An abnormality tracing system and method for chip carrier tray breakage

Abstract

This invention provides an anomaly tracing system and method for chip carrier tray damage, belonging to the field of semiconductor chip testing technology. The system includes: a cabinet storage detection system for acquiring and sending six-sided images of the damaged tray to a tray damage tracing system; the tray damage tracing system compares the six-sided images of the damaged tray with the six-sided images of the corresponding normal tray. If the deviation of any region on any face is greater than or equal to a preset value, the region is marked as a candidate damage region, and feature point coordinates are extracted to construct a polygon; the smallest bounding polygon covering all vertices of the polygon is compared with the rectangular region of the corresponding contact position of the mechanism. If there is a spatial intersection between the two, the mechanism is determined to be a highly correlated suspected mechanism causing the tray damage. This invention achieves automated identification and accurate correlation between damaged areas and interference points.

Description

Technical Field

[0001] This invention relates to the field of semiconductor chip testing technology, and in particular to an anomaly tracing system and method for chip carrier tray damage. Background Technology

[0002] In the mass production of semiconductor chips, trays (chip carriers) serve as the core transport medium for chips throughout the baking, testing, sorting, and packaging processes, requiring compatibility with contact operations at multiple stations and across various types of mechanisms. Trays must maintain their structural integrity during use; damage such as missing end tabs, cavity deformation, or side cracks can easily lead to scratches, positioning deviations, and even equipment jams during transport. Therefore, when tray damage is discovered, the source of the damage must be quickly located, and the stations and specific mechanisms in contact with the tray must be investigated to prevent mass breakage.

[0003] The current mainstream technology for tracing tray damage in the industry is as follows: When tray damage is discovered, engineers use their past experience to screen suspected sites and visually compare the damaged location with the contact area of ​​the mechanism. If there is no manpower available to investigate when tray damage occurs, production line personnel will temporarily store some of the damaged trays, manually fill out paper records (including marking the site where the damage occurred, information on products in operation, etc.), and place the physical trays on a shelf for unified management, awaiting investigation.

[0004] Clearly, the current solution suffers from drawbacks such as low efficiency of manual screening, large errors in experience-based judgment, and lack of data traceability. Summary of the Invention

[0005] This invention provides an anomaly tracing system and method for chip carrier tray damage, addressing the problems of low efficiency in manual investigation, large errors in experience-based judgment, and lack of data traceability in existing solutions. The technical solution is as follows: In a first aspect, embodiments of the present invention provide an anomaly tracing system for chip carrier tray damage, including a cabinet storage detection system and a tray damage tracing system; The cabinet storage detection system is used to collect six-sided images of the damaged tray and send the collected six-sided images to the tray damage tracing system. The Tray disk damage tracing system integrates a normal Tray disk database and a mechanism contact surface database. The normal Tray disk database includes the standard coordinates of the six sides of a normal Tray disk and the standard coordinates of the key structures on each side. The mechanism contact surface database includes the coordinates of the mechanism contact points that are in contact with any side of a normal Tray disk. The Tray damage tracing system receives six-sided images of the damaged Tray sent by the cabinet storage detection system, compares them with the six-sided images of the corresponding normal Tray stored in the normal Tray database, and marks the area as a candidate damage area if the deviation of any area on any face is greater than or equal to a preset value. The system then extracts the feature point coordinates of the candidate damage area and constructs a polygon based on these coordinates. It determines the smallest bounding polygon covering all vertices of the polygon and compares its coordinates with the rectangular area of ​​the corresponding mechanism contact position stored in the mechanism contact surface database. If there is a spatial intersection between the smallest bounding polygon and the rectangular area of ​​the mechanism contact position, the mechanism is identified as a highly correlated suspected mechanism causing the Tray damage.

[0006] Optionally, the tray damage tracing system aligns the six-sided images of the damaged tray with the six-sided images of the corresponding normal trays stored in the normal tray database. If the deviation of any region on any face is greater than or equal to a preset value, that region is marked as a candidate damaged region, including: The Tray damage tracing system performs Gaussian filtering on the six-sided images of the damaged Tray to eliminate noise and binarizes the images to highlight their contours. The Tray Disk Damage Tracing System retrieves the outline templates of the six sides of a normal Tray Disk with the corresponding PN code from the normal Tray Disk Database using the unique identifier PN code of the Tray Disk. It then uses an iterative nearest-point algorithm to align the image outlines with the outline templates. If the alignment deviation of any region on any face is greater than or equal to a preset value, then that region is marked as a candidate region for damage.

