Method for inspecting a lead frame

Abstract

A kind of lead frame flaw detection method, comprising: providing standard template image, initial deviation range and n lead frames, n is natural number;Deviation range iteration debugging is carried out several times based on the standard template image, initial deviation range and n lead frames, and first deviation range is obtained.The method can reduce false positives in the production process, lead frame flaw detection, improve the overkill situation of good lead frame, improve the detection speed of equipment, production efficiency and yield.

Description

Technical Field

[0001] This invention relates to the field of lead frame inspection technology, and more particularly to a method for detecting defects in lead frames. Background Technology

[0002] As the chip carrier for many integrated circuits, the reliability and stability of the lead frame determines the function and performance of the final semiconductor product. Therefore, quality inspection of lead frame materials is necessary during the actual production process.

[0003] Conventional lead frame inspection equipment checks the input lead frame materials for defects, and then determines the shipment category of the lead frame based on the defect inspection results: those with a high degree of hazard are directly scrapped, while those with a low degree of hazard or that do not affect the function are judged as good products.

[0004] In existing technologies, optical platforms are typically used to acquire source images of mass-produced leadframes. Furthermore, based on a relatively strict deviation standard, the source images of the leadframes are compared with standard template images to detect defects in the mass-produced leadframes.

[0005] However, due to factors such as reflections during the acquisition of the source image of the lead frame, the texture of the lead frame itself, and the material, the difference between the source image of a good lead frame and the standard template image may exceed the deviation standard. This results in a large number of false alarms during defect detection, leading to over-detection of good lead frames and lower equipment inspection speed, production efficiency, and yield. Summary of the Invention

[0006] The technical problem solved by this invention is to provide a defect detection method for lead frames, which reduces false alarms during lead frame defect detection in the production process, improves the over-detection of good lead frames, and increases equipment detection speed, production efficiency and yield.

[0007] To address the aforementioned technical problems, the present invention provides a method for detecting defects in lead frames, comprising: providing a standard template image, an initial deviation range, and m lead frames, where m is a natural number; performing several iterations of deviation range adjustment based on the standard template image, the initial deviation range, and the m lead frames to obtain a first deviation range, wherein the method for the nth iteration of deviation range adjustment includes: obtaining an nth source image corresponding to the standard template image based on the nth lead frame; detecting the nth defect based on the nth source image, the standard template image, and the (n-1)th first deviation range. For each nth defect, obtain the nth defect data, and when n=1, the first deviation range of the (n-1)th time is the initial deviation range; based on several nth defect data, obtain the nth defect false alarm rate, which refers to the proportion of the number of falsely reported nth defects in the total number of detected nth defects; when the nth defect false alarm rate is greater than the preset false alarm rate, based on several nth defect data corresponding to several falsely reported nth defects and a preset relaxation parameter, expand the first deviation range of the (n-1)th time to generate the first deviation range of the nth time.

[0008] Optionally, the method for iterative debugging of the nth deviation range further includes: when the nth defect false alarm rate is less than or equal to the preset false alarm rate, the first deviation range of the (n-1)th iteration is used as the first deviation range of the nth iteration.

[0009] Optionally, the method for the nth deviation range iteration debugging further includes: when the nth defect false alarm rate is less than or equal to the preset false alarm rate, after generating the first deviation range for the nth time, ending the deviation range iteration debugging.

[0010] Optionally, the method for the nth deviation range iteration debugging further includes: when the nth deviation range iteration debugging reaches a preset number, after generating the first deviation range for the nth time, ending the deviation range iteration debugging.

[0011] Optionally, the standard template image includes several independent standard regions, and the first deviation range of the (n-1)th time includes several region deviation ranges that correspond one-to-one with the several standard regions.

[0012] Optionally, the plurality of standard areas include at least one of electroplating and critical areas, semi-etched areas, functional areas, low-threshold areas and non-functional areas.

[0013] Optionally, the preset relaxation parameters include several region relaxation parameters that correspond one-to-one with several standard regions.

[0014] Optionally, the nth defect data corresponding to any nth defect includes at least the nth position information; the method for expanding the first deviation range of the (n-1)th time based on the nth defect data corresponding to the nth defect that was falsely reported and the preset relaxation parameter includes: obtaining a number of designated standard areas in a number of standard areas based on some or all of the nth position information that was falsely reported, wherein the designated standard areas have position information corresponding to the nth position information that was falsely reported; expanding the area deviation range of the (n-1)th time corresponding to each designated standard area based on the area relaxation parameter corresponding to each designated standard area, so as to generate the area deviation range of the nth time corresponding to each designated standard area.

[0015] Optionally, the preset relaxation parameters include individual relaxation parameters for partially expanding the area deviation range; the nth defect data corresponding to any nth defect includes: the corresponding nth location information and the nth defect image; the method for expanding the first deviation range of the (n-1)th time based on the nth defect data corresponding to the nth defect and the preset relaxation parameters includes: obtaining a number of nth defect areas containing the nth defect images in the nth source image based on some or all of the nth location information and the nth defect images; obtaining a number of corresponding nth defect regions in a number of standard regions based on the nth defect regions, and the area deviation range of the (n-1)th time corresponding to any nth defect region includes: the sub-area deviation range of the (n-1)th time corresponding to the nth defect region; expanding the sub-area deviation range of the (n-1)th time corresponding to each nth defect region based on the individual relaxation parameters to generate the corresponding area deviation range of the nth time.

[0016] Optionally, the method for expanding the sub-region deviation range corresponding to the (n-1)th defect region for each nth defect region according to the individual relaxation parameter includes: performing a number of false alarm reception steps, and the method for each false alarm reception step includes: expanding the sub-region deviation range corresponding to the (n-1)th defect region for one nth defect region according to the individual relaxation parameter.

[0017] Optionally, the method for expanding the sub-region deviation range of each nth defective region corresponding to the (n-1)th time according to the individual relaxation parameter includes: expanding the sub-region deviation range of all nth defective regions corresponding to the (n-1)th time in a single step according to the individual relaxation parameter.

[0018] Optionally, the preset relaxation parameter further includes an individual relaxation parameter for partially expanding the range of the area deviation; the nth defect data corresponding to any nth defect includes: the corresponding nth location information and the nth defect image; the method for expanding the first deviation range of the (n-1)th time based on the nth defect data corresponding to the falsely reported nth defects and the preset relaxation parameter includes: obtaining a number of corresponding nth falsely reported defect types based on the number of falsely reported nth defect data; obtaining a number of specified nth defect images from the number of falsely reported nth defect images based on the number of falsely reported nth location information, the number of falsely reported nth defect images, and the number of falsely reported nth defect types; and determining, based on the nth falsely reported defect type corresponding to any specified nth defect image, to expand the range of the area deviation of the (n-1)th time corresponding to the standard area corresponding to the arbitrary specified nth defect image according to the area relaxation parameter or the individual relaxation parameter.

