A method and apparatus for detecting chip defects

Abstract

This invention relates to the technical field of image detection, and provides a method and apparatus for detecting chip defects. The chip defect detection method includes: acquiring an image to be identified by a vision module; extracting a chip image region from the image to be identified, and extracting a current package region and multiple current pin regions from the chip image region; acquiring a standard package region, and based on the standard package region and the current package region, confirming whether the current package region has a package defect; when it is determined that the current package region does not have a defect, determining whether multiple current pin regions have pin defects based on the current package region; when it is determined that the current pin regions do not have defects, determining whether there is a positional distribution defect between the package region and the pin regions based on the current package region and the multiple current pin regions. The chip defect detection method provided by this invention offers a comprehensive and accurate detection solution for package defects, pin defects, and positional distribution defects.

Description

Technical Field

[0001] This invention belongs to the technical field of image detection, and particularly relates to a method and apparatus for detecting chip defects. Background Technology

[0002] With the rapid development of the semiconductor industry, chips, as core components of modern electronic products, directly affect the performance and stability of these products. The chip's packaging and pin areas are critical components determining its function and reliability. However, during the manufacturing process, due to technical, material, or operational issues, various defects can easily occur in the chip's packaging and pin areas. These defects can affect the chip's normal operation and even lead to product failure. Common defects include missing packages, damaged packages, abnormal printed patterns, pin misalignment, unevenness, or missing pins. These defects are often difficult to detect, but they can seriously affect the chip's performance and reliability.

[0003] Traditional chip defect detection methods mostly rely on manual inspection or simple automated testing techniques. These methods suffer from low efficiency, high error rates, and inability to detect all defects. Furthermore, with the continuous advancement of chip packaging technology and the increasing complexity of packaging, traditional testing techniques face significant challenges in handling large-scale, high-precision chip inspection. Therefore, improving the efficiency and accuracy of chip defect detection, especially in the detection of defects in packaged and pin areas, has become an urgent technical problem to be solved. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide a method and apparatus for detecting chip defects, in order to solve the technical problem that traditional detection technologies face significant challenges in handling large-scale, high-precision chip detection.

[0005] A first aspect of the present invention provides a method for detecting chip defects, the method comprising:

[0006] Acquire the image to be recognized captured by the vision module;

[0007] Extract the chip image region from the image to be identified, and extract the current package region and multiple current pin regions from the chip image region;

[0008] Obtain a standard packaging area, and based on the standard packaging area and the current packaging area, confirm whether there is a packaging defect in the current packaging area; the packaging defect includes missing packaging, damaged packaging, or abnormal printing pattern;

[0009] When it is determined that there are no defects in the current package area, it is determined whether pin defects occur in multiple current pin areas based on the current package area; the pin defects include pin offset defects, pin unevenness defects, and pin missing defects;

[0010] When it is determined that there are no defects in the current pin area, it is determined whether there are positional distribution defects between the package area and the pin area based on the current package area and multiple current pin areas.

[0011] Further, the step of extracting the chip image region from the image to be identified, and extracting the current package region and multiple current pin regions from the chip image region includes:

[0012] Obtain the first color distribution range corresponding to the base and the second color distribution range corresponding to the pin;

[0013] Obtain multiple first pixels corresponding to the first color distribution range in the image to be identified;

[0014] The chip image region is obtained by removing multiple consecutive first pixels from the image to be identified.

[0015] Obtain multiple second pixel points corresponding to the second color distribution range in the chip image region;

[0016] Obtain multiple image regions consisting of multiple consecutive second pixels, and use these multiple image regions as multiple current pin regions;

[0017] The current packaging area is obtained by removing multiple consecutive second pixels in the chip image area.

[0018] Furthermore, the step of obtaining a standard packaging area, and based on the standard packaging area and the current packaging area, confirming whether the current packaging area has a packaging defect includes:

[0019] Obtain the distribution range of the third color corresponding to the encapsulation material;

[0020] In the current encapsulation area, extract multiple third pixel points corresponding to the third color distribution range;

[0021] The image region composed of multiple third pixels is used as the first encapsulation material region;

[0022] Obtain the second encapsulation material region within the standard encapsulation region;

[0023] Align the first encapsulation material region with the second encapsulation material region, and calculate the pixel difference between each corresponding pixel position;

[0024] If the pixel difference between multiple consecutive pixel positions is greater than the first threshold, and the number of such consecutive pixel positions exceeds the second threshold, then it is confirmed that the current encapsulation area has encapsulation defects or encapsulation damage.

[0025] If the pixel difference corresponding to multiple consecutive pixel positions is not greater than the first threshold or the number of multiple consecutive pixel positions does not exceed the second threshold, then it is confirmed that there is no missing or damaged encapsulation in the current encapsulation area.

[0026] Based on the distribution relationship between the four diagonal points corresponding to the current packaging area and the preset feature points, it is confirmed whether there is any printing pattern abnormality in the current packaging area.

[0027] Furthermore, the step of confirming whether there is a printing pattern abnormality in the current packaging area based on the distribution relationship between the four diagonal points corresponding to the current packaging area and the preset feature points includes:

[0028] Obtain the four diagonal points corresponding to the current encapsulation area;

[0029] Extract two preset feature points corresponding to the printed pattern in the current packaging area;

[0030] Calculate the first distance between each preset feature point and the four diagonal points;

[0031] Obtain multiple second distances corresponding to the standard encapsulation region; the multiple second distances refer to the distances between preset feature points and four diagonal points in the standard encapsulation region;

[0032] Calculate the first difference between the first distance and the second distance corresponding to the same point;

[0033] If the first difference is less than the third threshold, it is confirmed that there is no printing pattern abnormality in the current packaging area;

[0034] If the first difference is not less than the third threshold, it is confirmed that there is an abnormal printing pattern in the current packaging area.

[0035] Furthermore, the step of extracting two preset feature points corresponding to the printed pattern in the current packaging area includes:

[0036] Obtain the standard feature matrix corresponding to the preset feature points;

[0037] In the current encapsulation region, with each current pixel as the center, extract the pixel matrix surrounding the current pixel; wherein, the pixel matrix has the same matrix size as the standard feature matrix;

[0038] Calculate the second difference between the pixel value of the current pixel and the pixel values ​​of other pixels in the pixel matrix;

[0039] The pixel values ​​of other pixels in the pixel matrix are replaced with the second difference to obtain the current feature matrix;

[0040] Calculate the similarity between each current feature matrix and the standard feature matrix;

[0041] The current pixel corresponding to the current feature matrix with the highest similarity is taken as the preset feature point.

[0042] Furthermore, the step of determining whether pin defects occur in multiple current pin regions based on the current package region when it is determined that there are no defects in the current package region includes:

[0043] When it is determined that there are no defects in the current packaging area, a preset first packaging pin ratio and a preset second packaging pin ratio are obtained; the first packaging pin ratio refers to the ratio between the first length of the standard packaging area and the second length of a single standard pin area, and the second packaging pin ratio refers to the ratio between the first length of the standard packaging area and the first width of a single standard pin area.

[0044] Obtain the third length of the current encapsulation region;

[0045] Based on the first package pin ratio and the third length, calculate the fourth length corresponding to the pin region;

[0046] Calculate the third width corresponding to the pin area based on the second package pin ratio and the third length;

[0047] Based on the third length, the fourth length, and the third width, it is determined whether pin defects occur in multiple current pin regions.

[0048] Further, the step of determining whether pin defects occur in the plurality of current pin regions based on the third length, the fourth length, and the third width includes:

[0049] Obtain the fifth length corresponding to each of the multiple current pin regions;

[0050] Calculate the third difference between the fourth length and the fifth length;

[0051] If the third difference is less than the fourth threshold, then the second width of multiple sampling points in each current pin region is obtained;

[0052] Calculate a fourth difference between the third width and the plurality of second widths;

[0053] If the fifth difference between multiple second widths in the same current pin region is less than the fourth threshold, and multiple fourth differences are less than the fifth threshold, then it is confirmed that there is no defect in the current pin region.

[0054] If the third difference is not less than the fourth threshold, then it is determined that there is a pin offset defect in the current pin area;

[0055] If the fifth difference is not less than the fourth threshold or the fourth difference is not less than the fifth threshold, then it is determined that there is a pin unevenness defect in the current pin area.

[0056] If the number of pins in the current pin area is less than the preset number, it is confirmed that there is a pin missing defect in the current pin area.