[0007] Optionally, the tray damage tracing system extracts the feature point coordinates of the candidate damage area, including: The Tray disk damage tracing system performs morphological dilation on the damaged candidate region and extracts the coordinates of feature points in the damaged candidate region according to preset rules. The preset rules include: if the rules are broken, extract the coordinates of the diagonal vertices; if the rules are broken irregularly, extract the feature point set.

[0008] Optionally, the Tray disc damage tracing system compares the coordinates of the minimum circumscribed polygon with the rectangular regions of the corresponding mechanism contact positions stored in the mechanism contact surface database. If there is a spatial intersection between the minimum circumscribed polygon and the rectangular regions of the mechanism contact positions, then the mechanism is determined to be a highly correlated suspected mechanism causing the Tray disc damage, including: The Tray damage tracing system determines whether the rectangular area where the smallest circumscribed polygon contacts the mechanism has overlapping regions in both the X and Y axes. If there are overlapping regions in both directions, it is determined that there is a spatial intersection between the minimum circumscribed polygon and the rectangular region at the contact position of the mechanism. This organization has been identified as a highly suspected entity linked to the damage to the Tray disc.

[0009] Optionally, it also includes: When the Tray disk damage tracing system determines that the rectangular regions of the contact positions of the minimum circumscribed polygon and multiple mechanisms overlap in both the X and Y axes, the multiple mechanisms are sorted according to a preset association priority rule. The preset association priority rules include: the higher the percentage of intersection area, the higher the priority; and / or, the higher the contribution of the institution's historical damage, the higher the priority.

[0010] Optionally, the tray damage tracing system is further used to: automatically generate a standardized tracing report; the standardized tracing report includes at least basic information, damage details, and tracing conclusions; wherein: The basic information includes at least: Tray disk PN code, supplier, product model, discovery site, and discovery time; The damage details include at least: the damaged contact surface, the coordinates of the damaged area, and the damage type; The source tracing conclusions include at least: a ranking of highly related suspected organizations.

[0011] Optionally, the cabinet storage detection system includes: a cabinet unlocking unit, a six-sided image acquisition unit, and a classification and storage unit; The sorting and storage unit is equipped with a testing area, a tested area, and a temporary storage area; The cabinet unlocking unit is used to compare the externally input Tray information with the Tray information stored in the cabinet storage detection system. Only when the comparison information matches, the cabinet unlocking unit controls the unlocking, allowing the damaged Tray to be placed into the detection area. The six-sided image acquisition unit is used to acquire images of the six sides of the damaged tray.

[0012] Optionally, the detection cavity of the six-sided image acquisition unit is symmetrically equipped with six sets of industrial cameras and strip LED light sources to achieve detection of the entire surface of the damaged tray without blind spots.

[0013] Optionally, the contact surface database of the mechanism adopts a six-level structure for storage: product model-Tray PN-site-contact mechanism-contact surface-contact point coordinates.

[0014] Secondly, embodiments of the present invention provide an anomaly tracing method for chip carrier tray damage, applied to the anomaly tracing system for chip carrier tray damage described in the first aspect, the anomaly tracing method comprising: Acquire images of all six sides of the damaged tray; The six-sided images of the damaged tray are compared with the six-sided images of the corresponding normal trays stored in the normal tray database. If the deviation of any region on any face is greater than or equal to a preset value, the region is marked as a candidate region for damage. The coordinates of the feature points of the candidate region for damage are extracted, and a polygon is constructed based on the extracted feature point coordinates. Determine the smallest bounding polygon that covers all vertices of the polygon, and compare its coordinates with the rectangular region corresponding to the mechanism contact position stored in the mechanism contact surface database; If there is a spatial intersection between the minimum circumscribed polygon and the rectangular area at the contact position of the mechanism, then the mechanism is determined to be a highly correlated suspected mechanism causing the damage to the Tray disk.

