Chip detection system and method based on image analysis
By using image analysis technology during chip processing, the operation processing index value and process contribution are evaluated, the problem of difficulty in traceability of chip quality problems is solved, and the chip yield and traceability accuracy are improved.
Patent Information
- Application Number
- CN202411639882.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-18
- Publication Date
- 2025-02-28
AI Technical Summary
The prior art is difficult to effectively trace the specific processes that lead to quality problems during chip processing, resulting in low chip yield and inaccurate traceability of quality defects.
Using a chip detection system based on image analysis, the image capturing components are set up to capture images of each processing process, identify the characteristic information before and after processing, evaluate the operation processing index value, judge the process continuity, and calculate the contribution degree of each process in the defect area to assist in the early warning of equipment performance detection.
It realizes accurate traceability of chip quality problems, improves chip yield, and enhances the traceability accuracy of quality defects.
Smart Images
Figure CN119540191A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of chip processing and detection, and in particular to a chip detection system and method based on image analysis. Background Art
[0002] Chip processing refers to further processing of manufactured chips to achieve more advanced functions and performance. The basic purpose of chip processing is to process the chip surface and add circuits, leads, etc. to the chip to further expand the function and scope of use of the chip. Through chip processing, various chips with different structures and complex functions related to the application field can be formed.
[0003] Chip processing usually includes: photolithography, evaporation, and constrained etching; among them, photolithography is a process that uses light and shadow to pattern the chip surface through equipment. Photolithography technology is one of the most important technologies in the field of chip processing, and can perform fine processing on the chip surface, such as fine-tuning the shape and size of the chip surface structure; evaporation is a processing technology that covers a layer of material on the chip surface, commonly used are metals or oxides, etc., through evaporation deposition, a layer of firmly bonded material is formed with the chip; constrained etching is a processing technology that etches the chip surface, but unlike traditional etching, constrained etching first performs certain processing on the chip surface, uses photolithography technology to show the desired chip morphology, and perfectly restores the chip surface pattern through constrained irradiation.
[0004] Chip quality testing is a key link in ensuring the performance and reliability of electronic products. Chip quality testing is a systematic process that covers multiple aspects, including appearance inspection, static and dynamic characteristics testing, functional verification, environmental adaptability testing, and reliability assessment. Summary of the invention
[0005] The object of the present invention is to provide a chip detection system and method based on image analysis to solve the problems raised in the prior art.
[0006] To achieve the above object, the present invention provides the following technical solution: a chip detection method based on image analysis, the method comprising: Step S1: sequentially setting a camera assembly on a corresponding processing station for each processing step in the target chip processing process; pre-acquiring all the to-be-processed areas of the target chip; sorting out the operation correspondence between each processing step and the corresponding to-be-processed area, and evaluating the operation processing index value of each processing step for the corresponding target operation area; Step S2: for the processing procedures of the target operation in the same area to be processed, determining whether there is a feature of process continuity based on the same area to be processed; Step S3: Collect historical defect records of target chips that fail quality inspection after completing corresponding processing steps; evaluate and calculate the contribution value of each processing step in the corresponding problem area of defect processing; Step S4: sort out the contribution value distribution of each processor in the corresponding historical defect record, and assist the management terminal in issuing an early warning prompt on whether to carry out performance testing of the equipment performing processing at the corresponding processing station.
[0007] Preferably, step S1 comprises: Step S1-1: capturing images of the target chip before and after each processing step is completed, respectively, to obtain a first characteristic image and a second characteristic image corresponding to each processing step; The image captured by the target chip before processing is the first characteristic image, and the image captured by the target chip after processing is the second characteristic image; Step S1-2: Based on the first characteristic image and the second characteristic image captured for each processing step, the characteristic information presented by each processing area before and after the processing step is performed is identified; for each processing step, the processing area whose corresponding characteristic information is not completely the same is captured, and it is determined that the processing area is the target operation area of each processing step, and each processing step is the characteristic process of the processing area; Step S1-3: Obtain the operation processing index value β=1-α of each processing procedure for each corresponding target operation area, wherein α is the similarity between the feature information presented by each corresponding target operation area before and after the execution of each processing procedure, α∈[0,1]; Among them, α represents the similarity value between corresponding feature information, with a value range of [0,1], and β represents the deviation value between corresponding feature information; The larger the value of the operation processing index β is, the greater the degree of change on the surface of the target chip is after the operation processing of the corresponding processing step.
