
[0078] S210. Obtain a surface image of the workpiece to be inspected.
[0079] The surface image includes an identifier for marking the type of workpiece to be inspected; the workpiece type is either good material or bad material.
[0080] Before sorting the workpieces to be inspected, their surfaces are marked according to the inspection results to distinguish whether they are good or bad, so that the sorting equipment can sort them accordingly. For example, after inspecting workpiece A and determining that it is a good workpiece, a mark indicating that it is a good workpiece is made on its surface. Similarly, after inspecting workpiece B and determining that it is a bad workpiece, a mark indicating that it is a bad workpiece is made on its surface.
[0081] In this step, by acquiring the surface image of the workpiece to be inspected, not only can the type identification on the workpiece be obtained, but also the characters or graphics on the workpiece other than the type identification can be retained, providing comprehensive raw data for the identification determination of the tool to be inspected, and realizing the recording of the identification on the workpiece to be inspected in the surface image in the original pixel form.
[0082] S220. Extract the image skeleton from the surface image to obtain the skeleton lines corresponding to the patterns in the surface image.
[0083] Among them, the skeleton line is the outline line with a single-layer pixel width.
[0084] Here, by extracting the image skeleton from the surface image, it is possible to... Figure 4 Irrelevant redundant information (such as lettering texture, stroke thickness, surface stains, device noise, etc.) in the surface image is stripped away to transform the patterns (including logos, lettering, etc.) in the surface image into single-pixel width images. Figure 5 The skeleton lines shown not only highlight the core topological structure of the pattern in the surface image, but also eliminate the interference of redundant information on subsequent detection. At the same time, the extracted skeleton lines can reduce the computational dimension of data processing in the subsequent step S230, providing a simple and effective feature input for the projection matching step in step S230, thereby accelerating the detection efficiency of the marking of semiconductor packaged workpieces.
[0085] S230. For each pixel on the skeleton line, if the minimum distance between the pixel and the preset identification reference line is less than the first preset distance, then project the pixel onto the identification reference line.
[0086] Among them, the identification reference line is drawn on the surface image according to the identification shape corresponding to the defective material type. That is, it is a continuous and regular standard reference line drawn on the surface image according to the identification shape corresponding to the defective material type. This identification reference line can be used as a reference template for matching defective material identification.
[0087] by Figure 5 Taking the skeleton line shown as an example, the identification and reference lines drawn in the surface image can be: Figure 6 The two intersecting straight lines in the skeleton (where the endpoints of both lines are located at the apex corners of the semiconductor package workpiece), and the pixels whose minimum distance between the skeleton line and the preset identification reference line is less than a first preset distance, such as... Figure 7 The remaining white pixels are shown in the image.
[0088] Here, the minimum distance between a pixel on the skeleton line and a preset marker line can be the vertical distance from that pixel to the marker line.
[0089] It should be noted that the workpiece to be inspected is an integrated circuit package. During the packaging process, the chip is typically printed with markings to prevent confusion between different batches of integrated circuit packages and to avoid errors in product orientation identification. These markings can include characters (such as letters, numbers, words, and punctuation marks) and non-character markings (such as graphic symbols). Therefore, the surface image of the workpiece to be inspected generally includes not only markings indicating the workpiece type but also the printed markings.
[0090] Since the surface image of the workpiece to be inspected also includes printed characters, in order to reduce the impact of the projection of the skeleton line corresponding to the printed characters onto the identification reference line on the proportion in step S240, in this step, the minimum distance between each pixel in the skeleton line and the preset identification reference line is compared with the first preset distance. Pixels in the skeleton line whose minimum distance to the preset identification reference line is greater than the first preset distance are regarded as pixels on the skeleton line corresponding to the printed characters, and pixels whose minimum distance to the preset identification reference line is less than the first preset distance are regarded as pixels on the skeleton line corresponding to the identification. Then, the pixels whose minimum distance to the preset identification reference line is less than the first preset distance are projected onto the identification reference line to obtain the projection result of the skeleton line corresponding to the identification on the identification reference line.
[0091] In this step, by comparing the minimum distance between each pixel in the skeleton line and the identification reference line, which serves as a reference template for matching defective material identification, with a first preset distance, effective pixels in the skeleton line related to the identification reference line are selected. The selected effective pixels are then projected onto the identification reference line. Even if handwritten markings have variations in thickness or laser-printed markings have partial breaks, the effective pixels in the skeleton line related to the identification reference line can still be aligned with the identification reference line. Thus, as long as most of the skeleton line overlaps with the outline of the identification reference line, the marking corresponding to the defective material type can be correctly identified. This solves the problem of inaccurate detection caused by the sensitivity of traditional methods to broken strokes and irregular strokes in the marking.
[0092] S240. Determine the workpiece type corresponding to the marking on the workpiece to be inspected based on the proportion of the area projected by the skeleton line in the marking reference line.
[0093] Here, the proportion of the area projected by the skeleton line in the identification comparison line can be compared with a preset threshold to determine the workpiece type corresponding to the identification on the workpiece to be inspected, thereby realizing the analysis of the normalized score of the identification on the workpiece to be inspected.
[0094] Specifically, when the proportion of the area projected by the skeleton line in the identification reference line reaches or exceeds a preset threshold, it indicates that the skeleton line matches the identification reference line corresponding to the defective material type well, and the workpiece type corresponding to the identification on the workpiece to be inspected can be determined to be a defective material type; when the proportion of the area projected by the skeleton line in the identification reference line is lower than the preset threshold, it indicates that the skeleton line matches the identification reference line corresponding to the defective material type poorly, and the workpiece type corresponding to the identification on the workpiece to be inspected can be determined to be a good material type.