[0015] The beneficial effects of the above-mentioned technical solution of the present invention are: In the anomaly tracing system and method for chip carrier tray damage provided in this embodiment of the invention, the cabinet storage detection system acquires six-sided images of the damaged tray and sends the acquired six-sided images to the tray damage tracing system. The tray damage tracing system integrates a normal tray database and a mechanism contact surface database. The normal tray database includes the standard coordinates of the six sides of a normal tray and the standard coordinates of key structures on each side. The mechanism contact surface database includes the coordinates of mechanism contact points that contact any side of a normal tray. Specifically, the tray damage tracing system is used to receive the six-sided images of the damaged tray sent by the cabinet storage detection system. The image is compared with the six-sided images of the corresponding normal tray stored in the normal tray database. If the deviation of any region on any face is greater than or equal to a preset value, the region is marked as a candidate damage region. The feature point coordinates of the candidate damage region are extracted, and a polygon is constructed based on the extracted feature point coordinates. The smallest bounding polygon covering all vertices of the polygon is determined, and its coordinates are compared with the rectangular region of the corresponding mechanism contact position stored in the mechanism contact surface database. If there is a spatial intersection between the smallest bounding polygon and the rectangular region of the mechanism contact position, the mechanism is determined to be a highly correlated suspected mechanism causing tray damage. This embodiment of the invention achieves accurate correlation between the damaged area and the interference point through standardized two-dimensional coordinate intersection judgment logic, thereby realizing automated identification of the damage location and rapid matching of suspected sites. It effectively solves the problems of low efficiency, long time consumption, large error, and high misjudgment rate in the traceability of tray damage caused by manual investigation in the existing technology. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of an anomaly tracing system for chip carrier tray damage disclosed in an embodiment of the present invention; Figure 2 This is a schematic diagram illustrating the coordinate system definition of a normal Tray disk surface 1 in an embodiment of the present invention; Figure 3 This is a schematic diagram of the six-level structure storage in the mechanism contact surface database in an embodiment of the present invention; Figure 4 This is a schematic diagram showing the marked damaged areas in an embodiment of the present invention; Figure 5 This is a schematic diagram illustrating the determination of the intersection of damaged area areas in an embodiment of the present invention; Figure 6 This is a flowchart of an anomaly tracing method for chip carrier tray damage disclosed in an embodiment of the present invention. Detailed Implementation

[0017] To make the technical problems, technical solutions, and advantages of this invention clearer, a detailed description will be provided below in conjunction with the accompanying drawings and specific embodiments. In the following description, specific details such as particular configurations and components are provided merely to aid in a comprehensive understanding of the embodiments of this invention. Therefore, those skilled in the art should understand that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this invention. Furthermore, for clarity and brevity, descriptions of known functions and structures have been omitted.

[0018] It should be understood that the phrase "one embodiment" or "an embodiment" throughout the specification means that a specific feature, structure, or characteristic related to the embodiment is included in at least one embodiment of the invention. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification do not necessarily refer to the same embodiment. Furthermore, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments.

[0019] In various embodiments of the present invention, it should be understood that the sequence number of each process described below 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.

[0020] This invention, through a collaborative architecture between a cabinet storage and detection system and a tray damage tracing system, achieves closed-loop management of damaged trays, including targeted storage, full-surface detection, and precise traceability. For example... Figure 1 As shown, the anomaly tracing system for chip carrier tray damage provided in this embodiment of the invention includes: cabinet storage detection system 1 and tray damage tracing system 2.

[0021] The cabinet storage detection system 1 is used to collect six-sided images of the damaged tray and send the collected six-sided images to the tray damage tracing system 2.

[0022] In practical applications, a tray has six surfaces: front (surface 1), back (surface 2), left (surface 3), right (surface 4), top (surface 5), and bottom (surface 6). The cabinet storage detection system 1 can quickly (e.g., within 20 seconds) acquire images of all six surfaces of the damaged tray and send the acquired images to the tray damage tracing system 2.

[0023] Tray disk damage tracing system 2 integrates a normal Tray disk database 21 and a mechanism contact surface database 22. The normal Tray disk database 21 includes the standard coordinates of the six faces of a normal Tray disk and the standard coordinates of the key structures on each face. The mechanism contact surface database 22 includes the coordinates of the mechanism contact points that are in contact with any face of a normal Tray disk.

[0024] In this embodiment of the invention, the Tray Damage Tracing System 2 includes two types of standardized data, which are also the core supporting data of the entire system. These are the standard coordinates of the six faces of a normal Tray and the standard coordinates of key structures on each face, stored in the Normal Tray Database 21; and the coordinates of the contact points of mechanisms that are in contact with any face of a normal Tray, stored in the Mechanism Contact Surface Database 22. In practical applications, these core supporting data can be updated and maintained by administrators through a web backend.

[0025] Normal Tray Database 21: All data in Normal Tray Database 21 is uniquely associated with the Tray PN code. Normal Tray Database 21 establishes a six-sided standardized two-dimensional coordinate system—with the geometric center of the Tray as the "global origin," projected onto six surfaces to form the "local origin" for each surface. Each surface independently establishes X and Y axes (e.g., the X-axis along the length of the Tray, and the Y-axis along the width). Figure 2 As shown, for example, the coordinates of face 1 (front) of PN KNTF10245-234 have the local origin coinciding with the global origin, with the upper left corner coordinates (-68, 161) and the lower right corner coordinates (68, -161). It also stores the standard coordinates of key structures on this face (such as the auricle and cavity), such as auricle 1 region: X: -40-20mm, Y: 30-50mm.