[0057] Furthermore, when it is determined that there is no defect in the current pin region, the step of determining whether there is a positional distribution defect between the package region and the pin region based on the current package region and multiple current pin regions includes:

[0058] Obtain the included angle fluctuation range corresponding to the standard reference image; the included angle fluctuation range is used to describe the distribution relationship between multiple pin regions;

[0059] Extract the first center of the current package region, and extract the second centers of multiple current pin regions;

[0060] Establish a first coordinate system with the first center as the origin;

[0061] In the first coordinate system, obtain the first coordinate positions of multiple second centers;

[0062] Extract multiple first line segments formed by multiple first coordinate positions and the first center;

[0063] Among multiple first line segments, the first minimum distance line segment corresponding to each first line segment is extracted; wherein, the first minimum distance line segment is the line segment with the smallest distance from the first line segment.

[0064] Calculate the first included angle between the first line segment and the first line segment with the minimum distance;

[0065] If multiple first included angles are all within the included angle fluctuation range, then it is determined that there is no positional distribution defect between the package area and the pin area;

[0066] If multiple first included angles are not uniformly within the included angle fluctuation range, then it is determined that there is a positional distribution defect between the package area and the pin area.

[0067] Furthermore, the step of obtaining the included angle fluctuation range corresponding to the standard reference image includes:

[0068] Obtain a standard reference image, and extract the standard package area and multiple standard pin areas from the standard reference image;

[0069] Extract the third center of the standard package region and extract the fourth center of multiple standard pin regions;

[0070] A second coordinate system is established with the third center as the origin.

[0071] In the second coordinate system, the second coordinate positions of the multiple fourth centers are obtained;

[0072] Extract multiple second line segments formed by multiple second coordinate positions and the third center;

[0073] Among multiple second line segments, extract the second minimum distance line segment corresponding to each second line segment; where the second minimum distance line segment is the line segment with the smallest distance from the second line segment.

[0074] Calculate the second included angle between the second line segment and the second minimum distance line segment respectively;

[0075] The first fluctuation range formed by multiple second included angles is statistically analyzed;

[0076] Based on a preset redundancy coefficient, the first fluctuation range is expanded to obtain the included angle fluctuation range.

[0077] A second aspect of the present invention provides a chip defect detection apparatus, comprising:

[0078] The acquisition unit is used to acquire the image to be recognized collected by the vision module;

[0079] The extraction unit is used to extract the chip image region in the image to be identified, and to extract the current package region and multiple current pin regions in the chip image region;

[0080] The first judgment unit is used to obtain a standard packaging area and, based on the standard packaging area and the current packaging area, to determine whether there is a packaging defect in the current packaging area; the packaging defect includes missing packaging, damaged packaging, or abnormal printing pattern.

[0081] The second determination unit is used to determine whether pin defects occur in multiple current pin areas based on the current packaging area when it is determined that there are no defects in the current packaging area; the pin defects include pin offset defects, pin unevenness defects and pin missing defects;

[0082] The third judgment unit is used to determine whether there is a positional distribution defect between the package area and the pin area based on the current package area and multiple current pin areas when it is determined that there is no defect in the current pin area.

[0083] A third aspect of the present invention provides a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps in the chip defect detection method described in the first aspect.

[0084] A fourth aspect of the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps in the chip defect detection method described in the first aspect.

[0085] The beneficial effects of this invention compared to existing technologies are as follows: By extracting a standard packaging area and comparing it with the current packaging area, packaging defects can be effectively identified. Packaging defects include missing packaging, damaged packaging, and abnormal printed patterns, which directly affect the chip's functionality and reliability. Traditional detection methods often struggle to accurately distinguish between different types of packaging defects, while this invention, through comparative analysis of the standard packaging area and the current packaging area, can precisely locate packaging defects, avoiding missed and false detections. Furthermore, when the packaging area is determined to be defect-free, this invention, through comprehensive analysis of the current packaging area and the current pin area, can identify various pin defects, such as pin misalignment, pin unevenness, and pin loss. These pin defects often directly affect the chip's electrical performance, while traditional detection methods typically only detect some defect types. This invention's method can comprehensively detect and accurately identify various defects in the pin area, improving the comprehensiveness and accuracy of detection. In addition to defect detection in the packaging and pin areas, this invention can further determine the existence of positional distribution defects by analyzing the positional distribution between the packaging and pin areas. Traditional inspection methods typically focus only on localized defects, neglecting the relative positional relationship between the package and pin areas. This can lead to some minor but serious defects going undetected. By detecting positional defects, the accuracy of overall chip quality inspection can be improved, reducing chip failures caused by positional errors. In summary, the chip defect detection method provided by this invention offers a comprehensive and accurate detection solution for package defects, pin defects, and positional defects, greatly improving the quality control level of chip production and overcoming many shortcomings of traditional inspection methods. Attached Figure Description

[0086] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0087] Figure 1 A schematic flowchart of a chip defect detection method provided by the present invention is shown;

[0088] Figure 2 A schematic diagram of a chip defect detection device according to an embodiment of the present invention is shown;

[0089] Figure 3 A schematic diagram of a terminal device provided in an embodiment of the present invention is shown. Detailed Implementation

[0090] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of the invention. However, those skilled in the art will understand that the invention can be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods are omitted so as not to obscure the description of the invention with unnecessary detail.

[0091] This invention provides a method and apparatus for detecting chip defects, addressing the technical challenge that traditional detection technologies face when dealing with large-scale, high-precision chip testing.

[0092] First, this invention provides a method for detecting chip defects. Please refer to [link / reference]. Figure 1 , Figure 1 A schematic flowchart of a chip defect detection method provided by the present invention is shown. Figure 1 As shown, the method for detecting defects in this chip may include the following steps:

[0093] Step 101: Acquire the image to be recognized from the vision module;

[0094] A vision inspection module is used to capture images of the chip, obtaining image data. Images are acquired from multiple angles or fields of view. The images to be identified contain various details of the chip, including its appearance, packaging, pins, and other structural information.

[0095] Step 102: Extract the chip image region from the image to be identified, and extract the current package region and multiple current pin regions from the chip image region;

[0096] The first step in acquiring the image is image processing and analysis to locate the area where the chip is located. The "chip region" in the image is the target area, which includes the chip's packaging and pins. Using image segmentation techniques (such as edge detection and region segmentation), the chip's packaging region and multiple pin regions can be accurately extracted. The packaging region refers to the chip's outer casing, while the pin regions refer to the pins on the chip used to connect to external circuits.

[0097] Specifically, step 102 includes steps 1021 to 1026:

[0098] Step 1021: Obtain the first color distribution range corresponding to the base and the second color distribution range corresponding to the pin;

[0099] At this stage, the first step is to obtain the predefined color ranges for different parts (base and pins) in the chip image. The "first color distribution range" corresponding to the base refers to the color range in the image representing the bottom or background of the chip; it is the color of the background or base outside the chip. The "second color distribution range" corresponding to the pins refers to the color range of the chip pins, which is a metallic color or other color that is distinct from the base area.

[0100] These two color distribution ranges can be obtained using color space thresholding. For example, the base plate's color may contain a certain range of red or gray, and a specific threshold range can be used to identify the base plate portion in the image; the pin region's color is usually distinct from the base plate and can be extracted using the corresponding color recognition range.

[0101] Step 1022: Obtain multiple first pixels corresponding to the first color distribution range in the image to be identified;

[0102] In this step, the system uses a predefined "first color distribution range" to perform pixel analysis on the image to be recognized, extracting all pixels that match the base color range. These pixels are distributed at the edges or in the background of the image, representing the base area of ​​the chip.

[0103] Based on color space thresholds (such as RGB, HSV, etc.), each pixel in the image is examined, and its belonging to the sill region is determined based on whether its color is within the "first color distribution range".

[0104] Step 1023: Remove multiple consecutive first pixels from the image to be identified to obtain the chip image region;

[0105] The purpose of this operation is to filter out areas in the image that do not belong to the chip itself, especially those that belong to the base plate. By removing consecutive first pixels from the image (that is, removing pixels in the base plate area), the "chip image region" can be obtained. The chip image region includes the chip's package area and pin area.

[0106] Step 1024: Obtain multiple second pixels corresponding to the second color distribution range in the chip image region;

[0107] After obtaining the chip image area, the next step is to identify the pin areas on the chip. To achieve this, the system uses a "second color distribution range" to extract pixels in the chip image area that match the colors of the pin areas. These second pixels are the colors of the chip pins and can be identified by setting a color threshold.