[0095] In this step, by determining the proportion of the area projected by the skeleton line in the identification reference line, the workpiece type corresponding to the identification on the workpiece to be inspected can be determined. This enables the analysis of the normalized score of the identification on the workpiece to be inspected, providing an objective and accurate quantitative indicator for identification inspection.
[0096] The above method projects the skeleton lines corresponding to the patterns in the surface image of the workpiece to be inspected onto a marker reference line drawn in the surface image according to the marker shape corresponding to the defect type. Then, based on the proportion of the area projected by the skeleton lines in the marker reference line, the workpiece type corresponding to the marker on the workpiece to be inspected is determined. This solves the problem of relying on "connected component analysis" or "Hough transform" for markers on workpieces in existing technologies. It also effectively addresses the problems of handwritten marker deformation and the fragility of laser-printed markers, significantly improving detection robustness and applicability to the detection of markers for defect types under various processes. Furthermore, by determining the workpiece type corresponding to the marker on the workpiece to be inspected based on the proportion of the area projected by the skeleton lines in the marker reference line, the detection result can be quantified into a normalized score, providing an accurate numerical basis for the detection of defect type markers. Therefore, this application solves the problems of low detection accuracy, weak anti-interference ability, and inability to quantify detection results in existing technologies, improving the accuracy and standardization of marker detection for semiconductor packaging workpieces.
[0097] In some possible implementations, the method further includes:
[0098] S310. Based on the contour information of the marking shape corresponding to the defective material type, obtain the contour feature points of the marking shape corresponding to the defective material type.
[0099] Here, the outline information of the identifier shape corresponding to the defective material type can include shape information; the outline feature points can include edge points, intersection points, and points that can be used to draw the identifier corresponding to the defective material type.
[0100] For example, when the outline information of the identifier shape corresponding to the defective material type is X-shaped, the outline feature points of the identifier shape corresponding to the defective material type can be the intersection point and the four endpoints of the X-shaped identifier.
[0101] In this step, by obtaining the outline feature points of the logo shape corresponding to the defective material type based on the outline information of the logo shape corresponding to the defective material type, key outline feature points (such as the intersection and endpoints of the X-shape) can be extracted from the complete outline of the logo (such as the X-shape) corresponding to the defective material type. Redundant pixels in the outline of the logo are removed, thus preserving the core points that determine the shape of the logo and ensuring that the subsequent generation of the logo reference line is not interfered with by irrelevant information.
[0102] S320. Based on the size information of the surface image, determine the position information of the contour feature points in the surface image.
[0103] Following the example in step S310, if the contour feature points of the identifier shape corresponding to the defective material type can be the intersection point and the four endpoints of the X-shaped identifier, and the four endpoints of the X-shaped identifier are respectively located at the positions corresponding to the four corners of the workpiece to be inspected, then the position information of the contour feature points in the surface image can be determined as: the positions corresponding to the four corners of the area corresponding to the workpiece to be inspected in the surface image.
[0104] Since different surface image sizes correspond to different sized markers, the position information of contour feature points in the surface image is determined based on the surface image size information. This ensures that the marker reference lines drawn in subsequent steps can match the size of the workpiece to be inspected, thereby solving the problems of incompatibility between fixed-size marker reference lines and surface images corresponding to workpieces of different specifications, as well as large comparison deviations.
[0105] Meanwhile, by combining the size information of the surface image, the position information of the contour feature points in the surface image can be determined. This allows the position of the contour feature points to be calibrated by the size information of the surface image when the packaging specifications of the workpiece to be inspected or the image acquisition ratio changes. This ensures that the marking and comparison lines drawn in subsequent steps always fit the actual scene of the current surface image, so that the obtained marking and comparison lines can be adapted to the inspection requirements of integrated circuit packages of different batches and specifications.
[0106] S330. For each contour feature point, the contour feature point is calibrated in the surface image according to its position information in the surface image.
[0107] Following the example in step S320, the four endpoints of the X-shaped marker are calibrated at the positions corresponding to the four corners of the area of the workpiece to be detected in the surface image.
[0108] In this step, by calibrating the contour feature points in the surface image according to the calculated position information, clear and accurate anchor points can be provided for the subsequent connection to generate the identification reference line. This avoids the deformation and offset of the identification reference line caused by the blurring of the position of the contour feature points in the surface image, thereby ensuring the contour consistency of the identification reference line with the standard defective material type.
[0109] S340. Based on the contour feature points marked in the surface image, draw lines in the surface image to obtain a marker line that matches the contour of the defective material type.
[0110] Following the example in step S330, the contour feature points marked at the four corners of the area corresponding to the workpiece to be detected in the surface image are connected to obtain the same identification line as the contour corresponding to the defective material type, which is the X-shaped identification line.
[0111] In this step, by connecting the contour feature points in the surface image, the standard contour of the defective material type identifier (such as an X-shape) can be accurately reproduced in the surface image, forming an identifier matching line that matches the surface image.
[0112] Using the above method, the contour feature points obtained from the contour information of the mark shape corresponding to the defective material type are combined with the size information of the surface image to determine the position information of the contour feature points in the surface image. Then, the contour feature points are calibrated in the surface image according to their position information in the surface image. This allows the mark comparison obtained from the contour feature points calibrated in the surface image to accurately match the surface image of the workpiece to be inspected, effectively avoiding detection misjudgment caused by the deviation of the comparison benchmark. Furthermore, the obtained mark comparison line can adapt to complex scenarios such as handwritten marks corresponding to the defective material type (uneven thickness, random intersection angle) and laser-printed marks corresponding to the defective material type (broken).
[0113] In some possible implementations, the identification shape corresponding to the defective material type includes a shape with two intersecting straight lines, the two straight lines being a first straight line and a second straight line.