[0026] Mechanism Contact Surface Database 22: This database stores the coordinates of contact points of mechanisms that contact any surface of the normal Tray disk. Only the contact point coordinates of mechanisms that contact the Tray disk are recorded and stored in the database. The contact point coordinates are mapped to a rectangular area of ​​the contact position in the coordinate system. Preferably, in this embodiment, the data in the mechanism contact surface database 22 can be stored using a six-level structure: product model-Tray PN-site-contact mechanism-contact surface-contact point coordinates, for example... Figure 3 As shown. Understandably, Figure 3 This is merely an example; other data structure storage methods, such as four-level or five-level storage, are also within the scope of protection of this invention.

[0027] The Tray Damage Tracing System 2 specifically receives six-sided images of a damaged Tray from the Cabinet Storage Detection System 1, compares them with the corresponding six-sided images of normal Trays stored in the Normal Tray Database 21, and marks any area on any face as a candidate damaged area if the deviation is greater than or equal to a preset value. It then extracts the feature point coordinates of the candidate damaged area and constructs a polygon based on these coordinates. Finally, it determines the smallest bounding polygon covering all vertices of the polygon and compares its coordinates with the rectangular area of ​​the corresponding mechanism contact position stored in the Mechanism Contact Surface Database 22. If there is a spatial intersection between the smallest bounding polygon and the rectangular area of ​​the mechanism contact position, the mechanism is identified as a highly correlated suspected mechanism causing the Tray damage.

[0028] In this embodiment of the invention, the Tray damage tracing system 2 mainly includes two core processing actions: 1. Damage identification and damage information collection and storage; 2. Data comparison enables automatic association between damaged areas and interference points of the mechanism.

[0029] The applicant will now provide a more detailed explanation of the two core processing steps mentioned above.

[0030] (I) Damage identification and damage information collection and storage In practical applications, the Tray damage tracing system 2 performs Gaussian filtering (e.g., 5×5 filter kernel) on the six sides of the received damaged Tray to eliminate noise, and binarizes the images (with a threshold set to 127, for example) to highlight the image contours. Using the unique identifier PN (part number) of the tray, the contour template of the six sides of the normal tray corresponding to the PN code is retrieved from the normal tray database 21. The ICP algorithm (Iterative Closest Point Algorithm) is used to align the contour of the binarized image with the contour template of the normal tray. If the alignment result of any region on any face deviates from the preset value (e.g., deviation ≥ 0.5 mm), then the region is marked as a candidate region for damage.

[0031] Furthermore, feature extraction is performed on the candidate damaged region. Specifically, the candidate damaged region is morphologically dilated (e.g., with a 3×3 dilation kernel), and then the coordinates of feature points in the candidate damaged region are extracted according to preset rules. The preset rules include: for regular damage, the coordinates of the diagonal vertices are extracted; for irregular damage, the set of feature points is extracted.

[0032] For example, regular damage, such as deformation of a rectangular cavity, is described using "coordinates of the diagonal vertices," while irregular damage, such as missing auricles, is described using "set of feature points."

[0033] Finally, polygons are constructed based on the extracted feature point coordinates.

[0034] For example, combined Figure 4 As shown, the damaged area b is an irregular deformation near the end ear of face 1. Five feature points are selected (clockwise order): Feature point 1 (leftmost point): (15, -68), Feature point 2 (upper turning point, local highest point of Y): (30, -40), Feature point 3 (rightmost point): (50, -50), Feature point 4 (lower turning point, local lowest point of Y): (53, -58), Feature point 5 (closed point, coinciding with feature point 1): (15, -68). Description of the damaged area: The outline is a polygon formed by connecting (15, -68) → (30, -40) → (50, -50) → (53, -58) → (15, -68).

[0035] It should be noted that in practical applications, a damaged tray database can be set up in the Tray Damage Tracing System 2. This damaged tray database is used to compare with the normal tray database to identify damage and to collect and store damage information.

[0036] (ii) Data comparison to achieve automatic association between damaged areas and mechanical interference points. Tray disc damage tracing system 2 employs a "two-dimensional coordinate intersection judgment" logic to automatically associate the damaged area with the mechanism interference point. Specifically, Tray disc damage tracing system 2 first retrieves the coordinates of all mechanism contact points corresponding to the surface from the mechanism contact surface database 22 based on "product model + damaged contact surface" to determine the rectangular area of ​​the mechanism contact position.

[0037] Furthermore, the smallest bounding polygon of the damaged area is determined. This smallest bounding polygon is the smallest polygon that covers all vertices of the polygon, and its coordinates are compared with the rectangular area of ​​the determined contact position of the mechanism.