[0108] This process is similar to the previous color recognition. The system will detect pixels in the image that belong to the pin color range and extract the pixels in these pin regions.

[0109] Step 1025: Obtain multiple image regions composed of multiple consecutive second pixels, and use the multiple image regions as multiple current pin regions;

[0110] By analyzing the multiple second pixels extracted from the second color distribution range, the system partitions the image based on the connectivity of these pixels, resulting in multiple independent image regions, each representing a pin on the chip. Here, "continuous second pixels" refers to pixels connected in the image through color similarity; these pixels constitute a complete pin region.

[0111] A pin region is typically composed of multiple consecutive pixels that are grouped into the same region based on their position and shape. Thus, each consecutive second pixel region is considered a current pin region.

[0112] Step 1026: Remove multiple consecutive second pixels in the chip image area to obtain the current packaging area.

[0113] Finally, to extract the package area, the system removes all identified pin areas from the chip image area. By removing these consecutive second pixels (i.e., the identified pin portions), the remaining area is the chip's package area.

[0114] The elimination method is also based on connected component analysis, removing all pixels belonging to the pin area from the chip image area, ultimately obtaining only the portion containing the chip package area.

[0115] In the embodiments corresponding to steps 1021 to 1026, the base, package, and pin areas of the chip are effectively distinguished by the color distribution range. The system first defines the color range of the base and pins, then extracts pixels that match these color ranges, eliminates irrelevant areas, and finally accurately extracts the package and pin areas of the chip. This method can efficiently and accurately identify and segment key areas in chip images.

[0116] Step 103: Obtain the standard packaging area. Based on the standard packaging area and the current packaging area, confirm whether there is a packaging defect in the current packaging area. The packaging defect includes missing packaging, damaged packaging, or abnormal printing pattern.

[0117] In this step, a "standard package area" is used as a reference template. The standard package area is a known defect-free chip package area. By comparing the characteristics of the current package area with the standard package area, the system can detect whether defects exist in the package area. Package defects typically include the following types:

[0118] Missing package: The package is missing or damaged;

[0119] Damaged packaging: cracks, breaks, or other damage to the packaging;

[0120] Printing pattern abnormality: Errors or missing printed circuit or marking patterns.

[0121] Specifically, step 103 includes steps 1031 to 1038:

[0122] Step 1031: Obtain the third color distribution range corresponding to the encapsulation material;

[0123] A predefined color range for the packaging material—the "third color distribution range"—is established. The color of the packaging material is the color of the outer covering or encapsulation layer of the chip casing, which differs from the color of the chip's interior or pin areas. This color range needs to be defined using color space threshold settings to ensure accurate identification of the packaging material portions. This color information will help the system locate the packaging area and further determine if defects are present.

[0124] Step 1032: Extract multiple third pixel points corresponding to the third color distribution range in the current encapsulation area;

[0125] Next, the system will use a color recognition algorithm to extract pixels within the current packaging area that match the "third color distribution range." These pixels represent the packaging material portion within the chip packaging area. By identifying these color-matching pixels, the system can pinpoint the area where the packaging material is located, thereby further analyzing whether there are any missing or damaged packages.

[0126] Step 1033: Use the image region composed of multiple third pixels as the first encapsulation material region;

[0127] By connecting multiple third pixels that match the color range, the system constructs an image region containing the packaging material, called the "first packaging material region." This region represents the packaging portion of the chip, and the quality of the packaging material can be analyzed through these pixels to confirm the presence of defects.

[0128] Step 1034: Obtain the second encapsulation material region within the standard encapsulation region;

[0129] Corresponding to the packaging material region of the current packaging region, the standard packaging region provides a known defect-free reference packaging material region, which is called the "second packaging material region". The standard packaging region is predefined and represents the ideal packaging material region, signifying the normal form that a chip package should have under defect-free conditions. The extraction logic of the second packaging material region is the same as that of the first packaging material region, and will not be repeated here. The second packaging material region is pre-stored data.

[0130] Step 1035: Align the first encapsulation material region with the second encapsulation material region, and calculate the pixel difference between each corresponding pixel position;

[0131] In this step, the system aligns the "first encapsulation material region" (the encapsulation portion within the current encapsulation region) with the "second encapsulation material region" (the encapsulation portion within the standard encapsulation region) to ensure a one-to-one correspondence between the pixel positions of the two regions. After alignment, the system calculates the difference between each pixel position, i.e., the difference in pixel values ​​between the current encapsulation region and the standard encapsulation region at corresponding positions.

[0132] The purpose of this step is to quantify the differences in the encapsulation area and help determine whether there is any damage, missing parts or other abnormalities in the current encapsulation area.

[0133] Step 1036: If the pixel difference corresponding to multiple consecutive pixel positions is greater than the first threshold, and the number of the multiple consecutive pixel positions exceeds the second threshold, then it is confirmed that there is a missing or damaged encapsulation in the current encapsulation area.

[0134] If, when comparing the pixel differences between the current encapsulation area and the standard encapsulation area, multiple consecutive pixel locations are found to have large pixel differences (i.e., the differences exceed a preset "first threshold"), and the number of pixels exhibiting these differences exceeds a set threshold (i.e., a "second threshold"), then it can be determined that the encapsulation is missing or damaged. Specifically, missing encapsulation usually manifests as the complete absence of certain areas, while damaged encapsulation manifests as damage or incompleteness of the encapsulation material. By detecting pixel differences and consecutive areas, these defects can be accurately identified.

[0135] The first and second thresholds are determined based on extensive experimental data, engineering experience, or actual test results. Their values ​​need to be optimized through a series of experiments to accurately identify defects and avoid false positives during detection. For example:

[0136] First threshold: can be set as a standard for the difference between a certain pixel. For example, if the color difference between two pixels exceeds 10 (according to the definition of color space, such as RGB or Lab color difference), then the two pixels are considered to have a significant difference.

[0137] The second threshold can be set as the minimum consecutive defect length. For example, if the number of consecutive pixels with differences greater than the first threshold exceeds 5, then a defect is considered to exist.

[0138] Step 1037: If the pixel difference corresponding to multiple consecutive pixel positions is not greater than the first threshold or the number of multiple consecutive pixel positions does not exceed the second threshold, then it is confirmed that there is no encapsulation missing or encapsulation damage in the current encapsulation area.

[0139] Conversely, if, during comparison, the pixel difference between multiple consecutive pixel positions is less than or equal to the first threshold, or even if the difference is large but the number of consecutive pixel positions is insufficient to exceed the second threshold, then the system will consider the current encapsulation area to be normal and there is no encapsulation defect or damage.

[0140] This process actually avoids misjudgments by setting reasonable thresholds, ensuring that a defect is only identified as a defect when it is severe enough and persists over a certain area.

[0141] Step 1038: Based on the distribution relationship between the four diagonal points corresponding to the current packaging area and the preset feature points, confirm whether there is any printing pattern abnormality in the current packaging area.

[0142] Finally, the system analyzes the distribution relationship between the four diagonal points of the current packaging area and preset feature points to determine if there are any printing pattern anomalies. The packaging area typically has printed patterns or markings, such as chip numbers, trademarks, and QR codes. If these printed patterns are irregularly distributed or deviated within the packaging area, printing anomalies may exist. Analyzing the spatial relationship between the diagonal points and feature points can detect such anomalies.

[0143] If the distribution of these points does not match the standard template, the system can determine that there is an anomaly in the printed pattern.

[0144] In the embodiments corresponding to steps 1031 to 1038, through precise color recognition, pixel difference calculation, and comparison of the packaged area with the standard packaged area, defects such as missing packages, damaged packages, and abnormal printed patterns can be effectively detected. This method can provide efficient and accurate quality control during chip manufacturing, ensuring that the packaged portion of the chip meets design requirements.

[0145] Specifically, step 1038 includes steps 10381 to 10387:

[0146] Step 10381: Obtain the four diagonal points corresponding to the current encapsulation region;

[0147] Within the current package area, the four diagonal points of the package area must first be identified, typically rectangular or similar shaped regions. By locating these diagonal points, the spatial position and shape of the package area can be determined.

[0148] Step 10382: Extract two preset feature points corresponding to the printed pattern in the current packaging area;

[0149] The printed pattern area contains specific logos, QR codes, or other design elements, which can be located using "preset feature points." These feature points have significant geometric characteristics within the pattern and can be used to help determine whether the pattern has been printed correctly. Extracting these feature points is for further analysis of their distribution within the packaging area.