[0019] Optionally, the method for expanding the range of the (n-1)th region deviation corresponding to the standard area corresponding to the arbitrary specified n-th defective image according to the region relaxation parameter includes: obtaining a specified standard area from several standard areas according to the n-th position information corresponding to the arbitrary specified n-th defective image, wherein the specified standard area has position information corresponding to the n-th position information; and expanding the range of the (n-1)th region deviation corresponding to the specified standard area according to the region relaxation parameter corresponding to the specified standard area.

[0020] Optionally, the method for expanding the (n-1)th region deviation range corresponding to the standard area corresponding to the arbitrary specified nth defective image according to the individual relaxation parameter includes: obtaining the nth defective area containing the arbitrary specified nth defective image in the nth source image according to the arbitrary specified nth defective image and the corresponding nth position information; obtaining the corresponding nth defective region in a plurality of standard areas according to the nth defective region, and the (n-1)th region deviation range corresponding to the standard area where the nth defective region is located includes: the (n-1)th sub-region deviation range corresponding to the nth defective region; expanding the (n-1)th sub-region deviation range corresponding to the nth defective region according to the individual relaxation parameter.

[0021] Optionally, the region widening parameters include at least one of the following: region diagonal length widening parameter, region area widening parameter, and region contrast widening parameter.

[0022] Optionally, the individual relaxation parameters include at least one of the individual diagonal length relaxation parameters, individual area relaxation parameters, and individual contrast relaxation parameters.

[0023] Optionally, the nth source image includes several repeating cell regions, each cell region having an nth source image corresponding to the standard template image, each cell region having origin coordinates, and the nth position information being the origin coordinates or relative coordinates to the origin coordinates.

[0024] Optionally, the method for detecting the nth defect based on the nth source image, the standard template image, and the (n-1)th first deviation range includes: comparing the nth source image of each unit region with the standard template image and the (n-1)th first deviation range.

[0025] Optionally, the type of the preset relaxation parameter includes a numerical value or a ratio.

[0026] Compared with the prior art, the technical solution of the embodiments of the present invention has the following beneficial effects:

[0027] In the defect detection method for the lead frame of the present invention, in the nth iteration of several deviation range iterations, the nth defect is detected based on the nth source image, the standard template image, and the first deviation range of the (n-1)th iteration, and the nth defect data corresponding to each nth defect is obtained; based on the nth defect data, the nth defect false alarm rate is obtained; when the nth defect false alarm rate is greater than the preset false alarm rate, the first deviation range of the (n-1)th iteration is expanded based on the nth defect data corresponding to the falsely reported nth defects and the preset relaxation parameter to generate the first deviation range of the nth iteration. Therefore, based on the falsely reported defects in the current defect detection (i.e., the nth falsely reported defect), the first deviation range of the previous (n-1th) defect can be locally expanded. Thus, the first deviation range obtained through several iterations of deviation range adjustments not only expands the deviation range corresponding to the falsely reported defects based on the initial deviation range, but also reduces or eliminates the impact on the deviation range corresponding to the actual defects, ensuring that the actual defects can still be detected. This reduces falsely reported defects during lead frame defect detection in the production process, improves the accuracy of defect detection, reduces over-detection of good lead frames, and increases equipment detection speed, production efficiency, and yield. Attached Figure Description

[0028] Figure 1 This is a flowchart illustrating a defect detection method for a lead frame according to an embodiment of the present invention.

[0029] Figure 2 This is a schematic diagram of a standard template image in one embodiment of the present invention;

[0030] Figure 3 This is a three-dimensional structural schematic diagram of a lead frame according to an embodiment of the present invention;

[0031] Figure 4 This is a flowchart illustrating the method for iterative debugging of the nth deviation range according to an embodiment of the present invention.

[0032] Figures 5 to 8 This is a schematic diagram of each step in the nth iteration of deviation range adjustment according to an embodiment of the present invention;

[0033] Figure 9 This is a schematic diagram of a designated standard area according to another embodiment of the present invention. Detailed Implementation

[0034] As described in the background art, due to factors such as reflections, textures, and materials of the lead frame itself, the difference between the source image of a good lead frame and the standard template image may exceed the deviation standard. This results in a large number of false alarms during defect detection, leading to over-detection of good lead frames and lower equipment detection speed, production efficiency, and yield.

[0035] To address the aforementioned technical problems, the present invention provides a defect detection method for lead frames. By iteratively adjusting the deviation range based on the standard template image, the initial deviation range, and m lead frames, a first deviation range is obtained. This method can reduce false alarms during lead frame defect detection in the production process, improve the over-detection of good lead frames, and increase equipment detection speed, production efficiency, and yield.

[0036] To make the above-mentioned objectives, features and beneficial effects of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0037] Figure 1 This is a flowchart illustrating a defect detection method for a lead frame according to an embodiment of the present invention.

[0038] Please refer to Figure 1 The defect detection method for the lead frame includes:

[0039] Step S100: Provide a standard template image, an initial deviation range, and m lead frames;

[0040] Step S200: Based on the standard template image, the initial deviation range, and m lead frames, perform several iterations of deviation range debugging to obtain the first deviation range.

[0041] The following is a detailed description in conjunction with the accompanying drawings.

[0042] Please refer to Figure 2 and Figure 3 , Figure 2 This is a schematic diagram of a standard template image according to an embodiment of the present invention. Figure 3This is a three-dimensional structural schematic diagram of a lead frame according to an embodiment of the present invention, providing a standard template image 100, an initial deviation range, and m lead frames 200.

[0043] m is a natural number.

[0044] In this embodiment, the standard template image 100 includes several independent standard areas 101.

[0045] Several standard areas 101 include one or more of the following: plating and key areas, half-etched areas, functional areas, large low-value areas, and non-functional areas.

[0046] Different standard areas 101 correspond to areas in the lead frame 200 that require different functions, areas formed by special processes, and so on. The standard template image 100 is used as a comparison benchmark and is compared with the source image of the lead frame 200 (e.g., the nth source image described later) to determine whether the deviation between the two is within the required deviation range (e.g., the initial deviation range, the first deviation range, etc.) to achieve defect detection of the lead frame 200.

[0047] Specifically, reference data can be obtained from the standard template image 100, and the standard template image 100 is configured with corresponding coordinate data.

[0048] The reference data includes at least one of several area parameters, several diagonal length parameters, and several contrast parameters corresponding to the standard template image 100.

[0049] The corresponding coordinate data includes: the origin coordinates of the standard template image 100, and several relative coordinates relative to the origin coordinates, so as to locate each standard area 101 or a specific local image in the standard template image 100.

[0050] In some practical application scenarios, the origin coordinates of the standard template image 100 can be set at the center of the standard template image 100 or at one of the four corners of the standard template image 100.