[0038] If there is a spatial intersection between the smallest circumscribed polygon and the rectangular area at the contact point of the mechanism, then the mechanism is identified as a highly correlated suspected mechanism that caused the damage to the Tray disk.

[0039] In this embodiment of the invention, when performing coordinate comparison, the Tray disc damage tracing system 2 specifically determines whether the rectangular area where the smallest circumscribed polygon contacts the mechanism overlaps in both the X and Y axes. If overlap exists in both directions, it is determined that there is a spatial intersection between the smallest circumscribed polygon and the rectangular area where the mechanism contacts the mechanism, thus identifying the mechanism as a highly correlated suspected mechanism causing the Tray disc damage. In practical applications, "site + mechanism" can be marked as a highly correlated suspected object.

[0040] More preferably, when the Tray disc damage tracing system 2 determines that the rectangular areas of the contact positions of the smallest circumscribed polygon and multiple mechanisms simultaneously overlap in both the X and Y axes, it can sort the multiple mechanisms according to preset association priority rules. These preset association priority rules include: a higher percentage of intersection area corresponds to a higher priority; and / or, a higher historical damage contribution from the mechanism corresponds to a higher priority. The percentage of intersection area is calculated as intersection area ÷ damaged area × 100%. The historical damage contribution from the mechanism is calculated as, for example, the number of damages caused by the mechanism in the past 3 months / total number of damages × 100%.

[0041] For example, combined Figure 5 As shown, Figure 5 The yellow area in the middle represents the damaged area, and the blue area represents the contact point between the mechanism and the tray (i.e., the rectangular area where the mechanism contacts the tray). In the left image, the two areas do not intersect, indicating that the damage at this location is unrelated to the mechanism. In the right image, the two areas intersect, and calculations show that the intersection area accounts for 85%, confirming that the mechanism at this location is highly suspected of causing the tray damage.

[0042] It should be noted that in practical applications, a comparison and matching database can be set up in Tray disk damage tracing system 2 to achieve automatic association between the damaged area and the interference point of the mechanism.

[0043] As a preferred implementation of this invention, the Tray disc damage tracing system 2 can also be used to: automatically generate a standardized tracing report; the standardized tracing report includes at least basic information, damage details, and tracing conclusions; wherein: Basic information should include at least: Tray PN code, supplier (associated with the PN code database), product model, discovery site, and discovery time; Damage details should include at least: damaged contact surface (e.g., surface 1), damaged area coordinates (feature point set / circumscribed rectangle), and damage type (e.g., "terminal ear missing" or "cavity deformation"). The conclusions of the source tracing should include at least: a ranking of highly correlated suspected institutions (e.g., including the percentage of overlapping areas and the contribution of historical damage to the institutions).

[0044] Tray disk damage tracing system 2 can also output an attachment, which includes a comparison diagram of the damaged area and the interference point of the mechanism (with the intersection part marked).

[0045] It should be noted that in practical applications, a report generation database can be set up in the Tray Disk Damage Tracing System 2 to automatically generate standardized traceability reports.

[0046] Furthermore, the applicant provides a detailed description of the cabinet storage detection system 1 in the embodiments of the present invention.

[0047] The cabinet storage detection system 1, serving as the physical carrier for "six-sided image acquisition + temporary storage" of damaged trays, ensures consistent detection through modular design. Its core comprises three functional units: a cabinet unlocking unit, a six-sided image acquisition unit, and a categorized storage unit; among which: The categorized storage unit includes a testing area, a tested area, and a temporary storage area. The cabinet unlocking unit is used to compare the externally input tray information with the tray information stored in the cabinet storage detection system. Only when the comparison information matches will the cabinet unlocking unit control the unlocking, allowing the damaged tray to be placed into the detection area. The six-sided image acquisition unit is used to acquire six-sided images of the damaged tray.

[0048] In practical applications, the detection chamber is divided into three removable partitions to form a detection area, a detection area, and a temporary storage area.

[0049] The cabinet unlocking unit can be equipped with an OCR recognition module and an input module. The input module allows employees to manually input tray information, while the OCR recognition module automatically acquires the tray information through scanning or other methods. In practical applications, when the tray information manually entered by the employee, the tray information acquired through scanning, and the tray information stored in the system match, the cabinet unlocking unit unlocks the tray, allowing the employee to place the damaged tray into the inspection area.

[0050] Six-sided image acquisition unit: corresponding to the six surfaces of the tray (surface 1: front, surface 2: back, surface 3: left, surface 4: right, surface 5: top, surface 6: bottom), six sets of "industrial cameras + strip LED light sources" are symmetrically deployed on the inner wall of the detection cavity, which can realize the detection of the entire surface of the damaged tray without blind spots.