[0150] Specifically, step 10382 includes steps A1 to A6:

[0151] Step A1: Obtain the standard feature matrix corresponding to the preset feature points;

[0152] In this step, a predefined standard feature matrix needs to be obtained. This is a feature matrix representing an ideal pattern (such as a printed pattern, logo, or QR code). The standard feature matrix can be a local image fragment containing information about the logo, pattern, or other feature points, and serves as a benchmark for comparison with the image of the current encapsulation area.

[0153] Step A2: In the current encapsulation region, extract the pixel matrix surrounding each current pixel as the center; wherein the pixel matrix has the same matrix size as the standard feature matrix;

[0154] Within the current encapsulation area, each time a pixel is selected as the "center pixel," the pixels surrounding that pixel are extracted to form a matrix. The size of this extracted matrix is ​​consistent with the standard feature matrix; that is, the size of the pixel matrix to be extracted perfectly matches the standard feature matrix.

[0155] For example, if the standard feature matrix is ​​a 3x3 region, then within the current encapsulation region, a 3x3 pixel matrix will be extracted, centered at each pixel. These matrices are used for comparison with the standard feature matrix.

[0156] This specifies that the size of the extracted pixel matrix is ​​consistent with the size of the standard feature matrix. This ensures that the structure and size of the two matrices are perfectly matched during comparison, thus enabling effective similarity calculation.

[0157] Step A3: Calculate the second difference between the pixel value of the current pixel and the pixel values ​​of other pixels in the pixel matrix;

[0158] The core of this step lies in calculating the differences between the current pixel and its surrounding pixels. For the currently selected "center pixel," the differences between its pixel value and the values ​​of other pixels in the surrounding pixel matrix need to be calculated. These differences are called "second differences," and they represent the differences in brightness, color, and other information between the center pixel and its neighboring pixels.

[0159] In this way, the differences in pixel values ​​within a local area can be quantified for subsequent comparison and analysis.

[0160] Step A4: Replace the pixel values ​​of other pixels in the pixel matrix with the second difference to obtain the current feature matrix;

[0161] After calculating the "second difference," these differences are used to replace the original pixel matrix, generating a new matrix called the "current feature matrix." Each pixel's value is replaced by its difference from the center pixel, thus creating a matrix that reflects the differences within a local region relative to the center pixel. This process allows for further analysis of the distribution of these differences across the entire image.

[0162] Understandably, the process of generating the standard feature matrix is ​​similar to the process of generating the current feature matrix.

[0163] Step A5: Calculate the similarity between each current feature matrix and the standard feature matrix;

[0164] After obtaining the current feature matrix, the system needs to compare it with the standard feature matrix. Calculating the similarity score is to assess the degree of matching between the current feature matrix and the standard feature matrix. This similarity calculation helps the system determine whether the current matrix is ​​very close to the standard matrix, thereby determining whether it conforms to the expected printing pattern or feature design.

[0165] Step A6: Take the current pixel point corresponding to the current feature matrix with the maximum similarity as the preset feature point.

[0166] After comparing all current feature matrices with the standard feature matrix, multiple similarity values ​​are obtained. The "center pixel" of the current feature matrix corresponding to the maximum similarity value is the preset feature point to be extracted. This is because the maximum similarity indicates that a certain pixel in the current encapsulation region has the highest matching degree with the standard pattern, which usually means that this position is a key feature point.

[0167] In the embodiments corresponding to steps A1 to A6, preset feature points are located by extracting the pixel matrix around each pixel within the current encapsulation area and comparing these matrices with a standard feature matrix to calculate similarity. By calculating the difference between each pixel and its surrounding pixels to generate a new feature matrix, and then comparing the similarity with the standard matrix, preset feature points consistent with the standard pattern in the current encapsulation area can be identified efficiently and accurately. This method, through meticulous local pixel comparison, helps to accurately identify key information in the encapsulation area and ensures the normality of the encapsulation pattern.

[0168] Step 10383: Calculate the first distance between each preset feature point and the four diagonal points;

[0169] Next, the system calculates the distances from the extracted preset feature points to the four diagonal points of the packaging area. These distances are key parameters for evaluating the relationship between the geometry of the packaging area and the distribution of the printed pattern. Calculating these distances helps determine the relative positions of the feature points and the packaging area, thereby confirming the correctness of the printed pattern.

[0170] Step 10384: Obtain multiple second distances corresponding to the standard packaging area; the multiple second distances refer to the distances between preset feature points and four diagonal points in the standard packaging area;

[0171] Within the standard package area, there are the same geometric features and printed patterns as the current package area. The standard package area provides a correct, defect-free package reference. Within the standard package area, there is also a set of preset distances between feature points and four diagonal points (i.e., "second distances"). These second distances represent the ideal standard positional relationship between feature points and diagonal points.

[0172] The second distance is defined as the actual distance between a feature point in the standard package area and its four diagonal points. These standard distances are used as the basis for comparison with the distances in the current package area.

[0173] Step 10385: Calculate the first difference between the first distance and the second distance corresponding to the same point;

[0174] In this step, the system compares the first distance and the second distance between the current packaged region and the standard packaged region at the same points (i.e., the same combination of feature points and corner points). Specifically, the system calculates the actual distance difference between each pair of feature points and corner points to obtain the "first difference". The goal of this step is to quantify the geometric differences between the current packaged region and the standard packaged region to see if these differences exceed a preset tolerance range.

[0175] Step 10386: If the first difference is less than the third threshold, then it is confirmed that there is no printing pattern abnormality in the current packaging area;

[0176] If all calculated first differences are less than a set threshold (i.e., the "third threshold"), it indicates that the geometry and printed pattern of the current packaging area are very close to the expected standard packaging area, and there are no obvious anomalies. This suggests that the printed pattern is without deviation and the printed portion within the packaging area is normal.

[0177] The third threshold is a key parameter used to determine whether the pattern deviation between the current packaging area and the standard packaging area exceeds the acceptable range. Assuming the packaging area is 10mm long, and the standard distance from the preset feature points of the standard packaging area to the four diagonal points is 3mm, and considering production errors, the deviation of the packaging area is typically between 0.1mm and 0.3mm, the third threshold can be set to 0.2mm. This means that when the pattern deviation is greater than 0.2mm, an anomaly is considered to exist.

[0178] Step 10387: If the first difference is less than the third threshold, it is confirmed that there is an abnormal printing pattern in the current packaging area.

[0179] If, during the comparison process, at least one of the first differences is found to be greater than the third threshold, then it can be determined that the printed pattern in the current packaging area is abnormal. That is, the distribution of the printed pattern differs significantly from that of the standard packaging area, which may be due to printing offset, scaling issues, or other processing problems causing the pattern abnormality.

[0180] In the embodiments corresponding to steps 10381 to 10387, the normality of the printed pattern is detected by comparing the geometric relationships between the feature points and diagonal points of the current packaging area and the standard packaging area. This method relies on precise geometric comparison and distance calculation, which can accurately identify anomalies in the printed pattern. In this way, printing defects can be detected in a timely manner, ensuring that the packaging quality meets standard requirements.

[0181] Step 104: When it is determined that there are no defects in the current package area, determine whether pin defects occur in multiple current pin areas based on the current package area; the pin defects include pin offset defects, pin unevenness defects, and pin missing defects;

[0182] Once the packaged area is confirmed to be defect-free, the system continues to inspect the chip's pin area. Pin area inspection primarily determines whether the pins have defects, including:

[0183] Pin misalignment defect: The pin position is inaccurate and deviates from its original position;

[0184] Uneven pin surface defect: Uneven pin surface may lead to poor electrical contact;

[0185] Pin missing defect: Some pins are completely missing or broken, which may cause the chip to malfunction.

[0186] Specifically, step 104 includes steps 1041 to 1045:

[0187] Step 1041: When it is determined that there are no defects in the current packaging area, obtain a preset first packaging pin ratio relationship and a second packaging pin ratio relationship; the first packaging pin ratio relationship refers to the ratio between the first length of the standard packaging area and the second length of a single standard pin area, and the second packaging pin ratio relationship refers to the ratio between the first length of the standard packaging area and the first width of a single standard pin area.

[0188] Before starting pin defect analysis, first confirm that there are no other defects in the current package area. Then, the system needs to obtain two preset ratio relationships:

[0189] The first package pin ratio refers to the ratio between the first length of the standard package area and the second length of a single standard pin area. This ratio reflects the proportional relationship in length between the standard package area and the pin area, helping to estimate the pin area length from the current package area length in subsequent analysis.