[0008] Preferably, step S2 comprises: Step S2-1: sort the characteristic processes corresponding to each area to be processed according to the execution order in the entire process of target chip processing, and obtain the characteristic process sequence corresponding to each area to be processed; extract each characteristic process of each area to be processed, and calculate the operation processing index value of each area to be processed; Step S2-2: If processing steps a and b are captured in adjacent order from a feature process sequence of a certain area to be processed, wherein the order of processing step a in the corresponding feature process sequence is before processing step b, and β(a)<β(b) is satisfied, wherein β(a) is the operation processing index value of processing step a for a certain area to be processed, and β(b) is the operation processing index value of processing step b for a certain area to be processed; it is determined that there is process continuity between processing step a and processing step b based on a certain area to be processed; There are two conditions for judging whether there is process continuity between processing steps a and processing steps b based on the same target operation area: first, whether they are arranged adjacently in the corresponding feature process sequence. If so, it means that after one processing step completes the operation processing on the same target operation area in the target chip, the next operation processing on the same target operation area is performed by another processing step; second, whether the corresponding operation processing index value shows a numerical growth trend with the advancement of the corresponding processing station sequence. If so, it means that after the operation processing of processing step b, the degree of change caused to the target chip in the same target operation area is greater than the degree of change caused to the target chip in the same target operation area after the operation processing of processing step a, that is, the specific operation position range involved in the completion of the processing content of processing step b in the same target operation area is larger than the specific operation position range involved in the completion of the processing content of processing step a in the same target operation area, that is, the possibility of overlap between the specific operation position ranges involved in processing steps a and processing steps b is higher.
[0009] Preferably, step S3 includes: Step S3-1: Identify all processing steps that have been completed for the corresponding target chip before each historical record is generated, and obtain a set Y of problematic processing steps corresponding to each historical record; wherein, after each processing step is completed for each pair of target chips, target chip quality inspection is required to be performed on the target chips in sequence to determine whether the target chip obtained up to the target processing step has defects; Step S3-2: extracting the area to be processed where each defect is located in each historical record, and setting the area to be processed as a problem area in the corresponding historical record; extracting the characteristic process sequence X corresponding to each problem area in each historical record; Step S3-3: extracting a defective processing procedure set Q=Y∩X for each corresponding problem area in each historical record; wherein the order of the processing procedures in the defective processing procedure set Q follows the execution sequence in the entire process of target chip processing; Step S3-3: Set a unit defect score value θ, and calculate the unit contribution value η of each processing procedure in the defective processing procedure set Q of each problem area when processing the defective problem area, where η=θ / card(Q), where card(Q) represents the total number of processing procedures included in the defective processing procedure set Q, and evaluate to obtain the contribution degree value δ=η×P of each processing procedure in each problem area of defective processing; where P represents the ranking value of each processing procedure in the defective processing procedure set Q; where, if there are two processing procedures in the defective processing procedure set Q that satisfy the process continuity based on the existence of each problem area, for the two processing procedures in turn, the ranking value when calculating the corresponding contribution degree value is adjusted to be the average ranking value of the two processing procedures.
[0010] Preferably, step S4 comprises: Step S4-1: The initial defect score value of each processing step is preset to 0, and the contribution value of each processing step in the corresponding historical defect record when the defect is processed in the corresponding problem area is accumulated to obtain the total contribution value of each processing step; Step S4-2: When it is detected that the total contribution value of a certain processing procedure is greater than the degree threshold, the feedback prompts the management terminal to perform performance testing on the equipment performing processing at the corresponding processing station of the certain processing procedure.