[0114] Here, the shape formed by the first and second straight lines can be an X shape.
[0115] Accordingly, based on the contour information of the identifier shape corresponding to the defective material type, the contour feature points of the identifier shape corresponding to the defective material type are obtained, which may include the following steps:
[0116] S410. Based on the contour information corresponding to the first straight line and the second straight line, determine the intersection point between the first straight line and the second straight line, as well as the direction of the first straight line and the second straight line.
[0117] Here, the intersection point determined from the contour information corresponding to the first and second intersecting straight lines is the core feature of the intersecting linear sign (such as an X-shape). This intersection point also determines the overall shape of the sign, providing a core anchor point for the subsequent selection of contour feature points and avoiding the problem of contour feature point extraction deviating from the core of the sign. At the same time, determining the directions of the first and second straight lines clearly defines their spatial orientation in the surface image, effectively avoiding the shape interference caused by the random intersection angle of handwritten signs (such as X-shapes) and the line offset of laser-printed signs. This ensures that the selection of target points in subsequent steps always conforms to the inherent geometric direction of the sign corresponding to the defective material type.
[0118] In this step, without processing all pixels on the contours corresponding to the first and second lines respectively, only the intersection point between the first and second lines and the directions of the first and second lines need to be determined. These two key pieces of information can then be used to extract contour feature points in subsequent steps. This simplifies the feature extraction process from the contour information and avoids interference from redundant pixels in the selection of contour feature points, thereby improving the efficiency and accuracy of contour feature point extraction.
[0119] S420. Based on the intersection point between the first straight line and the second straight line, and the directions of the first straight line and the second straight line, determine at least two first target points in the first straight line and at least two second target points in the second straight line.
[0120] Here, by selecting the first and second target points based on the intersection point between the first and second straight lines, and combining the directions of the first and second straight lines respectively, it is possible to ensure that the first and second target points are always located on the corresponding straight lines, while also ensuring that the distribution of the first and second target points conforms to the contour pattern of the marker. Regardless of how the intersection angle changes or whether the lines are slightly offset, the selection of the first and second target points has a unified standard, ensuring the consistency of contour feature point extraction in different scenarios.
[0121] S430. The intersection point, the first target point, and the second target point are respectively used as contour feature points.
[0122] Here, taking the intersection point between the first and second straight lines as the core, and combining the directions of the first and second straight lines respectively, at least two first target points can be determined in the first straight line and at least two second target points can be determined in the second straight line. This ensures that when the intersection point, the first target point, and the second target point are used as contour feature points, the contour feature points can completely cover the contour range of the mark corresponding to the defective material type, avoiding the situation where the mark reference line generated by the subsequent connection is missing or deformed due to insufficient contour feature points.
[0123] Using the above method, it can be ensured that the obtained contour feature points can fully reflect the core shape (such as X-shape) of "two intersecting straight lines" in the defective material type identification. At the same time, the directional constraints of the two straight lines respectively ensure the standardization of contour feature point selection, avoid the error of human selection of feature points, and enable the same and accurate contour feature point extraction results for intersecting straight line identifications of different batches and different shapes.
[0124] In some possible implementations, determining the positional information of contour feature points in the surface image based on the surface image's size information may include the following steps:
[0125] S510. Determine the position information of the intersection point in the surface image based on the size information of the surface image, a first ratio of the distance between the intersection point and the two endpoints of the first straight line, and / or a second ratio of the distance between the intersection point and the two endpoints of the second straight line.
[0126] Here, in order to accurately anchor the core contour feature points and avoid positional deviations of the intersection points, the actual size information of the surface image is combined with the proportional relationship (first ratio, second ratio) of the intersection points on the first and second straight lines. This ensures that the position of the intersection points fits the image size and that the relative position of the intersection points with the endpoints of the first and second straight lines conforms to the inherent shape of the defective material type identifier, thus avoiding positioning offset of the intersection points.
[0127] Meanwhile, by using the size information of the surface image, the first ratio of the distance between the intersection point and the two endpoints of the first straight line, and / or the second ratio of the distance between the intersection point and the two endpoints of the second straight line, the position information of the intersection point in the surface image can be accurately located, so that the determined intersection point can effectively adapt to the morphological variations of the intersecting straight line mark.
[0128] S520. Based on the positional correspondence between the first target point and the second target point and the intersection point, determine the positional information of the first target point and the second target point in the surface image.
[0129] Here, the intersection point located in step S510 is used as the reference, and the positional correspondence between the first target point, the second target point and the intersection point (such as distance and direction) is combined to make the positions of all the first target points and the second target points accurately associated with the intersection point, so as to avoid the offset and misalignment caused by the independent positioning of the first target point and the second target point.
[0130] Using the above method, with the intersection point as the core reference, the associated positioning of all contour feature points is achieved, ensuring that the relative positions between contour feature points conform to the inherent shape of the identifier corresponding to the defective material type, thus avoiding the problem of random positioning of individual contour feature points.
[0131] In some possible implementations, determining the workpiece type corresponding to the marking on the workpiece to be inspected based on the proportion of the area projected by the skeleton line in the marking reference line may include the following steps:
[0132] S610. Calculate the pixel length of the region in the first straight line that is projected by the skeleton line to obtain the first pixel length value.
[0133] In this step, by calculating the pixel length of the area projected by the skeleton line in the first straight line, the first pixel length value corresponding to the effective projection area of the skeleton line on the first straight line can be accurately calculated. At the same time, this step converts the degree of overlap between the skeleton line and the first straight line into a specific pixel length value, avoiding the subjective error of manually estimating the projection range between the skeleton line and the first straight line, and providing objective and traceable data support for the subsequent calculation of the first proportion.