[0190] The second package pin ratio refers to the ratio between the first length of the standard package area and the first width of a single standard pin area. This ratio provides the proportional relationship between the length of the standard package area and the width of the pin area, helping to estimate the width of the pin area from the length of the current package area.

[0191] This step clarifies the definitions of the first and second proportional relationships. The first length of the standard package area is a critical dimension, representing the total length of the package area or other major geometric dimensions. Using these two proportional relationships, the dimensions of the corresponding pin area can be calculated based on the current package area dimensions.

[0192] Step 1042: Obtain the third length of the current encapsulation region;

[0193] Within the current package area, a third length needs to be obtained, which typically refers to a specific length of the current package area. For example, this could be the total length of the current package area, the long side, or other lengths that need to be analyzed. This length, as an input parameter, along with a scaling factor, helps calculate the relevant dimensions of the pin area.

[0194] Step 1043: Calculate the fourth length corresponding to the pin region based on the first package pin ratio and the third length;

[0195] By using the first package pin ratio and the third length, the fourth length of the pin region can be calculated. The core of this step is using the ratio to determine the corresponding size of the pin region within the current package area size. That is, by multiplying the third length by the first ratio, the fourth length of the pin region can be obtained.

[0196] Step 1044: Calculate the third width corresponding to the pin area based on the pin ratio of the second package and the third length;

[0197] Similar to the calculation of the fourth length, the third width of the pin region is calculated using the second package pin ratio and the third length. The width of the pin region is obtained by multiplying the third length by the second ratio. These two calculations (the fourth length and the third width) are crucial steps in identifying whether defects exist in the pin region.

[0198] Step 1045: Determine whether pin defects occur in the plurality of current pin regions based on the third length, the fourth length, and the third width.

[0199] Ultimately, the system checks the geometry of the current pin region against the expected standards based on the third length, fourth length, and third width. If these calculated values ​​match the corresponding dimensions of the standard package area, the current pin region is considered defect-free. If they do not match, the system considers the current pin region defective. Here, a defect might be defined as dimensional inconsistency, abnormal shape, etc.

[0200] In the embodiments corresponding to steps 1041 to 1045, the relevant dimensions (such as length and width) of the pin area are calculated using several proportional relationships and the geometric dimensions of the current package area. By comparing the deviations of these calculated dimensions with preset standard dimensions, defects in the pins of the current package area can be detected. Specifically, the system first calculates the expected dimensions of the pin area and then compares them with the actual dimensions to determine if any anomalies exist. This method accurately determines pin defects in the package area through geometric calculations, avoiding errors from manual inspection.

[0201] Specifically, step 1045 includes steps 10451 to 10458:

[0202] Step 10451: Obtain the fifth length corresponding to each of the multiple current pin regions;

[0203] At this stage, the system needs to obtain the fifth length of the current pin region. The fifth length is a metric related to the geometry of the current pin region, and it may be related to the pin's length, position, or other standard parameters. By obtaining the fifth lengths of multiple pin regions, differences between them and potential defects can be further analyzed.

[0204] Step 10452: Calculate the third difference between the fourth length and the fifth length;

[0205] Next, the difference between the fourth and fifth lengths needs to be calculated, referred to as the "third difference." This difference reflects the discrepancy between the expected pin area length (calculated from the fourth length) and the actual pin area length (calculated from the fifth length). A smaller difference indicates that the pin area dimensions are more consistent, while a larger difference may indicate a dimensional anomaly or defect.

[0206] Step 10453: If the third difference is less than the fourth threshold, then obtain the second width of multiple sampling points in each current pin region;

[0207] If the calculated third difference is less than the preset fourth threshold, the system further analyzes the current pin region. At this point, it is necessary to obtain the second width of multiple sampling points within the current pin region. These sampling points represent several key locations within the region, used to more accurately analyze the width characteristics of the pin region.

[0208] The fourth threshold represents a tolerance or standard tolerance range, derived from experimental data or empirical standards. It determines the permissible range of difference between the fourth and fifth lengths. Specific data support may be based on empirical data, assuming, for example, 0.1 mm or 0.2 mm. If the difference is less than this threshold, it indicates that the dimensional error of the pin area is within acceptable limits, and subsequent testing can continue.

[0209] Step 10454: Calculate the fourth difference between the third width and the plurality of second widths;

[0210] Next, the difference between the third width (the width calculated for the current region) and multiple second widths (the widths corresponding to the sampling points) needs to be calculated. This "fourth difference" reflects the difference between the width of the current pin region and the width of the sampling points. By calculating the fourth difference across multiple sampling points, it can help to further determine whether the shape of the pin region is normal.

[0211] Step 10455: If the fifth difference between multiple second widths in the same current pin region is less than the fourth threshold, and multiple fourth differences are less than the fifth threshold, then it is confirmed that there is no defect in the current pin region.

[0212] At this stage, the stability of the pin region needs to be further verified. If the difference between the second widths of multiple sampling points in the same pin region (i.e., the fifth difference) is less than the fourth threshold, and the fourth difference of these sampling points is less than the fifth threshold, it indicates that the size of the pin region is uniform and stable within a certain range, meets the standard requirements, and it can be confirmed that there are no defects in the region.

[0213] The fifth threshold represents the tolerance range for pin area size variations. These thresholds are derived from manufacturing process standards, experimental data, or industry specifications. For example, 0.1 mm might be a common length tolerance range, and 0.05 mm might be the tolerance range for width variations, but the actual values ​​will depend on the specific manufacturing process and quality control standards.

[0214] Step 10456: If the third difference is not less than the fourth threshold, then it is determined that there is a pin offset defect in the current pin area;

[0215] If the third difference is greater than or equal to the fourth threshold, it indicates a significant deviation in the size of the pin area, possibly due to pin position offset. In this case, the system will determine that the pin area has a pin offset defect.

[0216] Step 10457: If the fifth difference is not less than the fourth threshold or the fourth difference is not less than the fifth threshold, then it is determined that there is a pin unevenness defect in the current pin area;

[0217] If, within the same pin area, the fifth difference between the second widths fails to consistently fall below the fourth threshold, or if the fourth difference is not uniformly less than the fifth threshold, it indicates a potential unevenness defect in the pin area. This defect suggests that the pin shape or surface does not meet expected standards, potentially affecting the overall quality and functionality of the package.

[0218] Step 10458: If the number of pins in the current pin area is less than the preset number, then it is confirmed that there is a pin missing defect in the current pin area.

[0219] Finally, if the number of pins detected is less than the predetermined preset number, it indicates a pin missing defect in the current package area. This could be because some pins were not correctly formed, were not identified, or were omitted, resulting in an insufficient number of pins.

[0220] In the embodiments corresponding to steps 10451 to 10458, the current pin region is analyzed layer by layer by comparing multiple threshold differences to determine whether different types of defects exist. Specifically: pin offset defects are determined by comparing the difference between the fourth and fifth lengths; pin unevenness defects are determined by analyzing the width differences of multiple sampling points; and pin missing defects are confirmed by checking whether the number of pin regions meets preset requirements. The determination of each defect is finely controlled through comparisons of multiple differences and thresholds, thereby improving detection accuracy and precision.

[0221] Step 105: When it is determined that there is no defect in the current pin area, determine whether there is a positional distribution defect between the package area and the pin area based on the current package area and multiple current pin areas.

[0222] If both the package and pin areas are free of defects, the system also needs to check for any abnormalities in the relative positions between them. For example, the mating positions of the chip's package and pin areas must meet specific geometric requirements to ensure that the pins can connect properly to external circuits. Any positional distribution defects between the package and pins (such as misalignment between the package and pin areas, which could lead to electrical connection problems or poor pin contact) also need to be detected.

[0223] Specifically, step 105 includes steps 1051 to 1059:

[0224] Step 1051: Obtain the included angle fluctuation range corresponding to the standard reference image; the included angle fluctuation range is used to describe the distribution relationship between multiple pin regions;

[0225] First, the system needs to acquire a standard reference image. This reference image represents a normal pin distribution. The angular fluctuation range is extracted from this image and describes the distribution relationship between the relative positions of the pin regions. This fluctuation range is used in subsequent steps to determine whether the current pin region deviates from the standard position distribution. It can be understood as the range of angular differences between pins in the standard distribution.