[0051] The initial deviation range is a deviation range based on the reference data. Specifically, the initial deviation range includes at least one of the following: the deviation range of the area parameter, the deviation range of the diagonal length parameter, and the deviation range of the contrast parameter.

[0052] In this embodiment, the type of the initial deviation range includes at least one of numerical and proportional values.

[0053] For example, it could be a numerical value used as the deviation value for the area parameter, or a ratio used as the deviation ratio for the contrast parameter, and so on.

[0054] It is important to understand that the unit of the numerical value is consistent with the unit of the reference value for that value. For example, the unit of the numerical value used as a deviation value for the area parameter is consistent with the unit of the area parameter.

[0055] In this embodiment, the initial deviation range includes deviation ranges for several areas that correspond one-to-one with several standard areas 101. This improves the accuracy of defect detection and reduces the risk of over-detection or under-detection.

[0056] The deviation range of the area also includes at least one of the deviation range of the area parameter, the deviation range of the diagonal length parameter, and the deviation range of the contrast parameter.

[0057] It's important to understand that the zone deviation range reflects the correspondence between the deviation ranges of the area parameter, the diagonal length parameter, and the contrast parameter in the initial deviation range, and the standard zone. In other words, each zone deviation range independently has its own deviation value or deviation ratio based on the area parameter, contrast parameter, and diagonal length parameter.

[0058] For example, the initial deviation range corresponding to the electroplating and critical areas includes the deviation range of the area parameter with a deviation value of 5. Meanwhile, the initial deviation range corresponding to the functional areas includes the deviation range of the area parameter with a deviation ratio of 10%, and the deviation range of the diagonal length parameter with a deviation value of 3.

[0059] In this embodiment, the lead frame 200 includes a plurality of repeating units 201.

[0060] In this embodiment, the plurality of repeating units 201 are arranged in an array along mutually perpendicular directions X and Y. Direction X is parallel to the width direction of the lead frame 200, and direction Y is parallel to the length direction of the lead frame 200.

[0061] In this embodiment, a source image of a flawless unit 201 is acquired using an optical platform to obtain the standard template image 100. The optical platform includes a CCD camera, etc.

[0062] It should be understood that the optical platform, related equipment, and methods for acquiring source images in this embodiment should not be considered as features that limit the scope of protection of this invention.

[0063] It should be noted that, for ease of explanation, Figure 2The image only schematically represents a portion of the standard region 101 within the standard template image 100. Figure 3 The diagram only schematically represents a portion of the units 201 in the lead frame 200.

[0064] It should be noted that the unit 201 can be the smallest repeating unit in the lead frame 200, or it can include more than two smallest repeating units. Furthermore, although... Figure 3 In both directions X and Y, there is a spacing between adjacent units 201. However, the spacing between adjacent units 201 can be determined according to actual needs, such as the specific area where defects need to be detected, the structural design of the lead frame, etc.

[0065] Please continue to refer to this. Figure 2 and Figure 3 Based on the standard template image 100, the initial deviation range, and m lead frames 200, several iterations of deviation range debugging are performed to obtain the first deviation range.

[0066] Please refer to Figure 4 , Figure 4 This is a flowchart illustrating a method for iterative debugging of the nth deviation range according to an embodiment of the present invention. The method for iterative debugging of the nth deviation range includes:

[0067] Step S210: Based on the nth lead frame, obtain the nth source image corresponding to the standard template image;

[0068] Step S220: Detect the nth defect based on the nth source image, the standard template image, and the first deviation range of the (n-1)th time, and obtain the nth defect data corresponding to each nth defect;

[0069] Step S230: Obtain the false alarm rate of the nth defect based on several nth defect data;

[0070] When the false alarm rate of the nth defect is greater than the preset false alarm rate, step S240 is executed, and the first deviation range of the (n-1)th time is expanded to generate the first deviation range of the nth time based on the nth defect data corresponding to the nth defect and the preset relaxation parameter.

[0071] By executing steps S210 to S230, the first deviation range from the previous (n-1) iteration can be locally expanded based on the falsely reported defect (i.e., the falsely reported nth defect) in the current defect detection. Thus, the first deviation range obtained through several iterations of deviation range adjustments not only expands the deviation range corresponding to the falsely reported defect based on the initial deviation range, but also reduces or eliminates the impact on the deviation range corresponding to the actual defect, ensuring that the actual defect can still be detected. This reduces falsely reported defects during lead frame defect detection in the production process, improves the accuracy of defect detection, reduces over-detection of good lead frames, and increases equipment detection speed, production efficiency, and yield.

[0072] In some practical applications, the m lead frames are lead frames from the trial production stage, meaning the nth deviation range iteration debugging is performed during the trial production stage of the lead frames. Therefore, by adopting the formed nth first deviation range, false positives during defect detection can be reduced during the mass production stage of the lead frames, improving the accuracy of defect detection in the mass production stage and mitigating over-detection of good lead frames during mass production. Furthermore, the mass production stage itself does not require time for the iterative debugging.

[0073] In other practical application scenarios, the m lead frames are not limited to lead frames in the trial production stage, but may also include lead frames in the mass production stage, so as to more flexibly deal with the risk of false alarms of defects that occur in the mass production stage.

[0074] In this embodiment, the method for iterative debugging of the nth deviation range further includes: when the nth defect false alarm rate is less than or equal to a preset false alarm rate, executing step S250, using the first deviation range of the (n-1)th iteration as the first deviation range of the nth iteration; and, after generating the first deviation range of the nth iteration, executing step S260 to end the deviation range iterative debugging. Since the determination of whether the debugging of the first deviation range is complete is entirely based on whether the nth defect false alarm rate is within the range of the preset false alarm rate, the accuracy of defect detection based on the first deviation range is further improved.

[0075] In some other embodiments, the method for iterative debugging of the nth deviation range further includes: when the nth deviation range iteration debugging reaches a preset number, after generating the first deviation range for the nth iteration, executing step S260. That is, when either the nth defect false alarm rate is less than or equal to a preset false alarm rate, or the nth deviation range iteration debugging reaches a preset number, the deviation range iteration debugging ends, and the final first deviation range is obtained. Thus, this method effectively balances improving defect detection accuracy with improving the efficiency of obtaining the first deviation range.

[0076] In some other embodiments, step S250 is not executed, and the method for iterative debugging of the nth deviation range includes: executing step S240 when the nth defect false alarm rate is less than or equal to a preset false alarm rate; and executing step S260 when the nth deviation range iterative debugging reaches a preset number of iterations. That is, regardless of whether the nth defect false alarm rate is less than or equal to the preset false alarm rate, the first deviation range of the (n-1)th iteration is expanded to generate the first deviation range of the nth iteration. Furthermore, when the number of iterations n reaches a preset number of iterations, the deviation range iterative debugging ends, and the final first deviation range is obtained. Therefore, by controlling the preset number of iterations, both the accuracy of defect detection and the efficiency of obtaining the first deviation range can be better balanced.