[0051] The anomaly tracing system for chip carrier tray damage provided in this invention achieves automated identification of the damage location and rapid matching of suspected sites, effectively solving the problems of low efficiency and long time consumption in traceability of tray damage caused by manual investigation in existing technologies. Through standardized two-dimensional coordinate intersection judgment logic, it achieves accurate association between the damaged area and the interference point, effectively solving the problems of large traceability errors and high misjudgment rate caused by experience-based judgment. By constructing a full-chain database of "Tray PN code - product model - damage location - institutional interference point - traceability conclusion", it achieves electronic storage of information and multi-dimensional retrieval, effectively solving the problems of scattered and untraceable damage information records.

[0052] Based on the anomaly tracing system for chip carrier tray damage provided in the foregoing embodiments of the present invention, the present invention also provides an anomaly tracing method for chip carrier tray damage, which is specifically applied to the anomaly tracing system for chip carrier tray damage described above.

[0053] To facilitate a better understanding of the embodiments of the present invention, the applicant uses the example of "a PHX FP9 product, Tray PNKNTF10245-234, showing damage to surface 1 at functional test station 1" to describe the anomaly tracing method for chip carrier tray damage provided by the embodiments of the present invention. Figure 6 As shown, the method includes: Step 101: Acquire six-sided images of the damaged tray.

[0054] In practical application, employees bring the damaged tray to the cabinet storage detection system, input "Product Model: PHX FP9" on the touchscreen, and scan the PN code KNTF10245-234 using a barcode scanner. The system verifies the PN code's validity, unlocks the door, and the employee places the damaged tray along the positioning protrusion on the platform into the inspection area, then closes the door. The infrared sensors within the system confirm the tray's placement and automatically activate the six-sided cameras and light source, completing the acquisition of six images within 20 seconds.

[0055] Step 2: Compare the six-sided images of the damaged tray with the corresponding six-sided images of normal trays stored in the normal tray database.

[0056] First, the six-sided image is preprocessed, including filtering and binarization, to highlight the image contours. Then, the contour template of a normal tray disk (PNKNTF10245-234) is used, and the two are aligned using the ICP algorithm.

[0057] Step 103: If the deviation of any region on any face is greater than or equal to a preset value, then the region is marked as a candidate region for damage. The feature point coordinates of the candidate region for damage are extracted, and a polygon is constructed based on the extracted feature point coordinates.

[0058] During alignment, an area with a deviation ≥0.5mm was found on face 1 (areas less than 0.5mm were ignored). This area was marked as a candidate area for damage, and the feature point coordinates of the candidate area were extracted as (15, -68) → (30, -40) → (50, -50) → (53, -58) → (15, -68). A polygon was constructed based on these feature point coordinates, and the damage type was determined to be "irregular missing end ear of face 1". The results were then stored in the damaged Tray disk database.

[0059] Step 104: Determine the smallest circumscribed polygon that covers all vertices of the polygon, and compare its coordinates with the rectangular region of the corresponding mechanism contact position stored in the mechanism contact surface database.

[0060] Step 105: If there is a spatial intersection between the minimum circumscribed polygon and the rectangular area at the contact position of the mechanism, then the mechanism is determined to be a highly correlated suspected mechanism that caused the damage to the Tray disk.

[0061] Based on "Product PHX FP9+ Damaged Surface 1", retrieve the coordinates of all interference points (cylinder Y1, pressure bar Y5, support block S1) of surface 1 from the mechanism contact surface database, and determine the smallest circumscribed polygon covering all vertices of the polygon. Then, determine whether spatial intersection exists by comparing the coordinates. Specifically: Cylinder Y1 (rectangular area of ​​contact position: X20~40mm, Y30~50mm): no intersection with the smallest circumscribed polygon → excluded; Pressure strip Y5 (rectangular area at contact position: X5~15mm, Y70~90mm): no intersection with the smallest circumscribed polygon → excluded; Support block S1 (rectangular area at contact position: X10~55mm, Y-70~-35mm): the area of ​​intersection with the smallest circumscribed polygon accounts for 85% → high correlation.

[0062] As a preferred embodiment of the present invention, the order is as follows: support block S1 is the only highly associated object, and has caused similar damage twice in the past 3 months, with a historical contribution of 40%, so it is ranked as the first priority.

[0063] Furthermore, as a preferred embodiment of the present invention, the method may further include: generating and outputting a traceability report.