[0226] Specifically, step 1051 includes steps 10511 to 10519:

[0227] Step 10511: Obtain a standard reference image and extract the standard package area and multiple standard pin areas from the standard reference image;

[0228] The first step in this step is to obtain a standard reference image, which represents the ideal package and pin layout. This image includes a standard package area and multiple standard pin areas.

[0229] Extract the standard package region (i.e., the boundary between the pins and the package in the image) and multiple standard pin regions (i.e., the location of each pin) from this image. These regions will be used for subsequent calculations.

[0230] Step 10512: Extract the third center of the standard package region and extract the fourth center of the multiple standard pin regions;

[0231] The third center of the standard package area is the geometric center or centroid of the package area in the standard reference image, which is usually the center of symmetry of the package.

[0232] The fourth center of multiple standard pin regions refers to the geometric center or centroid of each pin region, which is the center point of each pin.

[0233] Step 10513: Establish a second coordinate system with the third center as the origin;

[0234] In this stage, a second coordinate system is established, with the third center of the standard encapsulation region as the origin of the new coordinate system. All subsequent calculations will then be based on this new coordinate system.

[0235] Step 10514: In the second coordinate system, obtain the second coordinate positions of the multiple fourth centers;

[0236] After establishing the second coordinate system, the coordinates of the fourth center (i.e., the center of each pin area) of each standard pin area will be obtained and transformed into the second coordinate system to obtain the coordinate position of each pin area.

[0237] Step 10515: Extract multiple second line segments formed by multiple second coordinate positions and the third center;

[0238] By connecting the third center of the standard package area to the fourth center of each standard pin area, a series of line segments are obtained. These line segments represent the connection from the center of the package area to the center of each pin area, and are called the second line segments.

[0239] Step 10516: Among the multiple second line segments, extract the second minimum distance line segment corresponding to each second line segment; wherein, the second minimum distance line segment refers to the line segment with the smallest distance from the second line segment;

[0240] For each second line segment, it is necessary to further analyze the shortest distance between this line segment and other line segments. By calculation, the line segment with the minimum distance between each second line segment and other line segments is extracted, which is called the second minimum distance line segment.

[0241] This step is mainly to determine the geometric relationship between each line segment and other line segments, and to find their relative positions.

[0242] Step 10517: Calculate the second included angle between the second line segment and the second minimum distance line segment respectively;

[0243] Calculate the angle between each second line segment and its corresponding second minimum distance line segment. These angles reflect the angular distribution of each pin area within the package area.

[0244] This included angle can be used to describe the stability and symmetry of the pin distribution within a standard package area.

[0245] Step 10518: Calculate the first fluctuation range formed by multiple second included angles;

[0246] After calculating the angle between each second line segment and its corresponding second minimum distance line segment, it is necessary to count all the calculated angles to obtain an angle fluctuation range (i.e., the fluctuation range of all angles).

[0247] The first fluctuation range is a statistical result of these included angles, representing the range between the maximum and minimum values ​​of all included angles. It is used to describe the angular variation of the standard pin area distribution.

[0248] Step 10519: Based on a preset redundancy coefficient, the first fluctuation range is expanded to obtain the included angle fluctuation range.

[0249] In this step, a redundancy coefficient is introduced to extend the first fluctuation range. The redundancy coefficient is used to adjust the tolerance range of the included angle fluctuation so as to cope with small changes that may occur in actual production processes.

[0250] This extended processing yields an angle fluctuation range, which represents the tolerance range of the angle between the standard pin area and the package area, and provides a standard reference for subsequent defect detection.

[0251] In the embodiments corresponding to steps 10511 to 10519, an angle fluctuation range is derived through the following steps to describe the stability of the pin area distribution: extracting the package area and pin area from a standard reference image; establishing a coordinate system with the center of the package area as the origin; calculating the geometric relationship between the centers of the package area and the pin area; calculating the angle between line segments and statistically analyzing the angle fluctuation range; and expanding this fluctuation range based on a redundancy coefficient to obtain a standard angle fluctuation range. This method helps to set an ideal pin distribution range for comparison with the actual pin distribution in testing to determine whether defects exist.

[0252] Step 1052: Extract the first center of the current package region, and extract the second centers of multiple current pin regions;

[0253] At this stage, two center points need to be extracted from the current package area and pin area:

[0254] The first center is the geometric center of the current encapsulation region.

[0255] The second center is the geometric center of each pin in the multiple pin regions, that is, the center point of each pin region.

[0256] Step 1053: Establish a first coordinate system with the first center as the origin;

[0257] In this stage, a first coordinate system is established with the first center of the encapsulated region as the origin. This coordinate system provides a unified reference framework for all subsequent calculations.

[0258] Step 1054: In the first coordinate system, obtain the first coordinate positions of multiple second centers;

[0259] Next, the second centers of the multiple pin regions (i.e., the center point of each pin) are mapped onto the first coordinate system to obtain their coordinate positions. These coordinate values ​​represent the specific position of each pin center relative to the center of the package region.

[0260] Step 1055: Extract multiple first line segments formed by multiple first coordinate positions and the first center;

[0261] Based on the coordinates obtained in step 1054, the center point of each pin is connected to the first center of the package area to form several first line segments. Each first line segment represents a line connecting the center of the package area to the center of a certain pin area.

[0262] Step 1056: Among the multiple first line segments, extract the first minimum distance line segment corresponding to each first line segment; wherein, the first minimum distance line segment refers to the line segment with the smallest distance from the first line segment;

[0263] Within these line segments, further analysis of the relative positions of each segment is needed, particularly the distances between each segment and other segments. To do this, the minimum distance between each first segment and other segments must be calculated, and the shortest segment is extracted; this is called the first minimum distance segment. This represents the minimum distance from one pin to other pins, and it is crucial for calculating the included angle.

[0264] Step 1057: Calculate the first included angle between the first line segment and the first minimum distance line segment respectively;

[0265] Next, the angle between each first line segment and its corresponding first minimum distance line segment is calculated. By calculating the angle between each pair of line segments, the angular relationship between the pin areas can be obtained. This angle reflects the positional distribution of the pins within the package area.

[0266] Step 1058: If multiple first included angles are all within the included angle fluctuation range, then it is determined that there is no positional distribution defect between the package area and the pin area;

[0267] At this stage, the system checks whether all calculated first included angles fall within the standard included angle fluctuation range. If each included angle is within this range, it indicates that the pin area distribution meets the standard requirements and there are no positional distribution defects.

[0268] Step 1059: If multiple first included angles are not uniformly within the included angle fluctuation range, then it is determined that there is a positional distribution defect between the package area and the pin area.

[0269] If a portion or most of the calculated included angle is outside the included angle fluctuation range, it indicates that the positional distribution of the pin area does not conform to the standard and there is a deviation. In this case, the system will determine that there is a positional distribution defect between the package area and the pin area.

[0270] In the embodiments corresponding to steps 1051 to 1059, the positional distribution between the package area and the pin area is analyzed in detail through the following geometric steps: a standard reference image is used to set the range of angle fluctuations; the coordinates of the center of the package area and the center of the pin area are extracted; the relative positions between the pins are calculated in the coordinate system to form connecting lines; the angles between these connecting lines are calculated; and it is determined whether these angles conform to the standard range, thereby determining whether there are positional distribution defects. This method determines whether the pin distribution is normal through geometric analysis, and is a precise defect detection method.

[0271] In the embodiments corresponding to steps 101 to 105, by extracting the standard packaging area and comparing it with the current packaging area, packaging defects can be effectively identified. Packaging defects include missing packaging, damaged packaging, and abnormal printed patterns, which directly affect the chip's functionality and reliability. Traditional detection methods often struggle to accurately distinguish between different types of packaging defects, while this invention, through comparative analysis of the standard packaging area and the current packaging area, can accurately locate packaging defects, avoiding missed and false detections. Furthermore, when the packaging area is determined to be defect-free, this invention can identify various pin defects, such as pin offset, pin unevenness, and pin missing, through comprehensive analysis of the current packaging area and the current pin area. These pin defects often directly affect the chip's electrical performance, while traditional detection methods typically only detect some defect types. This invention's method can comprehensively detect and accurately identify various defects in the pin area, improving the comprehensiveness and accuracy of detection. In addition to defect detection in the packaging and pin areas, this invention can further determine the existence of positional distribution defects by analyzing the positional distribution between the packaging and pin areas. Traditional inspection methods typically focus only on localized defects, neglecting the relative positional relationship between the package and pin areas. This can lead to some minor but serious defects going undetected. By detecting positional defects, the accuracy of overall chip quality inspection can be improved, reducing chip failures caused by positional errors. In summary, the chip defect detection method provided by this invention offers a comprehensive and accurate detection solution for package defects, pin defects, and positional defects, greatly improving the quality control level of chip production and overcoming many shortcomings of traditional inspection methods.