[0011] In order to better implement the above method, a chip detection system is also proposed, which includes: an operation processing index value calculation module, a process continuity judgment management module, a contribution degree value calculation module, and a detection prompt management module; The operation processing index value calculation module is used to sequentially set a camera component on the corresponding processing station for each processing step in the target chip processing process; pre-acquire all the to-be-processed areas of the target chip; sort out the operation correspondence between each processing step and the corresponding to-be-processed area, and evaluate the operation processing index value of each processing step for the corresponding target operation area; The process continuity judgment management module is used to judge whether the process continuity exists in the same processing area for the target operation based on the characteristics of the same processing area; The contribution value calculation module is used to collect historical defect records of target chips that failed quality inspection after completing corresponding processing procedures; evaluate and calculate the contribution value presented by each processing procedure when processing the corresponding problem area of the defect; The detection prompt management module is used to sort out the distribution of contribution values of each processing worker in the corresponding historical defect records, and assist the management terminal in issuing early warning prompts on whether to carry out performance testing of the equipment performing processing at the corresponding processing station.
[0012] Preferably, the operation processing index value calculation module includes an operation processing information combing unit and a calculation evaluation unit; The operation processing information sorting unit is used to sort out the operation correspondence between each processing step and the corresponding area to be processed; The calculation and evaluation unit is used to evaluate the operation processing index value of each processing procedure for the corresponding target operation area.
[0013] Preferably, the contribution value calculation module includes a historical chip defect information collection unit and a contribution value calculation unit; A historical chip defect information collection unit, used to collect historical defect records of target chips that failed quality inspection after completing corresponding processing procedures; The contribution value calculation unit is used to evaluate and calculate the contribution value presented by each processing step when the defective processing corresponding problem area is processed.
[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention can trace the problematic process of chips that have quality problems after processing, and specifically trace and locate the processing process that may cause the defect problem. At the same time, while tracing the problematic process, the present application takes into account that in the actual processing operation, there are multiple processing processes that are superimposed on the same operation area in processing content, thereby avoiding the traditional practice of arbitrarily judging that there is a processing defect in the processing process that has a defect problem at the moment. The present application can effectively improve the yield of processed chips, and at the same time can effectively improve the accuracy of tracing the quality defects that occur in the chip processing process. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] Figure 1 A schematic diagram of a chip detection method based on image analysis according to the present invention; Figure 2 The figure is a schematic diagram of the structure of a chip detection system based on image analysis of the present invention. DETAILED DESCRIPTION
[0016] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present invention.
[0017] Example: Figure 1-Figure 2 As shown, the present invention provides a technical solution, a chip detection method based on image analysis, the method comprising: Step S1: sequentially setting a camera assembly on a corresponding processing station for each processing step in the target chip processing process; pre-acquiring all the to-be-processed areas of the target chip; sorting out the operation correspondence between each processing step and the corresponding to-be-processed area, and evaluating the operation processing index value of each processing step for the corresponding target operation area; Step S2: for the processing procedures of the target operation in the same area to be processed, determining whether there is a feature of process continuity based on the same area to be processed; Step S3: Collect historical defect records of target chips that fail quality inspection after completing corresponding processing steps; evaluate and calculate the contribution value of each processing step in the corresponding problem area of defect processing; Step S4: sort out the contribution value distribution of each processor in the corresponding historical defect record, and assist the management terminal in issuing an early warning prompt on whether to carry out performance testing of the equipment performing processing at the corresponding processing station.