[0064] The system automatically generates a traceability report, which is synchronized to the touchscreen of the cabinet storage inspection system and / or the engineer's APP within 5 seconds. The engineer reviews the report, confirms support block S1 as a highly relevant suspected component, and issues a maintenance work order. Maintenance personnel disassemble support block S1, find surface wear protrusions (matching the location of the damage), repair them, and enter the maintenance record. The employee removes the tray from the inspected area, determines that "the end ear is missing and cannot be repaired," enters "Disposal Result: Scrap," and the system updates the final status of PN KNTF10245-234, completing the closed loop.

[0065] Compared with existing technologies, the anomaly tracing system and method for chip carrier tray damage provided in this invention achieves breakthroughs in three major dimensions: efficiency, accuracy, and traceability. 1. Significant improvement in traceability efficiency: Automated image acquisition (20 seconds) + intelligent comparison (50 seconds) + report generation (10 seconds) reduces the traceability time for a single batch of damaged trays from the current 2-4 hours to less than 5 minutes, meeting the production line's control requirement of "rectification within 2 hours"; 2. Significantly improved traceability accuracy: Standardized two-dimensional coordinate intersection judgment replaces experience-based judgment, the accuracy of correlation between damage and mechanism is ≥99.8%, the misjudgment rate is reduced to below 0.2%, avoiding ineffective disassembly and maintenance (such as misremoving a locating pin that has no problem), and saving 6-8 hours of maintenance time per batch; 3. Full-chain data traceability: Construct a data closed loop of "Tray disk PN code - product - damage - institution - maintenance", support multi-dimensional retrieval (such as "PN KNTF10245-234 historical damage", "number of times the functional test station support block S1 is damaged"), and provide data support for the optimization of the institution's maintenance cycle (such as shortening the support block maintenance interval to 20 days).

[0066] It should be noted that in the system and method of the present invention, it is obvious that the components or steps can be decomposed and / or recombined. These decompositions and / or recombinations should be considered equivalent solutions of the present invention. Furthermore, the steps performing the above series of processes can naturally be executed in the order described, but are not necessarily required to be executed in chronological order; some steps can be executed in parallel or independently of each other. Those skilled in the art will understand that all or any step or component of the method and apparatus of the present invention can be implemented in any computing device (including processors, storage media, etc.) or network of computing devices, in hardware, firmware, software, or a combination thereof. This is something that those skilled in the art can achieve by using their basic programming skills after reading the description of the present invention.

[0067] Therefore, the object of the present invention can also be achieved by running a program or a set of programs on any computing device. The computing device can be a known general-purpose device. Therefore, the object of the present invention can also be achieved simply by providing a program product containing program code implementing the method or apparatus. That is, such a program product also constitutes the present invention, and the storage medium storing such a program product also constitutes the present invention. Obviously, the storage medium can be any known storage medium or any storage medium developed in the future. It should also be noted that in the apparatus and method of the present invention, it is obvious that the components or steps can be decomposed and / or recombined. These decompositions and / or recombinations should be considered equivalent to the present invention. Furthermore, the steps performing the above series of processes can naturally be performed in the order described, but are not necessarily required to be performed in chronological order. Some steps can be performed in parallel or independently of each other.

[0068] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. An abnormality traceability system for a chip carrier tray breakage, characterized by, This includes a cabinet storage detection system and a tray damage tracing system; The cabinet storage detection system is used to collect six-sided images of the damaged tray and send the collected six-sided images to the tray damage tracing system. The Tray disk damage tracing system integrates a normal Tray disk database and a mechanism contact surface database. The normal Tray disk database includes the standard coordinates of the six sides of a normal Tray disk and the standard coordinates of the key structures on each side. The mechanism contact surface database includes the coordinates of the mechanism contact points that are in contact with any side of a normal Tray disk. The Tray damage tracing system receives six-sided images of the damaged Tray sent by the cabinet storage detection system, compares them with the six-sided images of the corresponding normal Tray stored in the normal Tray database, and marks the area as a candidate damage area if the deviation of any area on any face is greater than or equal to a preset value. The system then extracts the feature point coordinates of the candidate damage area and constructs a polygon based on these coordinates. It determines the smallest bounding polygon covering all vertices of the polygon and compares its coordinates with the rectangular area of ​​the corresponding mechanism contact position stored in the mechanism contact surface database. If there is a spatial intersection between the smallest bounding polygon and the rectangular area of ​​the mechanism contact position, the mechanism is identified as a highly correlated suspected mechanism causing the Tray damage.