[0134] S620: Calculate the length value of the first pixel and the length value of the first straight line to obtain the first proportion of the area projected by the skeleton line in the first straight line.
[0135] It should be noted that the first percentage can intuitively reflect the projection coverage ratio of the skeleton line on the first straight line. Therefore, the higher the first percentage, the higher the fit between the skeleton line and the first straight line, and the closer it is to the standard form of the defective material type mark.
[0136] Here, by comparing the length value of the first pixel with the length of the first straight line itself, the length value of the first pixel can be converted into a proportion, eliminating the scale differences caused by different specifications of workpieces to be inspected and different sizes of marking lines. This provides a unified quantitative evaluation standard for the inspection of workpieces to be inspected in different batches and of different specifications, thus solving the problem of inconsistent marking inspection standards on workpieces to be inspected in different scenarios.
[0137] S630. Calculate the pixel length of the region in the second straight line that is projected by the skeleton line to obtain the second pixel length value.
[0138] Since the identifier corresponding to the defective material type with intersecting straight lines consists of two straight lines, quantizing only the projection of the first straight line cannot fully reflect the shape of the identifier. Therefore, in this step, the projection pixel length of the skeleton line on the second straight line is supplemented to achieve a comprehensive capture of the identifier projection situation corresponding to the entire defective material type.
[0139] S640: Calculate the length value of the second pixel and the length value of the second line to obtain the second proportion of the area projected by the skeleton line in the second line.
[0140] Here, by calculating the second proportion, it can complement the first proportion and avoid misjudgment caused by the incomplete evaluation of a single straight line projection (such as misjudging a single line as the label corresponding to the defective material type).
[0141] S650, compare the first percentage with the first preset percentage, and compare the second percentage with the second preset percentage.
[0142] S660. If the first proportion is greater than the first preset proportion and the second proportion is greater than the second preset proportion, then the workpiece type corresponding to the mark on the workpiece to be inspected is determined to be a defective material type.
[0143] Here, the first preset percentage and the second preset percentage can be the same value, such as both being 0.9; the first preset percentage and the second preset percentage can also be different values, such as the first preset percentage being 0.9 and the second preset percentage being 0.83. Figure 8 As shown, in Figure 8 The thicker area in each straight line is the area projected by the skeleton line. The first preset proportion S1 is 0.91 and the second preset proportion S2 is 0.84. Then it can be determined that the workpiece type corresponding to the mark on the workpiece to be inspected is a defective material type.
[0144] Here, by requiring that the first proportion be greater than the first preset proportion and the second proportion be greater than the second preset proportion, the workpiece type corresponding to the mark on the workpiece to be inspected is determined to be a defective material type. This creates a dual constraint on the determination result, thereby improving the accuracy of the inspection result judgment.
[0145] Using the above method, the detection results of defective material type identification can be quantified into the proportion of the first and second categories, forming a clear normalized score. This avoids the problem that existing methods (connected component analysis, Hough transform) can only make qualitative judgments and cannot quantify the detection accuracy.
[0146] In some possible implementations, for each pixel on the skeleton line, if the minimum distance between the pixel and the preset identification reference line is less than a first preset distance, the following steps may be included before projecting the pixel onto the identification reference line:
[0147] S710. For each contour feature point, determine whether there is a partial line segment of the skeleton line in the area centered on the contour feature point in the surface image.
[0148] S720. If there is no part of the skeleton line in any region centered on the contour feature point, then the workpiece type corresponding to the mark on the workpiece to be inspected is determined as the good material type.
[0149] Here, the contour feature points are the core anchor points of the identifier (intersecting straight lines) corresponding to the defective material type. If there are no line segments in the skeleton line in the area centered on a certain contour feature point, it is determined that the skeleton line does not fit the core features of the identifier corresponding to the defective material type, and it can be directly and preliminarily judged as a non-defective material, thereby improving the overall efficiency of identifier detection.
[0150] S730. If all regions centered on the contour feature point contain a portion of the skeleton line, then for each region centered on the contour feature point, calculate the deviation between the portion of the skeleton line located within that region and the portion of the marker line located within that region.
[0151] After confirming that all regions centered on contour feature points contain segments of the skeleton line, it can be preliminarily determined that the marking corresponding to this workpiece type may be a defective material type. In order to accurately capture the positional deviation between the skeleton line and the identification reference line and clarify the calibration direction of the identification reference line, in this step, the deviation value between the corresponding segments of the skeleton line and the identification reference line in the region centered on the contour feature points is calculated. This transforms the visual phenomenon of "the two positions not coinciding" into a quantifiable value, clearly defining the magnitude and direction of the deviation, and providing a clear basis for the subsequent calibration of the identification reference line. S740, If the deviation value is greater than the preset deviation value, the position of the identification reference line is adjusted in the surface image until the deviation value between the corresponding segments of the skeleton line and the identification reference line in each region centered on the contour feature point is less than the preset deviation value.
[0152] This step enables dynamic calibration of the marker reference line position, ensuring that the dynamically calibrated marker reference line accurately aligns with the skeleton line, thereby improving the accuracy of subsequent projection inspection.
[0153] By adopting the above method, not only can the identification of non-defective material types be eliminated in advance through pre-verification, avoiding misjudgment of good material types, but also the deviation of the identification reference line that can be corrected by dynamic calibration can avoid misjudgment of defective materials due to the offset of the reference line. The double protection greatly reduces the overall misjudgment rate, ensures the accurate distinction between good and bad materials, and reduces the risk of loss of good products and defective products flowing into subsequent processes.