[0077] Figures 5 to 8 This is a schematic diagram of each step in the nth iteration of deviation range adjustment according to an embodiment of the present invention. The following is in conjunction with... Figures 5 to 8 The method for iterative debugging of the nth deviation range is described in detail.

[0078] Please refer to Figure 5 According to the nth lead frame 200 (e.g. Figure 3 As shown), obtain the standard template image 100 (e.g. Figure 2 The nth source image (as shown) is 301. n .

[0079] Specifically, the nth source image refers to the source image of the nth lead frame 200.

[0080] Similar to the acquisition method of standard template image 100, this embodiment utilizes an optical platform to acquire the nth source image 301. n Obtain the standard template image 100.

[0081] In this embodiment, the nth source image 301 n Includes several repeating cell regions 300 n 300 per unit area n Image n, source image 301 n It corresponds to the standard template image 100.

[0082] Specifically, the unit area 300 n Image n, source image 301 n It is the source image of unit 201 of the nth lead frame 200.

[0083] In this embodiment, each unit area is 300 n Configure the corresponding coordinate data.

[0084] Similarly, the corresponding coordinate data includes: located in cell region 300 n The origin coordinates and several relative coordinates relative to the origin coordinates are used to define each unit region 300. n The standard areas 101 or specific local images in the image are used for positioning.

[0085] In some practical application scenarios, unit area 300 n The origin coordinates can be set in cell 300. n The central location, or set in unit area 300 n One of the four corners.

[0086] It should be noted that unit area 300 n The origin coordinates are in cell 300. n The position of the origin coordinates in the standard template image 100 corresponds to the position of the origin coordinates in the standard template image 100. For example, when the origin coordinates of the standard template image 100 are at the center of the standard template image 100, then the unit area 300... n The origin coordinates are in cell 300. n The center.

[0087] Therefore, in the process of defect detection of the lead frame, the coordinate data of the standard template image 100 and the coordinate data of each unit area 300 can be used. n The corresponding coordinate data enables image-to-image mapping (standard template image 100 and each unit area 300). n The nth source image 301 n The comparison and alignment between images. It is important to understand that, in actual comparison, the comparison between images can also refer to the comparison between parameters formed by the transformation of images.

[0088] Furthermore, since each unit area in this embodiment has 300 n The corresponding coordinate data is 300 for each cell area. n The origin coordinates, or based on each unit area of ​​300 n The relative coordinates of the origin are used to reduce the computational load of data processing, thereby reducing the time required to acquire defect images and data and improving the efficiency of defect detection.

[0089] In some other embodiments, the nth source image 301 n Configure the corresponding coordinate data. The corresponding coordinate data includes: located in the nth source image 301. n The origin coordinates and absolute coordinates relative to the origin coordinates are required. Based on this, data processing of the absolute coordinates is needed to establish the relationship between the standard template image 100 and each unit region 300. nThe nth source image 301 n Positioning and comparison between them.

[0090] Please refer to Figure 6 According to the nth source image 301 n The standard template image 100 and the first deviation range detection of the nth defect (not shown) are used to obtain the nth defect data corresponding to each nth defect.

[0091] Specifically, when n=1, the first deviation range of the (n-1)th iteration is the initial deviation range; when n>1, the first deviation range of the (n-1)th iteration is the first deviation range generated during the (n-1)th iteration of deviation range debugging in history.

[0092] Therefore, the first deviation range of the (n-1)th time includes several zone deviation ranges of the (n-1)th time, which correspond one-to-one with several standard zones 101. Wherein, when n=1, the zone deviation range of the (n-1)th time is the same as the zone deviation range in the initial deviation range.

[0093] The type of the first deviation range for the (n-1)th time also includes at least one of numerical and proportional values.

[0094] The nth defect is the defect of the nth lead frame 200.

[0095] In this embodiment, the step of using the nth source image 301 n The method for detecting the nth defect using the standard template image 100 and the first deviation range of the (n-1)th iteration includes: comparing each unit region 300 based on the standard template image 100 and the first deviation range of the (n-1)th iteration. n The nth source image 301 n .

[0096] In other words, by using the standard template image 100 as a reference, each unit area is 300. n The nth source image 301 n In comparison, obtain the 300 used to represent each unit area. n The nth source image 301 n Several deviation feature parameters are defined for several deviations that occur between the image and the standard template image 100; when any deviation feature parameter exceeds the first deviation range of the (n-1)th time, it is determined that the nth defect corresponding to the arbitrary deviation feature parameter is detected, and the nth defect data corresponding to the nth defect is obtained.

[0097] In this embodiment, the deviation characteristic parameters include at least one of the following: diagonal length deviation parameter, area deviation parameter, and contrast deviation parameter.

[0098] To facilitate comparison with the first deviation range of the (n-1)th time, the type of the deviation characteristic parameter also includes at least one of numerical or proportional values.

[0099] It is important to understand that due to factors such as reflections during the acquisition of the nth source image, the texture of the nth lead frame 200 itself, and its material composition, the detected nth defect may be a false nth defect. That is, the detected nth defects include both real nth defects and falsely reported nth defects. Correspondingly, the acquired nth defect data includes: the correct nth defect data corresponding to the real nth defect, and the falsely reported nth defect data corresponding to the falsely reported nth defect.

[0100] In this embodiment, the nth defect data corresponding to any nth detected defect includes: the corresponding nth location information and the nth defect image 302. n .

[0101] The nth position information is used to locate the position of any nth defect.

[0102] In this embodiment, the nth position information is cell region 300. n The origin coordinates, or relative to the unit region 300 n The relative coordinates of the origin are used to directly locate the position of any nth defect in unit 301.

[0103] In some other embodiments, the nth location information can also be the nth source image 301. n The origin coordinates, or relative to the nth source image 301 n The absolute coordinates of the origin.

[0104] It should be understood that, for ease of explanation, Figure 6 The image 302 only schematically represents a portion of the nth defective image. n .

[0105] In another embodiment, the nth defect data corresponding to any nth defect detected includes: nth location information.

[0106] Please continue to refer to this. Figure 6 Based on several defect data points, the false alarm rate for the nth defect is obtained.

[0107] The false alarm rate for the nth defect refers to the proportion of the number of falsely reported nth defects to the total number of all nth defects detected.

[0108] Specifically, the method for obtaining the false alarm rate of the nth defect includes: judging whether each detected nth defect is a false alarm based on several nth defect data; and obtaining the false alarm rate of the nth defect based on the number of false alarms of the nth defect and the total number of nth defects.

[0109] In this embodiment, the method for determining whether each detected nth defect is a false alarm based on several nth defect data includes: observing the nth defect image 302. n The system determines whether the detected nth defect is a false alarm.