2. The anomaly tracing system according to claim 1, characterized in that, The tray damage tracing system aligns the six-sided images of the damaged tray with the corresponding six-sided images of normal trays stored in the normal tray database. If the deviation of any region on any face is greater than or equal to a preset value, that region is marked as a candidate damage region, including: The Tray damage tracing system performs Gaussian filtering on the six-sided images of the damaged Tray to eliminate noise and binarizes the images to highlight their contours. The Tray Disk Damage Tracing System retrieves the outline templates of the six sides of a normal Tray Disk with the corresponding PN code from the normal Tray Disk Database using the unique identifier PN code of the Tray Disk. It then uses an iterative nearest-point algorithm to align the image outlines with the outline templates. If the alignment deviation of any region on any face is greater than or equal to a preset value, then that region is marked as a candidate region for damage.

3. The anomaly tracing system according to claim 2, characterized in that, The Tray disc damage tracing system extracts the feature point coordinates of the candidate damage area, including: The Tray disk damage tracing system performs morphological dilation on the damaged candidate region and extracts the coordinates of feature points in the damaged candidate region according to preset rules. The preset rules include: if the rules are broken, extract the coordinates of the diagonal vertices; if the rules are broken irregularly, extract the feature point set.

4. The anomaly tracing system according to claim 1, characterized in that, The Tray disc damage tracing system compares the coordinates of the minimum circumscribed polygon with the rectangular regions corresponding to the contact positions of the mechanisms stored in the mechanism contact surface database. If there is a spatial intersection between the minimum circumscribed polygon and the rectangular regions of the mechanism contact positions, then the mechanism is determined to be a highly correlated suspected mechanism causing the Tray disc damage, including: The Tray damage tracing system determines whether the rectangular area where the smallest circumscribed polygon contacts the mechanism has overlapping regions in both the X and Y axes. If there are overlapping regions in both directions, it is determined that there is a spatial intersection between the minimum circumscribed polygon and the rectangular region at the contact position of the mechanism. This organization has been identified as a highly suspected entity linked to the damage to the Tray disc.

5. The anomaly tracing system according to claim 4, characterized in that, Also includes: When the Tray disk damage tracing system determines that the rectangular regions of the contact positions of the minimum circumscribed polygon and multiple mechanisms overlap in both the X and Y axes, the multiple mechanisms are sorted according to a preset association priority rule. The preset association priority rules include: the higher the percentage of intersection area, the higher the priority; And / or, the higher the institution's historical contribution to damage, the higher its priority.

6. The anomaly tracing system according to claim 1, characterized in that, The tray damage tracing system is also used to: automatically generate standardized tracing reports; the standardized tracing reports include at least basic information, damage details, and tracing conclusions; wherein: The basic information includes at least: Tray disk PN code, supplier, product model, discovery site, and discovery time; The damage details include at least: the damaged contact surface, the coordinates of the damaged area, and the damage type; The source tracing conclusions include at least: a ranking of highly related suspected organizations.

7. The anomaly tracing system according to claim 1, characterized in that, The cabinet storage detection system includes: a cabinet unlocking unit, a six-sided image acquisition unit, and a classification and storage unit; The sorting and storage unit is equipped with a testing area, a tested area, and a temporary storage area; The cabinet unlocking unit is used to compare the externally input Tray information with the Tray information stored in the cabinet storage detection system. Only when the comparison information matches, the cabinet unlocking unit controls the unlocking, allowing the damaged Tray to be placed into the detection area. The six-sided image acquisition unit is used to acquire images of the six sides of the damaged tray.

8. The anomaly tracing system according to claim 7, characterized in that, The six-sided image acquisition unit has six sets of industrial cameras and strip LED light sources symmetrically deployed on the inner wall of the detection cavity, enabling detection of the entire surface of the damaged tray without blind spots.

9. The anomaly tracing system according to claim 1, characterized in that, The contact surface database of the mechanism adopts a six-level structure for storage: product model-Tray PN-site-contact mechanism-contact surface-contact point coordinates.

10. A method for tracing the source of anomalies in chip carrier tray damage, characterized in that, The anomaly tracing system for chip carrier tray damage as described in any one of claims 1-9, wherein the anomaly tracing method comprises: Acquire images of all six sides of the damaged tray; The six-sided images of the damaged tray are compared with the six-sided images of the corresponding normal trays stored in the normal tray database. If the deviation of any region on any face is greater than or equal to a preset value, the region is marked as a candidate region for damage. The coordinates of the feature points of the candidate region for damage are extracted, and a polygon is constructed based on the extracted feature point coordinates. Determine the smallest bounding polygon that covers all vertices of the polygon, and compare its coordinates with the rectangular region corresponding to the mechanism contact position stored in the mechanism contact surface database; If there is a spatial intersection between the minimum circumscribed polygon and the rectangular area at the contact position of the mechanism, then the mechanism is determined to be a highly correlated suspected mechanism causing the damage to the Tray disk.

Images