[0272] like Figure 2 This invention provides a chip defect detection device; please refer to [link / reference]. Figure 2 , Figure 2 A schematic diagram of a chip defect detection device provided by the present invention is shown, as follows. Figure 2 The chip defect detection device shown includes:

[0273] Acquisition unit 21 is used to acquire the image to be recognized collected by the vision module;

[0274] Extraction unit 22 is used to extract the chip image region in the image to be identified, and to extract the current package region and multiple current pin regions in the chip image region;

[0275] The first judgment unit 23 is used to obtain a standard packaging area and, based on the standard packaging area and the current packaging area, to confirm whether there is a packaging defect in the current packaging area; the packaging defect includes missing packaging, damaged packaging, or abnormal printing pattern;

[0276] The second determination unit 24 is used to determine whether pin defects occur in multiple current pin areas based on the current packaging area when it is determined that there are no defects in the current packaging area; the pin defects include pin offset defects, pin unevenness defects and pin missing defects;

[0277] The third judgment unit 25 is used to determine whether there is a positional distribution defect between the package area and the pin area based on the current package area and multiple current pin areas when it is determined that there is no defect in the current pin area.

[0278] This invention provides a chip defect detection device that effectively identifies packaging defects by extracting a standard packaging area and comparing it with the current packaging area. Packaging defects include missing packages, damaged packages, and abnormal printed patterns, which directly affect the chip's functionality and reliability. Traditional detection methods often struggle to accurately distinguish between different types of packaging defects, while this invention, through comparative analysis of the standard packaging area and the current packaging area, can precisely locate packaging defects, avoiding missed and false detections. Furthermore, assuming the packaging area is defect-free, this invention can identify various pin defects, such as pin misalignment, pin unevenness, and missing pins, through comprehensive analysis of the current packaging area and the current pin area. These pin defects often directly affect the chip's electrical performance, while traditional detection methods typically only detect some defect types. This invention's method can comprehensively detect and accurately identify various defects in the pin area, improving the comprehensiveness and accuracy of detection. In addition to defect detection in the packaging and pin areas, this invention can further determine the existence of positional distribution defects by analyzing the positional distribution between the packaging and pin areas. Traditional inspection methods typically focus only on localized defects, neglecting the relative positional relationship between the package and pin areas. This can lead to some minor but serious defects going undetected. By detecting positional defects, the accuracy of overall chip quality inspection can be improved, reducing chip failures caused by positional errors. In summary, the chip defect detection method provided by this invention offers a comprehensive and accurate detection solution for package defects, pin defects, and positional defects, greatly improving the quality control level of chip production and overcoming many shortcomings of traditional inspection methods.

[0279] Figure 3 This is a schematic diagram of a terminal device provided in an embodiment of the present invention. Figure 3 As shown, a terminal device 3 in this embodiment includes: a processor 30, a memory 31, and a computer program 32 stored in the memory 31 and executable on the processor 30, such as a chip defect detection program. When the processor 30 executes the computer program 32, it implements the steps in the various chip defect detection method embodiments described above, for example... Figure 1 Steps 101 to 105 are shown. Alternatively, when the processor 30 executes the computer program 32, it implements the functions of each unit in the above-described device embodiments, for example... Figure 2 The function of the unit shown.

[0280] For example, the computer program 32 can be divided into one or more units, which are stored in the memory 31 and executed by the processor 30 to complete the present invention. The one or more units can be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program 32 in the terminal device 3. For example, the specific functions of each unit of the computer program 32 can be divided as follows:

[0281] The acquisition unit is used to acquire the image to be recognized collected by the vision module;

[0282] The extraction unit is used to extract the chip image region in the image to be identified, and to extract the current package region and multiple current pin regions in the chip image region;

[0283] The first judgment unit is used to obtain a standard packaging area and, based on the standard packaging area and the current packaging area, to determine whether there is a packaging defect in the current packaging area; the packaging defect includes missing packaging, damaged packaging, or abnormal printing pattern.

[0284] The second determination unit is used to determine whether pin defects occur in multiple current pin areas based on the current packaging area when it is determined that there are no defects in the current packaging area; the pin defects include pin offset defects, pin unevenness defects and pin missing defects;

[0285] The third judgment unit is used to determine whether there is a positional distribution defect between the package area and the pin area based on the current package area and multiple current pin areas when it is determined that there is no defect in the current pin area.

[0286] The terminal device includes, but is not limited to, a processor 30 and a memory 31. Those skilled in the art will understand that... Figure 3 This is merely an example of a terminal device 3 and does not constitute a limitation on a terminal device 3. It may include more or fewer components than shown, or combine certain components, or different components. For example, the terminal device may also include input / output devices, network access devices, buses, etc.

[0287] The processor 30 can be a Central Processing Unit (CPU), or other 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. The general-purpose processor can be a microprocessor or any conventional processor.

[0288] The memory 31 can be an internal storage unit of the terminal device 3, such as a hard disk or memory of the terminal device 3. The memory 31 can also be an external storage device of the terminal device 3, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the terminal device 3. Furthermore, the memory 31 can include both internal and external storage units of the terminal device 3. The memory 31 is used to store the computer program and other programs and data required by the roaming control device. The memory 31 can also be used to temporarily store data that has been output or will be output.

[0289] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0290] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of the present invention. For details on their specific functions and technical effects, please refer to the method embodiments section, which will not be repeated here.

[0291] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this invention. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0292] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps described in the various method embodiments above.

[0293] This invention provides a computer program product that, when run on a mobile terminal, enables the mobile terminal to implement the steps described in the above-described method embodiments.

[0294] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying the computer program code to a photographing device / terminal device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0295] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0296] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0297] In the embodiments provided by this invention, it should be understood that the disclosed apparatus / network devices and methods can be implemented in other ways. For example, the apparatus / network device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another 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 units may be electrical, mechanical, or other forms.

[0298] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units.

[0299] It should be understood that, when used in this specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0300] It should also be understood that the term “and / or” as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0301] As used in this specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if [the described condition or event] is detected" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once [the described condition or event] is detected," or "in response to detection of [the described condition or event]."

[0302] Furthermore, in the description of this invention and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0303] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of the invention include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0304] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for detecting chip defects, characterized in that, The chip defect detection method includes: Acquire the image to be recognized captured by the vision module; Extract the chip image region from the image to be identified, and extract the current package region and multiple current pin regions from the chip image region according to the first color distribution range corresponding to the base and the second color distribution range corresponding to the pin; A standard encapsulation area is obtained. Based on the pixel difference between the same pixel position in the standard encapsulation area and the current encapsulation area, and the distribution relationship between the four diagonal points of the current encapsulation area and the preset feature points, it is confirmed whether there is an encapsulation defect in the current encapsulation area. The encapsulation defect includes encapsulation missing, encapsulation damaged, or abnormal printing pattern. When it is determined that there are no defects in the current package area, based on the first package pin ratio relationship, the second package pin ratio relationship, the current package area length, the current package area width, the current pin area length, and the current pin area width, it is determined whether pin defects occur in multiple current pin areas. The pin defects include pin offset defects, pin unevenness defects, and pin missing defects. When it is determined that there is no defect in the current pin area, the presence of a positional distribution defect between the package area and the pin area is determined based on whether the multiple first included angles obtained from the first center of the current package area and the multiple second centers of the current pin areas are within the included angle fluctuation range. The included angle fluctuation range is used to describe the distribution relationship between the multiple pin areas.

2. The chip defect detection method as described in claim 1, characterized in that, The step of extracting the chip image region from the image to be identified, and extracting the current package region and multiple current pin regions from the chip image region according to the first color distribution range corresponding to the base and the second color distribution range corresponding to the pin, includes: Obtain the first color distribution range corresponding to the base and the second color distribution range corresponding to the pin; Obtain multiple first pixels corresponding to the first color distribution range in the image to be identified; The chip image region is obtained by removing multiple consecutive first pixels from the image to be identified. Obtain multiple second pixel points corresponding to the second color distribution range in the chip image region; Obtain multiple image regions consisting of multiple consecutive second pixels, and use these multiple image regions as multiple current pin regions; The current packaging area is obtained by removing multiple consecutive second pixels in the chip image area.