[0110] Preferably, the method for determining whether each detected nth defect is a false alarm based on several nth defect data includes: observing the nth defect image 302. n By combining the corresponding nth position information (i.e., the location of the nth defect image), it is determined whether the detected nth defect is a false alarm. This improves the accuracy of the judgment.

[0111] In another embodiment, the method for determining whether each detected nth defect is a false alarm based on a plurality of nth defect data includes: observing the nth source image 301. n Based on the information at the nth position, it is determined whether the detected nth defect is a false alarm.

[0112] Next, when the false alarm rate of the nth defect is greater than the preset false alarm rate, the first deviation range of the (n-1)th time is expanded to generate the first deviation range of the nth time based on the nth defect data corresponding to the nth defect and the preset relaxation parameter.

[0113] Preferably, the preset false alarm rate is 10%.

[0114] In this embodiment, the type of the preset relaxation parameter includes numerical value or ratio.

[0115] In this embodiment, the preset relaxation parameters include individual relaxation parameters, which are used to partially expand the range of the deviation of the region.

[0116] It should be noted that the aforementioned partial expansion refers to: for a portion of the standard area 101, the corresponding area deviation range is locally expanded.

[0117] Specifically, the individual relaxation parameters can be either numerical or proportional.

[0118] Specifically, the individual broadening parameters include at least one of the following: individual diagonal length broadening parameter, individual area broadening parameter, and individual contrast broadening parameter. The individual diagonal length broadening parameter broadens the deviation range of the diagonal length parameter within the first deviation range; the individual area broadening parameter broadens the deviation range of the area parameter within the first deviation range; and the individual contrast broadening parameter broadens the deviation range of the contrast parameter within the first deviation range.

[0119] Furthermore, it should be understood that in practical applications, due to the different optical platforms used to acquire the standard template image 100 and the nth source image, the contrast parameter may include one or more of the following: red pixel value (R value), green pixel value (G value), blue pixel value (B value), and grayscale value. Correspondingly, the individual contrast relaxation parameter may include one or more of the following: an individual red pixel value relaxation parameter corresponding to the red pixel value, an individual green pixel value relaxation parameter corresponding to the green pixel value, an individual blue pixel value relaxation parameter corresponding to the blue pixel value, and an individual grayscale value relaxation parameter corresponding to the grayscale value.

[0120] The following will combine Figures 7 to 8 The steps in the method of expanding the first deviation range of the (n-1)th time to generate the first deviation range of the nth time based on the nth defect data corresponding to the nth defect that is falsely reported and the preset relaxation parameter will be explained.

[0121] Please refer to Figure 7 Based on several false alarms of the nth position information and several false alarms of the nth defect image 302 n Part or all of the nth source image 301 n 302 images of nth defects containing false positives are obtained. n Several nth defective areas 303 n .

[0122] It is important to understand that since the purpose of expanding the first deviation range is to ensure that the false alarm rate of defects in the lead frame 200 is below a preset false alarm rate when detecting defects, it is not necessary to expand the first deviation range for all falsely reported nth defects. Instead, it can be selected for some or all of them based on the actual situation. In other words, depending on the actual situation, the first deviation range can be expanded based on several falsely reported nth position information and several falsely reported nth defect images 302. n In the part, obtain several nth defective regions 303 n To expand the first deviation range of the (n-1)th time to generate the first deviation range of the nth time; alternatively, it can be based on several nth position information of all false alarms and several nth defect images 302 of false alarms. nObtain several nth defective regions 303 n .

[0123] Furthermore, due to the nth source image 301 n Includes several unit areas 300 n Therefore, the plurality of nth defective regions 103 n Possibly located in the same unit area 300 n It may also be located in different unit areas 300 n middle.

[0124] Specifically, based on any nth defective image 302 that is falsely reported n and the nth defective image 302 n The corresponding nth position information can be found in any nth defect image 302. n Unit 300 n Obtain the corresponding nth defect region 303 from the middle. n The corresponding nth defect area 303 n Image 302 containing any nth defect n .

[0125] In this embodiment, the nth defect region 303 n The range is greater than that of the nth defect region 303. n Image 302 of the nth defect n .

[0126] In this embodiment, the nth defect region 303 n These are regions with regular shapes to reduce computational load. Such regions are, for example, rectangular regions, symmetrical polygonal regions, etc.

[0127] In other embodiments, the nth defect region 303 n It can also be an area with an irregular shape.

[0128] In this embodiment, based on the nth defective image 302 n The edges form an initial nth defect region (not shown) with a regular shape, the initial nth defect region containing the nth defect image 302. n ; Along the vertical direction of the edge of the initial nth defect area, the edge of the initial nth defect area is expanded outward by a predetermined distance to form the nth defect area 303. n Therefore, the nth defective image 302 is better protected. n In the nth defect area 303 n Within the range.

[0129] In some other embodiments, along the nth defect image 302 nThe defect area is formed by expanding outwards by a predetermined distance in the vertical direction of the edge. In some other embodiments, this expansion is not performed.

[0130] It is important to understand that in practical applications, the nth defect image 302 can be used as a basis. n The resolution is determined, and the number of pixels corresponding to the preset size is obtained for the expansion.

[0131] For example, when the preset distance is 0.05mm, when the nth defective image 302 n When the resolution is 16k, the number of pixels corresponding to the preset distance is 4. Similarly, when the resolution is 4k, the number of pixels corresponding to the preset distance is 1; when the resolution is 8k, the number of pixels corresponding to the preset distance is 2, and so on.

[0132] Please refer to Figure 8 According to several nth defect areas 303 n Within several standard areas 101, obtain several corresponding nth defective areas 103. n .

[0133] Specifically, with any nth defect region 303 n The corresponding nth defective region 103 n This refers to the corresponding nth defective region 103. n The position and extent in the standard template image 100, and the arbitrary nth defect region 303 n In unit area 300 n The position and range are the same.

[0134] It should be understood that, since the standard template image 100 includes several standard regions 101, any nth defective region 103 n It will also be located within at least one of several standard areas 101. Specifically, being located within two or more standard areas 101 represents the nth defective area 103. n Cross-region. Furthermore, since the standard template image 100 is compared with each unit region 300n to detect the nth defect, therefore, based on several nth defect regions 303... n Obtain the corresponding nth defective region 103 n This refers to: forming, within several standard areas 101 of the standard template image 100, elements that correspond to all unit areas 300. n All nth defective regions 303 n One-to-one correspondence of the nth defective region 103 n .

[0135] In this embodiment, any nth defective region 103 nThe range of the area deviation corresponding to the (n-1)th time in the standard area 101 includes: the range of the area deviation corresponding to the arbitrary nth defect area 103. n The corresponding sub-region deviation range for the (n-1)th time.