[0154] In some possible implementations, before projecting the pixel onto the marker reference line for each pixel on the skeleton line if the minimum distance between the pixel and the preset marker reference line is less than a first preset distance, the following steps may also be included:
[0155] S810. Remove pixels from the skeleton line whose minimum distance to the identification comparison line is greater than the second preset distance, and obtain the set of pixels to be compared.
[0156] Here, pixels whose minimum distance to the identification line is greater than the second preset distance (such as image noise, skeleton pixels of irrelevant printed characters, and redundant pixels at the edge of the defective material identification) are removed. This ensures that the set of pixels to be compared only contains valid pixels related to the defective material identification, and avoids irrelevant pixels from interfering with the distance statistics and preset distance calculation in subsequent steps, thereby improving the pertinence and reliability of the first preset distance calculation in subsequent steps.
[0157] S820: Calculate the minimum distance between each pixel in the set of pixels to be compared and the identification line.
[0158] Here, by statistically analyzing the minimum distance between each pixel in the set of pixels to be compared and the marker line, we can fully grasp the distribution of the fit between the skeleton line and the marker line, and clearly present the deviation between the skeleton line and the marker line.
[0159] S830. Based on the value corresponding to the minimum distance between each pixel and the identification line in the statistical set of pixels to be compared, calculate the average value corresponding to the minimum distance between the pixel and the identification line.
[0160] Here, the average value corresponding to the minimum distance between the calculated pixel and the marker line can comprehensively reflect the overall distance level between all the pixels to be compared and the marker line, avoiding the influence of extreme deviations (such as local noise or slight breaks) of a single pixel on the calculation of the first preset distance. Therefore, by using the average value calculated in this step as the core calculation parameter of the first preset distance, the overall fit between the skeleton line and the marker line can be accurately reflected, providing a reliable basis for subsequent optimization of the first preset distance by combining the variance value.
[0161] S840. Calculate the variance of the minimum distance and average distance between each pixel in the set of pixels to be compared and the identification line to obtain the variance value.
[0162] Since the variance value reflects the minimum distance between each pixel in the pixel set and the identification reference line, and the degree of deviation from the average value, a large variance value indicates that the skeleton line and the identification reference line are not evenly attached (such as local offset of handwritten marks or local breakage of laser marks); a small variance value indicates that the skeleton line and the identification reference line are evenly attached. Therefore, the variance value calculated in this step can be used as an important calculation parameter for the first preset distance, which can quickly identify local abnormalities in the attachment of the skeleton line and the identification reference line, and provide a clear direction for the dynamic adjustment of the first preset distance.
[0163] S850, Calculate the mean and variance values, and use the calculation results as the first preset distance.
[0164] Here, by combining the average value (overall fit level) and the variance value (dispersion) for dynamic calculation, the obtained first preset distance can be adapted to the specific shape of the current skeleton line, avoiding the situation of projection omission (threshold too small) or misjudgment (threshold too large) caused by a fixed threshold.
[0165] By using the above method, the limitations of the traditional fixed threshold can be overcome by dynamically calculating the first preset distance. The first preset distance can be flexibly adjusted according to the current fit between the skeleton line and the marking reference line (overall level, degree of dispersion). This allows the obtained first preset distance to be adapted to markings corresponding to different types of defective materials, such as handwritten and laser-printed markings, and significantly improves the accuracy of the skeleton line projecting onto the marking reference line.
[0166] In some possible implementations, image skeleton extraction of the surface image to obtain the skeleton lines corresponding to the pattern in the surface image may include the following steps:
[0167] S910. Perform binarization processing on the surface image to obtain a binarized image.
[0168] See Figure 3 and Figure 4 It can be seen that by binarizing the surface image, redundant grayscale information can be extracted from the surface image, thereby simplifying the processing difficulty of the surface image in subsequent steps. That is, it can convert the color or grayscale surface image into a black and white binary image, retaining only the two pixel states of "logo and text" and "background", thereby removing irrelevant grayscale gradients and color differences in the surface image, greatly simplifying the complexity of subsequent image processing.
[0169] Meanwhile, by binarizing the surface image, the markings (X-shaped for defective materials, O-shaped for good materials), irrelevant printed words and the background of the workpiece can be distinguished, avoiding the blurring of the target caused by the grayscale interference of the background in the surface image, and ensuring that the outlines of markings, printed words and other features can be accurately identified.
[0170] S920. Extract the contour map corresponding to the pattern from the binarized image.
[0171] In this step, by extracting the contour map corresponding to the pattern from the binarized image, the target pattern and the background can be separated, avoiding interference from background pixels in the surface image on the subsequent extraction of skeleton lines, and ensuring that the processing object focuses on the logo and printed words in the surface pattern.
[0172] S930. Extract the skeleton from the outline of the pattern to obtain the skeleton line of the pattern.
[0173] Among them, the skeleton line is the outline line with a single-layer pixel width.
[0174] Here, by extracting the skeleton from the outline of the pattern, the outline of the sign with uneven thickness (such as a handwritten X) can be simplified into a skeleton line with a single pixel width. This solves the problem of uneven thickness of handwritten signs and inconsistent line width of laser-printed signs, so that signs of different shapes can be converted into skeleton lines with a single pixel width.
[0175] Using the above method, the skeleton lines corresponding to the patterns in the obtained surface image are...
[0176] It can effectively address the uneven thickness and random intersection angles of handwritten markings (such as X-shaped markings), as well as the brokenness and varying line widths of laser-printed markings, ensuring that the extracted skeleton lines can accurately and completely reflect the core characteristics of the markings corresponding to the defective material type, thus laying a solid foundation for marking detection.