3. The chip defect detection method as described in claim 1, characterized in that, The step of obtaining the standard encapsulation region, and confirming whether there is an encapsulation defect in the current encapsulation region based on the pixel difference between the same pixel positions in the standard encapsulation region and the current encapsulation region, and the distribution relationship between the four diagonal points of the current encapsulation region and the preset feature points, includes: Obtain the distribution range of the third color corresponding to the encapsulation material; In the current encapsulation area, extract multiple third pixel points corresponding to the third color distribution range; The image region composed of multiple third pixels is used as the first encapsulation material region; Obtain the second encapsulation material region within the standard encapsulation region; Align the first encapsulation material region with the second encapsulation material region, and calculate the pixel difference between each corresponding pixel position; If the pixel difference between multiple consecutive pixel positions is greater than the first threshold, and the number of such consecutive pixel positions exceeds the second threshold, then it is confirmed that the current encapsulation area has encapsulation defects or encapsulation damage. If the pixel difference corresponding to multiple consecutive pixel positions is not greater than the first threshold or the number of multiple consecutive pixel positions does not exceed the second threshold, then it is confirmed that there is no missing or damaged encapsulation in the current encapsulation area. Based on the distribution relationship between the four diagonal points corresponding to the current packaging area and the preset feature points, it is confirmed whether there is any printing pattern abnormality in the current packaging area.

4. The chip defect detection method as described in claim 3, characterized in that, The step of confirming whether there is a printing pattern abnormality in the current packaging area based on the distribution relationship between the four diagonal points corresponding to the current packaging area and the preset feature points includes: Obtain the four diagonal points corresponding to the current encapsulation area; Extract two preset feature points corresponding to the printed pattern in the current packaging area; Calculate the first distance between each preset feature point and the four diagonal points; Obtain multiple second distances corresponding to the standard encapsulation region; the multiple second distances refer to the distances between preset feature points and four diagonal points in the standard encapsulation region; Calculate the first difference between the first distance and the second distance corresponding to the same point; If the first difference is less than the third threshold, it is confirmed that there is no printing pattern abnormality in the current packaging area; If the first difference is not less than the third threshold, it is confirmed that there is an abnormal printing pattern in the current packaging area.

5. The chip defect detection method as described in claim 4, characterized in that, The step of extracting two preset feature points corresponding to the printed pattern in the current packaging area includes: Obtain the standard feature matrix corresponding to the preset feature points; In the current encapsulation region, with each current pixel as the center, extract the pixel matrix surrounding the current pixel; wherein, the pixel matrix has the same matrix size as the standard feature matrix; Calculate the second difference between the pixel value of the current pixel and the pixel values ​​of other pixels in the pixel matrix; The pixel values ​​of other pixels in the pixel matrix are replaced with the second difference to obtain the current feature matrix; Calculate the similarity between each current feature matrix and the standard feature matrix; The current pixel corresponding to the current feature matrix with the highest similarity is taken as the preset feature point.

6. The chip defect detection method as described in claim 1, characterized in that, When it is determined that there are no defects in the current package area, the step of determining whether pin defects occur in multiple current pin areas based on the first package pin ratio relationship, the second package pin ratio relationship, the current package area length, the current package area width, the current pin area length, and the current pin area width, wherein the pin defects include pin offset defects, pin unevenness defects, and pin missing defects, includes: When it is determined that there are no defects in the current packaging area, a preset first packaging pin ratio and a preset second packaging pin ratio are obtained; the first packaging pin ratio refers to the ratio between the first length of the standard packaging area and the second length of a single standard pin area, and the second packaging pin ratio refers to the ratio between the first length of the standard packaging area and the first width of a single standard pin area. Obtain the third length of the current encapsulation region; Based on the first package pin ratio and the third length, calculate the fourth length corresponding to the pin region; Calculate the third width corresponding to the pin area based on the second package pin ratio and the third length; Based on the third length, the fourth length, and the third width, it is determined whether pin defects occur in multiple current pin regions.

7. The chip defect detection method as described in claim 6, characterized in that, The step of determining whether pin defects occur in the plurality of current pin regions based on the third length, the fourth length, and the third width includes: Obtain the fifth length corresponding to each of the multiple current pin regions; Calculate the third difference between the fourth length and the fifth length; If the third difference is less than the fourth threshold, then the second width of multiple sampling points in each current pin region is obtained; Calculate a fourth difference between the third width and the plurality of second widths; If the fifth difference between multiple second widths in the same current pin region is less than the fourth threshold, and multiple fourth differences are less than the fifth threshold, then it is confirmed that there is no defect in the current pin region. If the third difference is not less than the fourth threshold, then it is determined that there is a pin offset defect in the current pin area; If the fifth difference is not less than the fourth threshold or the fourth difference is not less than the fifth threshold, then it is determined that there is a pin unevenness defect in the current pin area. If the number of pins in the current pin area is less than the preset number, it is confirmed that there is a pin missing defect in the current pin area.

8. The chip defect detection method as described in claim 1, characterized in that, The step of determining whether there is a positional distribution defect between the package area and the pin area when it is determined that there is no defect in the current pin area, based on whether the multiple first included angles obtained from the first center of the current package area and the multiple second centers of the current pin areas are within the included angle fluctuation range, and the included angle fluctuation range is used to describe the distribution relationship between the multiple pin areas, includes: Obtain the included angle fluctuation range corresponding to the standard reference image; the included angle fluctuation range is used to describe the distribution relationship between multiple pin regions; Extract the first center of the current package region, and extract the second centers of multiple current pin regions; Establish a first coordinate system with the first center as the origin; In the first coordinate system, obtain the first coordinate positions of multiple second centers; Extract multiple first line segments formed by multiple first coordinate positions and the first center; Among multiple first line segments, the first minimum distance line segment corresponding to each first line segment is extracted; wherein, the first minimum distance line segment is the line segment with the smallest distance from the first line segment. Calculate the first included angle between the first line segment and the first line segment with the minimum distance; If multiple first included angles are all within the included angle fluctuation range, then it is determined that there is no positional distribution defect between the package area and the pin area; If multiple first included angles are not uniformly within the included angle fluctuation range, then it is determined that there is a positional distribution defect between the package area and the pin area.

9. The chip defect detection method as described in claim 8, characterized in that, The step of obtaining the included angle fluctuation range corresponding to the standard reference image includes: Obtain a standard reference image, and extract the standard package area and multiple standard pin areas from the standard reference image; Extract the third center of the standard package region and extract the fourth center of multiple standard pin regions; A second coordinate system is established with the third center as the origin. In the second coordinate system, the second coordinate positions of the multiple fourth centers are obtained; Extract multiple second line segments formed by multiple second coordinate positions and the third center; Among multiple second line segments, extract the second minimum distance line segment corresponding to each second line segment; where the second minimum distance line segment is the line segment with the smallest distance from the second line segment. Calculate the second included angle between the second line segment and the second minimum distance line segment respectively; The first fluctuation range formed by multiple second included angles is statistically analyzed; Based on a preset redundancy coefficient, the first fluctuation range is expanded to obtain the included angle fluctuation range.

10. A chip defect detection device, characterized in that, The chip defect detection device includes: The acquisition unit is used to acquire the image to be recognized collected by the vision module; The extraction unit is used to extract the chip image region in the image to be identified, and extract the current packaging region and multiple current pin regions in the chip image region according to the first color distribution range corresponding to the base and the second color distribution range corresponding to the pin. The first judgment unit is used to obtain a standard encapsulation area and, based on the pixel difference between the same pixel position in the standard encapsulation area and the current encapsulation area, and the distribution relationship between the four diagonal points of the current encapsulation area and the preset feature points, confirm whether there is an encapsulation defect in the current encapsulation area; the encapsulation defect includes encapsulation missing, encapsulation damaged, or abnormal printing pattern. The second determination unit is used to determine whether pin defects occur in multiple current pin areas based on the first package pin ratio relationship, the second package pin ratio relationship, the current package area length, the current package area width, the current pin area length, and the current pin area width when it is determined that there are no defects in the current package area. The pin defects include pin offset defects, pin unevenness defects, and pin missing defects. The third judgment unit is used to determine whether there is a positional distribution defect between the package area and the pin area when it is determined that there is no defect in the current pin area. This is based on whether multiple first angles obtained from the first center of the current package area and the second centers of multiple current pin areas are within the angle fluctuation range. The angle fluctuation range is used to describe the distribution relationship between multiple pin areas.

Images