[0136] Specifically, when n=1, since the first deviation range of the (n-1)th iteration is the initial deviation range that has not yet been adjusted, it is related to the arbitrary nth defect region 103. n The corresponding sub-region deviation range for the (n-1)th time is the arbitrary nth defect region 103 n The zone deviation range within the initial deviation range corresponding to standard zone 101.

[0137] When n > 1, due to the iterative adjustments of the deviation range based on individual relaxation parameters from the 1st to the (n-1st)th time in history, the deviation range corresponding to any standard region 101 may be partially expanded due to the adjustments. That is, any nth defective region 103 n The (n-1)th deviation range corresponding to the standard area 101 is not necessarily consistent throughout the entire standard area 101. For the (n-1)th deviation range, different local areas of the standard area 101 may have the same or different deviation ranges based on the reference data.

[0138] Based on this, when n>1, and with the aforementioned arbitrary nth defect region 103 n The corresponding sub-region deviation range for the (n-1)th time refers to: the range relative to the arbitrary nth defective region 103 n The deviation ranges corresponding to several local regions in the data, with the reference data as the reference.

[0139] Please continue to refer to this. Figure 8 Based on the individual relaxation parameters, each nth defect region is expanded by 103. n The corresponding (n-1)th sub-region deviation range is used to generate the corresponding nth region deviation range.

[0140] By expanding each nth defect region by 103 n The corresponding sub-region deviation range for the (n-1)th time improves the targeting of deviation adjustment, thereby further improving the accuracy of defect detection through the first deviation range.

[0141] The range of all deviations in the nth iteration constitutes the first deviation range of the nth iteration.

[0142] It should be noted that, based on the individual relaxation parameters, the arbitrary nth defect region 103 is expanded. n The corresponding (n-1)th sub-region deviation range refers to: based on the individual relaxation parameter, the deviation range relative to the arbitrary nth defect region 103n The deviation ranges corresponding to several local areas in the data, based on the reference data, are all expanded. However, this expansion does not apply to any nth defective region 103. n The deviation range of the area corresponding to the standard area 101 outside the standard area is affected.

[0143] In this embodiment, each nth defect region is expanded by 103 according to the individual relaxation parameter. n The method for determining the sub-region deviation range for the (n-1)th time includes performing several false alarm reception steps.

[0144] The method for each false alarm reception step includes: expanding the sub-region deviation range of the (n-1)th time corresponding to the nth defective region according to the individual relaxation parameter.

[0145] Specifically, through several false alarm reception steps, the debugging status of the sub-region deviation range corresponding to each nth defective region in the (n-1)th iteration can be confirmed separately.

[0146] In other embodiments, each nth defect region is expanded by 103 according to the individual relaxation parameter. n The method for determining the sub-region deviation range for the corresponding (n-1)th time includes: expanding the entire nth defect region 103 in a single step according to the individual relaxation parameter. n The corresponding sub-region deviation range for the (n-1)th time.

[0147] Compared to methods that employ multiple false alarm reception steps, this method, based on the individual relaxation parameters, expands the entire nth defect region 103 in a single step. n The corresponding sub-region deviation range for the (n-1)th iteration can better improve debugging efficiency.

[0148] It is important to understand that when n > 1, several false alarm reception steps can be used for debugging during the (n-1)th iteration of deviation range debugging in history. However, during the nth iteration of deviation range debugging, the deviation range of the sub-region corresponding to the n-1th defect area is expanded in a single step for debugging. This allows for more flexible selection of the appropriate debugging method in each iteration based on the actual situation.

[0149] In another embodiment, the preset relaxation parameters include: a plurality of region relaxation parameters corresponding one-to-one with a plurality of standard regions 101. Furthermore, the nth defect data corresponding to any nth defect includes at least nth location information. For a detailed explanation of the nth location information in the other embodiment, please refer to the description of the nth location information in the foregoing embodiments; it will not be repeated here.

[0150] In another embodiment, the type of the preset relaxation parameter includes a numerical value or a ratio.

[0151] In another embodiment, the region widening parameter includes at least one of a region diagonal length widening parameter, a region area widening parameter, and a region contrast widening parameter. Specifically, the region diagonal length widening parameter is used to widen the deviation range of the diagonal length parameter within the first deviation range; the region area widening parameter is used to widen the deviation range of the area parameter within the first deviation range; and the region contrast widening parameter is used to widen the deviation range of the contrast parameter within the first deviation range.

[0152] Similar to individual contrast relaxation parameters, regional contrast relaxation parameters may include one or more of the following: red pixel value relaxation parameters for the region corresponding to the red pixel value, green pixel value relaxation parameters for the region corresponding to the green pixel value, blue pixel value relaxation parameters for the region corresponding to the blue pixel value, and grayscale value relaxation parameters for the region corresponding to the grayscale value.

[0153] Figure 9 This is a schematic diagram of a designated standard area according to another embodiment of the present invention, which will be discussed below in conjunction with... Figure 9 In another embodiment of the present invention, the steps of the method for expanding the first deviation range of the (n-1)th time to generate the first deviation range of the nth time based on the nth defect data corresponding to the nth defect that is falsely reported and the preset relaxation parameter will be described.

[0154] Please Figure 6 Continue to refer to Figure 9 Based on some or all of the nth position information from the false alarms, a number of designated standard areas 111 are obtained from a number of standard areas 101, wherein the designated standard areas 111 contain position information corresponding to the nth position information from the false alarms.

[0155] Specifically, the corresponding location information is used to locate the position in the standard template image 100 to: the nth false alarm location information in cell region 300. n The location is determined. Based on this, the standard area 101 in the standard template image 100 that includes this location is the corresponding designated standard area 111.

[0156] It is important to understand that since the purpose of expanding the first deviation range is to ensure that the false alarm rate of defects in the lead frame 200 is below the preset false alarm rate when detecting defects, it is not necessary to expand the first deviation range for all false alarms of the nth defect. Instead, it can be selected for some or all of them based on the actual situation. In other words, depending on the actual situation, several designated standard areas 111 can be obtained based on the partial nth position information of the false alarms; or several designated standard areas 111 can be obtained based on all nth position information of the false alarms.

[0157] Please continue to refer to this. Figure 9Based on the area relaxation parameter corresponding to each specified standard area 111, the area deviation range of the (n-1)th time corresponding to each specified standard area 111 is expanded to generate the area deviation range of the nth time corresponding to each specified standard area 111.

[0158] Therefore, all the deviation parameters of the nth iteration constitute the first deviation parameter of the nth iteration.

[0159] It should be noted that, according to the area relaxation parameter corresponding to each specified standard area 111, expanding the area deviation range of the (n-1)th time corresponding to each specified standard area 111 means: expanding the area deviation range of the (n-1)th time as a whole.

[0160] It should be understood that regardless of whether any specified standard area 111 contains one or more corresponding location information, the range of the area deviation corresponding to the (n-1)th time will be expanded by one deviation range.