[0177] Figure 9 This is a schematic diagram of a marking and detection device 1000 for semiconductor packaged workpieces according to an embodiment of the present invention. The marking and detection device 1000 for semiconductor packaged workpieces includes:
[0178] The image acquisition module 1010 is used to acquire a surface image of the workpiece to be inspected; the surface image includes an identifier for marking the type of workpiece to be inspected; the workpiece type is either good material or bad material.
[0179] The image skeleton extraction module 1020 is used to extract the image skeleton from the surface image to obtain the skeleton lines corresponding to the patterns in the surface image.
[0180] The projection module 1030 is used to project each pixel on the skeleton line onto the identification reference line if the distance between the pixel and the preset identification reference line is less than a first preset distance; the identification reference line is drawn in the surface image according to the identification shape corresponding to the defective material type.
[0181] The marking determination module 1040 is used to determine the workpiece type corresponding to the marking on the workpiece to be inspected based on the proportion of the area projected by the skeleton line in the marking reference line.
[0182] In some implementations, the marking and detection device 1000 for the semiconductor packaged workpiece further includes:
[0183] The contour feature point determination module is used to obtain the contour feature points of the identifier shape corresponding to the defective material type based on the contour information of the identifier shape corresponding to the defective material type.
[0184] The position information determination module is used to determine the position information of contour feature points in the surface image based on the size information of the surface image;
[0185] The contour feature point calibration module is used to calibrate each contour feature point in the surface image according to its position information in the surface image.
[0186] The identification reference line drawing module is used to draw lines in the surface image based on the contour feature points marked in the surface image to obtain an identification reference line with the same contour as the identification of the defective material type.
[0187] In some implementations, the identifier shape corresponding to the defective material type includes a shape with two intersecting straight lines, the two straight lines being a first straight line and a second straight line; correspondingly, the contour feature point determination module includes:
[0188] The information determination unit is used to determine the intersection point between the first straight line and the second straight line, as well as the direction of the first straight line and the second straight line, based on the contour information corresponding to the first straight line and the second straight line respectively.
[0189] The target point determination unit is used to determine at least two first target points in the first straight line and at least two second target points in the second straight line based on the intersection point between the first straight line and the second straight line, and the respective directions of the first straight line and the second straight line.
[0190] The feature point determination unit is used to identify the intersection point, the first target point, and the second target point as contour feature points, respectively.
[0191] In some implementations, the location information determination module includes:
[0192] The first position information determining unit is used to determine the position information of the intersection point in the surface image based on the size information of the surface image, a first ratio of the distance between the intersection point and the two endpoints of the first straight line, and / or a second ratio of the distance between the intersection point and the two endpoints of the second straight line.
[0193] The second position information determination unit is used to determine the position information of the first target point and the second target point in the surface image based on the position correspondence between the first target point and the second target point and the intersection point, respectively.
[0194] In some implementations, the identifier determination module 1040 includes:
[0195] The first pixel length value determination unit is used to calculate the pixel length of the region in the first straight line that is projected by the skeleton line to obtain the first pixel length value.
[0196] The first proportion determination unit is used to calculate the length value of the first pixel and the length value of the first straight line to obtain the first proportion of the area projected by the skeleton line in the first straight line.
[0197] The second pixel length value determination unit is used to calculate the pixel length of the area projected by the skeleton line in the second straight line to obtain the second pixel length value.
[0198] The second proportion determination unit is used to calculate the length value of the second pixel and the length value of the second line to obtain the second proportion of the area projected by the skeleton line in the second line.
[0199] The comparison unit is used to compare the first proportion with the first preset proportion, and to compare the second proportion with the second preset proportion;
[0200] The identification determination unit is used to determine that the workpiece type corresponding to the identification on the workpiece to be inspected is a defective material type if the first proportion is greater than the first preset proportion and the second proportion is greater than the second preset proportion.
[0201] In some implementations, the marking and detection device 1000 for the semiconductor packaged workpiece further includes:
[0202] The judgment module is used to determine, for each contour feature point, whether there is a part of the skeleton line segment in the area centered on the contour feature point in the surface image.
[0203] The good material type determination module is used to determine the workpiece type corresponding to the mark on the workpiece to be inspected as good material type if there is no part of the skeleton line in any region centered on the contour feature point.
[0204] The deviation value calculation module is used to calculate the deviation value between the partial line segment of the skeleton line located in the region centered on the contour feature point and the partial line segment of the identification control line located in the region for each region centered on the contour feature point.
[0205] The position adjustment module is used to adjust the position of the marker line in the surface image if the deviation value is greater than the preset deviation value, so that the deviation value between the corresponding part of the skeleton line and the marker line in each region centered on the contour feature point is less than the preset deviation value.
[0206] In some implementations, the marking and detection device 1000 for the semiconductor packaged workpiece further includes:
[0207] The pixel set generation module is used to remove pixels from the skeleton line whose minimum distance to the identification comparison line is greater than a second preset distance, so as to obtain the pixel set to be compared.
[0208] The statistics module is used to calculate the minimum distance between each pixel in the set of pixels to be compared and the identification control line.
[0209] The average value calculation module is used to calculate the average value corresponding to the minimum distance between each pixel and the identification control line in the statistical set of pixels to be compared.
[0210] The variance calculation module is used to calculate the variance between the minimum distance and the average distance between each pixel in the set of pixels to be compared and the identification control line, and obtain the variance value.
[0211] The mean and variance values are calculated, and the results are used as the first preset distance.
[0212] In some implementations, the image skeleton extraction module 1020 includes:
[0213] The binarization processing unit is used to perform binarization processing on the surface image to obtain a binarized image;
[0214] The contour extraction unit is used to extract the contour map corresponding to the pattern from the binarized image;
[0215] The skeleton line acquisition unit is used to extract the skeleton from the contour map corresponding to the pattern to obtain the skeleton line corresponding to the pattern; the skeleton line is a contour line with a single-layer pixel width.