[0161] By expanding the range of the deviation for the (n-1)th iteration as a whole, the efficiency of iterative debugging of the deviation range can be better balanced.

[0162] In another embodiment, the preset relaxation parameters include: a plurality of regional relaxation parameters corresponding one-to-one with a plurality of standard regions 101; and individual relaxation parameters for partially expanding the range of deviation of the regions. For a detailed explanation of the regional relaxation parameters and the individual relaxation parameters, please refer to the foregoing related descriptions, which will not be repeated here.

[0163] In yet another embodiment, the nth defect data corresponding to any nth defect includes: the corresponding nth location information and the nth defect image 302. n For the nth position information and the nth defect image 302 n For a detailed explanation, please refer to the aforementioned relevant instructions, which will not be repeated here.

[0164] The method for expanding the first deviation range of the (n-1)th time based on the nth defect data corresponding to the nth falsely reported defects and the preset relaxation parameter includes: obtaining the corresponding nth falsely reported defect types based on the nth falsely reported defect data; and based on the nth falsely reported location information and the nth falsely reported defect images 302... n And several nth false positive defect types, in several false positive defect images 302 n Several specified nth defect images (not shown) are obtained; based on the nth false alarm defect type corresponding to any specified nth defect image, it is determined that the range of the area deviation corresponding to the standard area 101 corresponding to the specified nth defect image is expanded according to the area relaxation parameter or the individual relaxation parameter.

[0165] Specifically, the nth false alarm defect type is the defect type of the nth false alarm defect.

[0166] The nth false alarm defect type includes: outline reflection of stamped sheet, dent reflection of dented product, sheet texture, non-straight outline, color difference of surface material or semi-etched area.

[0167] It is important to understand that since the purpose of expanding the first deviation range is to ensure that the false alarm rate of defects in the lead frame 200 is below a preset false alarm rate when detecting defects, it is not necessary to expand the first deviation range for all falsely reported nth defects. Instead, it can be selected for some or all of them based on the actual situation. In other words, depending on the actual situation, the first deviation range can be expanded for all falsely reported nth defect images 302. n Select a portion or all of the data to proceed with the subsequent steps of expanding the deviation range.

[0168] The specified nth defective image is a plurality of nth defective images 302 n Image 302 of the nth defect selected n .

[0169] Since the nth false alarm defect type corresponding to any specified nth defect image is used, the region deviation range corresponding to the standard region 101 of the specified nth defect image is expanded according to the region widening parameter or the individual widening parameter. Therefore, during a single deviation range iteration adjustment process, based on the characteristics of the actual false alarm defect type, it is possible to flexibly choose to adjust the overall deviation range of the corresponding standard region 101 through the region widening parameter, or to choose to adjust the deviation range of a more local area within the range of the corresponding standard region 101 through the individual widening parameter.

[0170] It should be noted that during a deviation range iteration debugging process, based on the characteristics of the defect types of the actual false alarms, the deviation range of the (n-1)th iteration corresponding to the standard area 101 corresponding to all specified n-th defect images can be expanded according to the region relaxation parameter; or, the deviation range of the (n-1)th iteration corresponding to the standard area 101 corresponding to all specified n-th defect images can be expanded according to the individual relaxation parameter; or, the deviation range of the (n-1)th iteration corresponding to the standard area 101 corresponding to a portion of specified n-th defect images can be expanded according to the region relaxation parameter, and the deviation range of the (n-1)th iteration corresponding to the standard area 101 corresponding to the remaining portion of specified n-th defect images (excluding the portion of specified n-th defect images) can be expanded according to the individual relaxation parameter.

[0171] In another embodiment, the method for expanding the (n-1)th region deviation range corresponding to the standard region 101 corresponding to the arbitrary specified nth defect image according to the region relaxation parameter includes: obtaining a specified standard region from a plurality of standard regions 101 according to the nth position information corresponding to the arbitrary specified nth defect image, wherein the specified standard region has position information corresponding to the nth position information; and expanding the (n-1)th region deviation range corresponding to the specified standard region according to the region relaxation parameter corresponding to the specified standard region.

[0172] For an explanation of each step in the method of expanding the range of the (n-1)th region deviation corresponding to the standard region 101 corresponding to any specified nth defect image according to the region relaxation parameter, please refer to the relevant description in the aforementioned other embodiment, which will not be repeated here.

[0173] In another embodiment, the method of expanding the region deviation range corresponding to the (n-1)th time of the standard region 101 corresponding to the arbitrary specified nth defective image according to the region widening parameter includes: based on the arbitrary specified nth defective image and the corresponding nth position information, in the nth source image 301 n The process involves obtaining the nth defect region containing the arbitrary specified nth defect image; obtaining the corresponding nth defect region in several standard regions based on the nth defect region, and the (n-1)th region deviation range corresponding to the standard region 101 where the nth defect region is located includes: the (n-1)th sub-region deviation range corresponding to the nth defect region; and expanding the (n-1)th sub-region deviation range corresponding to the nth defect region based on the individual relaxation parameter.

[0174] In another embodiment, several false alarm reception steps can be performed, and in each false alarm reception step, the sub-region deviation range corresponding to the n-1th defective region is expanded, or, according to the individual relaxation parameter, the sub-region deviation range corresponding to the n-1th defective region is expanded in a single step.

[0175] Similarly, for a detailed explanation of each step in the method of expanding the range of the (n-1)th region deviation corresponding to the standard region 101 corresponding to the arbitrarily specified nth defect image according to the individual relaxation parameter, please refer to the foregoing. Figures 7 to 8 The relevant descriptions of the embodiments shown will not be repeated here.

[0176] While the present invention has been disclosed above, it is not limited thereto. Any person skilled in the art can make various modifications and alterations without departing from the spirit and scope of the invention; therefore, the scope of protection of the present invention should be determined by the scope defined in the claims.

Claims

1. A method for detecting defects in a lead frame, characterized in that, include: Provide a standard template image, an initial deviation range, and m lead frames, where m is a natural number; The method for obtaining the first deviation range by performing several iterations of deviation range debugging based on the standard template image, the initial deviation range, and m lead frames, and the method for the nth deviation range iteration debugging includes: Based on the nth lead frame, obtain the nth source image corresponding to the standard template image; The nth defect is detected based on the nth source image, the standard template image, and the first deviation range of the (n-1)th time, and the nth defect data corresponding to each nth defect is obtained. When n=1, the first deviation range of the (n-1)th time is the initial deviation range. Based on several nth defect data, the nth defect false alarm rate is obtained, where the nth defect false alarm rate refers to the proportion of the number of falsely reported nth defects in the total number of all nth defects detected. When the false alarm rate of the nth defect is greater than the preset false alarm rate, the first deviation range of the (n-1)th time is expanded to generate the first deviation range of the nth time based on the data of the nth defect corresponding to the false alarms and the preset relaxation parameter.