[0216] It should be understood that the embodiments of the marking and detection apparatus for semiconductor packaged workpieces and the embodiments of the marking and detection method for semiconductor packaged workpieces can correspond to each other, and similar descriptions can be found in the embodiments of the marking and detection method for semiconductor packaged workpieces. To avoid repetition, further details are omitted here. Specifically, Figure 9The semiconductor package workpiece marking detection device 1000 shown can execute the above-described semiconductor package workpiece marking detection method embodiment. The aforementioned and other operations and / or functions of each module in the semiconductor package workpiece marking detection device 1000 are respectively for implementing the corresponding process in the above-described semiconductor package workpiece marking detection method. For the sake of brevity, they will not be described in detail here.
[0217] The semiconductor packaged workpiece identification and detection device 1000 of this invention has been described above from the perspective of functional modules in conjunction with the accompanying drawings. It should be understood that this functional module can be implemented in hardware, in software instructions, or in a combination of hardware and software modules. Specifically, the steps of the semiconductor packaged workpiece identification and detection method and the detection method embodiments of this invention can be completed by integrated logic circuits in the processor's hardware and / or by software instructions. The steps of the semiconductor packaged workpiece identification and detection method disclosed in this invention can be directly manifested as being executed by a hardware decoding processor, or by a combination of hardware and software modules in the decoding processor. Optionally, the software module can be located in a mature storage medium in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. This storage medium is located in memory, and the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above-described semiconductor packaged workpiece identification and detection method embodiments.
[0218] Figure 10 This is a schematic block diagram of an electronic device 110 according to an embodiment of the present invention.
[0219] like Figure 10 As shown, the electronic device 110 may include:
[0220] The system includes a memory 111 and a processor 112. The memory 111 stores computer programs and transfers the program code to the processor 112. In other words, the processor 112 can retrieve and run the computer programs from the memory 111 to implement the methods described in the embodiments of the present invention.
[0221] For example, the processor 112 can be used to execute the above-described method embodiments according to instructions in the computer program.
[0222] In some embodiments of the present invention, the electronic device 110 may include, but is not limited to:
[0223] General-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
[0224] In some embodiments of the present invention, the memory 111 includes, but is not limited to:
[0225] Volatile memory and / or non-volatile memory. Non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory can be random access memory (RAM), used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDR SDRAM), Enhanced Synchronous DRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), and Direct Rambus RAM (DR RAM).
[0226] In some embodiments of the present invention, the computer program may be divided into one or more modules, which are stored in the memory 111 and executed by the processor 112 to perform the method provided by the present invention. The one or more modules may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the controller.
[0227] like Figure 10 As shown, the electronic device 110 may further include:
[0228] Transceiver 113, which can be connected to processor 112 or memory 111.
[0229] The processor 112 can control the transceiver 113 to communicate with other devices; specifically, it can send information or data to other devices or receive information or data sent by other devices. The transceiver 113 may include a transmitter and a receiver. The transceiver 113 may further include antennas, and the number of antennas may be one or more.
[0230] It should be understood that the various components in the electronic device are connected through a bus system, which includes a data bus, a power bus, a control bus, and a status signal bus.
[0231] The present invention also provides a cutting and sorting integrated machine, including the above-mentioned marking and detection device for semiconductor packaged workpieces or the above-mentioned electronic equipment.
[0232] The present invention also provides a computer storage medium having a computer program stored thereon, which, when executed by a computer, enables the computer to perform the methods of the above-described method embodiments. Alternatively, one embodiment of the present invention also provides a computer program product containing instructions that, when executed by a computer, cause the computer to perform the methods of the above-described method embodiments.
[0233] When implemented using software, it can be implemented entirely or partially as a computer program product. This computer program product includes one or more computer instructions. When these computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., Digital Video Disc (DVD)), or a semiconductor medium (e.g., Solid State Disk (SSD)).
[0234] Those skilled in the art will recognize that the modules 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.
[0235] In the several embodiments provided by this invention, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or modules may be electrical, mechanical, or other forms.
[0236] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. For example, the functional modules in the various embodiments of this application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.
[0237] The above are merely specific embodiments of the present invention, but 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 scope of the technology disclosed in this application should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.
Claims
1. A method for identifying and detecting semiconductor packaged workpieces, characterized in that, include: Acquire a surface image of the workpiece to be inspected; the surface image includes an identifier for marking the workpiece type; the workpiece type is either a good material type or a defective material type; Image skeleton extraction is performed on the surface image to obtain the skeleton lines corresponding to the patterns in the surface image; For each pixel on the skeleton line, if the minimum distance between the pixel and the preset identification reference line is less than a first preset distance, then the pixel is projected onto the identification reference line; the identification reference line is drawn in the surface image according to the identification shape corresponding to the defective material type; The type of workpiece corresponding to the marking on the workpiece to be inspected is determined based on the proportion of the area projected by the skeleton line in the marking reference line.
2. The identification and detection method for semiconductor packaged workpieces according to claim 1, characterized in that, Also includes: Based on the contour information of the identifier shape corresponding to the defective material type, the contour feature points of the identifier shape corresponding to the defective material type are obtained; Based on the size information of the surface image, determine the position information of the contour feature points in the surface image; For each of the contour feature points, the contour feature points are calibrated in the surface image according to their position information in the surface image; Based on the contour feature points marked in the surface image, lines are drawn in the surface image to obtain a marker line that corresponds to the contour of the defective material type.
3. The identification and detection method for semiconductor packaged workpieces according to claim 2, characterized in that, The identification shape corresponding to the defective material type includes a shape with two intersecting straight lines, wherein the two straight lines include a first straight line and a second straight line; correspondingly, based on the contour information of the identification shape corresponding to the defective material type, the contour feature points of the identification shape corresponding to the defective material type are obtained, including: Based on the contour information corresponding to the first straight line and the second straight line, determine the intersection point between the first straight line and the second straight line, as well as the direction of the first straight line and the second straight line respectively; Based on the intersection point between the first line and the second line, and the respective directions of the first line and the second line, determine at least two first target points in the first line and at least two second target points in the second line; The intersection point, the first target point, and the second target point are respectively used as the contour feature points.
4. The identification and detection method for semiconductor packaged workpieces according to claim 3, characterized in that, Determining the position information of the contour feature points in the surface image based on the size information of the surface image includes: The position information of the intersection point in the surface image is determined based on the size information of the surface image, a first ratio of the distance between the intersection point and the two endpoints of the first straight line, and / or a second ratio of the distance between the intersection point and the two endpoints of the second straight line. Based on the positional correspondence between the first target point and the second target point and the intersection point, the positional information of the first target point and the second target point in the surface image is determined.
5. The identification and detection method for semiconductor packaged workpieces according to claim 3, characterized in that, Determining the workpiece type corresponding to the marking on the workpiece to be inspected based on the proportion of the area projected by the skeleton line in the marking reference line includes: The pixel length of the region in the first straight line that is projected by the skeleton line is calculated to obtain the first pixel length value; The length values of the first pixel and the first straight line are calculated to obtain the first proportion of the area in the first straight line that is projected by the skeleton line. The pixel length of the region in the second straight line that is projected by the skeleton line is calculated to obtain the second pixel length value; The second pixel length value and the second line length value are calculated to obtain the second proportion of the area in the second line that is projected by the skeleton line; Compare the first percentage with the first preset percentage, and compare the second percentage with the second preset percentage; If the first proportion is greater than the first preset proportion and the second proportion is greater than the second preset proportion, then the workpiece type corresponding to the mark on the workpiece to be tested is determined to be a defective material type.
6. The identification and detection method for semiconductor packaged workpieces according to claim 2, characterized in that, Before projecting the pixel onto the marker reference line for each pixel on the skeleton line, if the minimum distance between the pixel and the preset marker reference line is less than a first preset distance, the method further includes: For each of the contour feature points, determine whether there is a partial line segment of the skeleton line in the area centered on the contour feature point in the surface image. If there is no part of the skeleton line in any region centered on the contour feature point, then the workpiece type corresponding to the mark on the workpiece to be inspected is determined to be a good material type. If all regions centered on the contour feature point contain a portion of the skeleton line, then for each region centered on the contour feature point, calculate the deviation between the portion of the skeleton line located within that region and the portion of the identification reference line located within that region. If the deviation value is greater than a preset deviation value, the position of the marker reference line in the surface image is adjusted so that the deviation value between the corresponding line segments of the skeleton line and the marker reference line in each region centered on the contour feature point is less than the preset deviation value.
7. The identification and detection method for semiconductor packaged workpieces according to claim 1, characterized in that, Before projecting the pixel onto the marker reference line for each pixel on the skeleton line if the minimum distance between the pixel and the preset marker reference line is less than a first preset distance, the method further includes: Pixels whose minimum distance from the skeleton line to the identification comparison line is greater than a second preset distance are removed from the skeleton line to obtain a set of pixels to be compared. The minimum distance between each pixel in the set of pixels to be compared and the identification line is statistically analyzed. Based on the value corresponding to the minimum distance between each pixel in the set of pixels to be compared and the identification reference line, calculate the average value corresponding to the minimum distance between the pixel and the identification reference line. The variance value is obtained by calculating the variance between the minimum distance between each pixel in the set of pixels to be compared and the identification line and the average value. The average value and the variance value are calculated, and the calculation result is used as the first preset distance.
8. The identification and detection method for semiconductor packaged workpieces according to claim 1, characterized in that, The step of extracting the image skeleton from the surface image to obtain the skeleton lines corresponding to the pattern in the surface image includes: The surface image is binarized to obtain a binarized image; Extract the contour map corresponding to the pattern from the binarized image; Skeleton extraction is performed on the contour map corresponding to the pattern to obtain the skeleton line corresponding to the pattern; the skeleton line is a contour line with a single-layer pixel width.
9. A marking and detection device for semiconductor packaged workpieces, characterized in that, include: An image acquisition module is used to acquire a surface image of a workpiece to be inspected; the surface image includes an identifier for marking the type of the workpiece to be inspected; the workpiece type is either a good material type or a defective material type; The image skeleton extraction module is used to extract the image skeleton from the surface image to obtain the skeleton lines corresponding to the patterns in the surface image. The projection module is used to project each pixel on the skeleton line onto the identification reference line if the distance between the pixel and the preset identification reference line is less than a first preset distance; the identification reference line is drawn in the surface image according to the identification shape corresponding to the defective material type. The identification determination module is used to determine the workpiece type corresponding to the identification on the workpiece to be inspected based on the proportion of the area projected by the skeleton line in the identification reference line.
10. An electronic device, characterized in that, include: A processor and a memory, the memory being used to store a computer program, the processor being used to invoke and run the computer program stored in the memory to perform the method of any one of claims 1-8.
11. A computer-readable storage medium, characterized in that, Used to store a computer program that causes a computer to perform the method as described in any one of claims 1-8.
12. A cutting and sorting integrated machine, characterized in that, This includes the identification and detection device for semiconductor packaged workpieces as described in claim 9, or the electronic device as described in claim 10.