2. The defect detection method for lead frame as described in claim 1, characterized in that, The method for iterative debugging of the nth deviation range further includes: when the nth defect false alarm rate is less than or equal to the preset false alarm rate, the first deviation range of the (n-1)th iteration is used as the first deviation range of the nth iteration.

3. The defect detection method for lead frames as described in claim 2, characterized in that, The method for iterative debugging of the nth deviation range further includes: when the nth defect false alarm rate is less than or equal to the preset false alarm rate, the deviation range iterative debugging is terminated after generating the first deviation range for the nth time.

4. The defect detection method for lead frames as described in claim 1 or 2, characterized in that, The method for the nth deviation range iteration debugging further includes: when the nth deviation range iteration debugging reaches a preset number, after generating the first deviation range for the nth time, the deviation range iteration debugging is terminated.

5. The defect detection method for lead frames as described in claim 1, characterized in that, The standard template image includes several independent standard regions, and the first deviation range of the (n-1)th time includes several region deviation ranges that correspond one-to-one with the several standard regions.

6. The defect detection method for lead frames as described in claim 5, characterized in that, The aforementioned standard areas include at least one of the following: electroplating and critical areas, semi-etched areas, functional areas, low-threshold areas, and non-functional areas.

7. The defect detection method for lead frames as described in claim 5, characterized in that, The preset relaxation parameters include several region relaxation parameters that correspond one-to-one with several standard regions.

8. The defect detection method for lead frames as described in claim 7, characterized in that, The data for the nth defect corresponding to any nth defect includes at least the nth position information; The method for expanding the first deviation range of the (n-1)th time based on the nth defect data corresponding to the nth falsely reported defects and the preset relaxation parameter includes: obtaining a number of designated standard areas in a number of standard areas based on part or all of the nth falsely reported position information, wherein the designated standard areas have position information corresponding to the nth falsely reported position information; expanding the area deviation range of the (n-1)th time corresponding to each designated standard area based on the area relaxation parameter corresponding to each designated standard area, so as to generate the area deviation range of the nth time corresponding to each designated standard area.

9. The defect detection method for lead frames as described in claim 5, characterized in that, The preset relaxation parameters include individual relaxation parameters used to partially expand the range of deviation in the region; The data for the nth defect corresponding to any nth defect includes: the corresponding nth location information and the nth defect image; The method for expanding the first deviation range of the (n-1)th time based on the nth defect data corresponding to the nth defect and the preset relaxation parameter includes: obtaining a number of nth defect areas containing the nth defect images in the nth source image based on the nth location information and part or all of the nth defect images; obtaining a number of corresponding nth defect regions in a number of standard regions based on the nth defect regions, and the (n-1)th region deviation range corresponding to any nth defect region includes: the (n-1)th sub-region deviation range corresponding to the any nth defect region; expanding the (n-1)th sub-region deviation range corresponding to each nth defect region based on the individual relaxation parameter to generate the corresponding nth region deviation range.

10. The defect detection method for lead frames as described in claim 9, characterized in that, The method for expanding the sub-region deviation range of each nth defective region corresponding to the (n-1)th time based on the individual relaxation parameter includes: performing a number of false alarm reception steps, and the method for each false alarm reception step includes: expanding the sub-region deviation range of one nth defective region corresponding to the (n-1)th time based on the individual relaxation parameter.

11. The defect detection method for lead frames as described in claim 9, characterized in that, The method for expanding the sub-region deviation range of each nth defective region corresponding to the (n-1)th time according to the individual relaxation parameter includes: expanding the sub-region deviation range of all nth defective regions corresponding to the (n-1)th time in a single step according to the individual relaxation parameter.

12. The defect detection method for lead frames as described in claim 7, characterized in that, The preset relaxation parameters also include individual relaxation parameters for partially expanding the range of deviation in the region; The data for the nth defect corresponding to any nth defect includes: the corresponding nth location information and the nth defect image; The method for expanding the first deviation range of the (n-1)th time based on the nth defect data corresponding to the nth falsely reported defects and the preset relaxation parameter includes: obtaining the corresponding nth falsely reported defect types based on the nth falsely reported defect data; obtaining the nth specified defect images from the nth falsely reported defect images based on the nth falsely reported location information, the nth falsely reported defect images, and the nth falsely reported defect types; and determining, based on the nth falsely reported defect type corresponding to any specified nth defect image, to expand the (n-1)th time area deviation range corresponding to the standard area corresponding to the arbitrary specified nth defect image according to the area relaxation parameter or the individual relaxation parameter.

13. The defect detection method for lead frames as described in claim 12, characterized in that, The method for expanding the range of the (n-1)th region deviation corresponding to the standard region corresponding to any specified n-th defective image according to the region relaxation parameter includes: obtaining a specified standard region from several standard regions according to the n-th position information corresponding to the specified n-th defective image, wherein the specified standard region has position information corresponding to the n-th position information; and expanding the range of the (n-1)th region deviation corresponding to the specified standard region according to the region relaxation parameter corresponding to the specified standard region.

14. The defect detection method for lead frames as described in claim 12, characterized in that, The method for expanding the (n-1)th region deviation range corresponding to the standard region corresponding to the arbitrary specified nth defective image according to the individual relaxation parameter includes: obtaining the nth defective region containing the arbitrary specified nth defective image in the nth source image according to the arbitrary specified nth defective image and the corresponding nth position information; obtaining the corresponding nth defective region in a plurality of standard regions according to the nth defective region, and the (n-1)th region deviation range corresponding to the standard region where the nth defective region is located includes: the (n-1)th sub-region deviation range corresponding to the nth defective region; expanding the (n-1)th sub-region deviation range corresponding to the nth defective region according to the individual relaxation parameter.

15. The defect detection method for lead frames as described in claims 7, 8, 12, 13 or 14, characterized in that, The region widening parameters include at least one of the following: region diagonal length widening parameter, region area widening parameter, and region contrast widening parameter.

16. The defect detection method for lead frames as described in any one of claims 9 to 14, characterized in that, The individual relaxation parameters include at least one of the individual diagonal length relaxation parameters, individual area relaxation parameters, and individual contrast relaxation parameters.

17. The defect detection method for lead frames as described in any one of claims 8 to 14, characterized in that, The nth source image includes several repeating cell regions. The nth source image of each cell region corresponds to the standard template image. Each cell region has origin coordinates, and the nth position information is the origin coordinates or relative coordinates to the origin coordinates.

18. The defect detection method for lead frames as described in claim 17, characterized in that, The method for detecting the nth defect based on the nth source image, the standard template image, and the (n-1)th first deviation range includes: comparing the nth source image of each unit region with the standard template image and the (n-1)th first deviation range.

19. The defect detection method for lead frames as described in claim 1, characterized in that, The preset relaxation parameters can be either numerical or proportional.

Images