[0018] Wherein, step S1 comprises: Step S1-1: capturing images of the target chip before and after each processing step, respectively, to obtain a first characteristic image and a second characteristic image corresponding to each processing step; wherein the image captured for the target chip before processing is the first characteristic image, and the image captured for the target chip after processing is the second characteristic image; Step S1-2: Based on the first characteristic image and the second characteristic image captured for each processing step, the characteristic information presented by each processing area before and after the processing step is performed is identified; for each processing step, the processing area whose corresponding characteristic information is not completely the same is captured, and it is determined that the processing area is the target operation area of each processing step, and each processing step is the characteristic process of the processing area; Step S1-3: Obtain the operation processing index value β=1-α of each processing procedure for each corresponding target operation area, wherein α is the similarity between the feature information presented by each corresponding target operation area before and after the execution of each processing procedure, α∈[0,1]; Wherein, step S2 comprises: Step S2-1: sort the characteristic processes corresponding to each area to be processed according to the execution order in the entire process of target chip processing, and obtain the characteristic process sequence corresponding to each area to be processed; extract each characteristic process of each area to be processed, and calculate the operation processing index value of each area to be processed; Step S2-2: If processing steps a and b are captured in adjacent order from a feature process sequence of a certain area to be processed, wherein the order of processing step a in the corresponding feature process sequence is before processing step b, and β(a)<β(b) is satisfied, wherein β(a) is the operation processing index value of processing step a for a certain area to be processed, and β(b) is the operation processing index value of processing step b for a certain area to be processed; it is determined that there is process continuity between processing step a and processing step b based on a certain area to be processed; For example, the entire process of processing the target chip is: processing step a→processing step b→processing step c→processing step d→processing step e; The characteristic process sequence extracted from a certain area to be processed is {processing process a, processing process d, processing process e}; It is known that the operation processing index value of processing procedure a to a certain area to be processed is β=0.6, the operation processing index value of processing procedure d to a certain area to be processed is β=0.4, and the operation processing index value of processing procedure e to a certain area to be processed is β=0.5; In summary, it can be seen that the processing procedure a is executed before the processing procedure d in the whole process of target chip processing, and the processing procedure d is executed before the processing procedure e in the whole process of target chip processing; Therefore, it can be seen from the above that the operation processing index value corresponding to processing procedure a is greater than the operation processing index value corresponding to processing procedure d, and the operation processing index value corresponding to processing procedure e is greater than the operation processing index value corresponding to processing procedure d; therefore, it is determined that there is process continuity between processing procedure d and processing procedure e based on a certain area to be processed.
[0019] Wherein, step S3 comprises: Step S3-1: Identify all processing steps that have been completed for the corresponding target chip before each historical record is generated, and obtain a set Y of problematic processing steps corresponding to each historical record; Step S3-2: extracting the area to be processed where each defect is located in each historical record, and setting the area to be processed as a problem area in the corresponding historical record; extracting the characteristic process sequence X corresponding to each problem area in each historical record; Step S3-3: extracting a defective processing procedure set Q=Y∩X for each corresponding problem area in each historical record; wherein the order of the processing procedures in the defective processing procedure set Q follows the execution sequence in the entire process of target chip processing; Step S3-3: Set a unit defect score value θ, and calculate the unit contribution value η of each process in the defective process set Q of each problem area when processing each problem area, where η=θ / card(Q), where card(Q) represents the total number of processes included in the defective process set Q, and evaluate the contribution value δ=η×P of each process in processing each problem area; where P represents the ranking value of each process in the defective process set Q; Among them, if there are two processing steps in the defective processing step set Q that satisfy the process continuity based on each problem area, the ranking value of the two processing steps is adjusted to the average ranking value of the two processing steps when calculating the corresponding contribution value; Wherein, step S4 comprises: Step S4-1: The initial defect score value of each processing step is preset to 0, and the contribution value of each processing step in the corresponding historical defect record when the defect is processed in the corresponding problem area is accumulated to obtain the total contribution value of each processing step; Step S4-2: When it is detected that the total contribution value of a certain processing procedure is greater than the degree threshold, the feedback prompts the management terminal to perform performance testing on the equipment performing processing at the corresponding processing station of the certain processing procedure.
[0020] In order to better implement the above method, a chip detection system is also proposed, which includes: an operation processing index value calculation module, a process continuity judgment management module, a contribution degree value calculation module, and a detection prompt management module; The operation processing index value calculation module is used to sequentially set a camera component on the corresponding processing station for each processing step in the target chip processing process; pre-acquire all the to-be-processed areas of the target chip; sort out the operation correspondence between each processing step and the corresponding to-be-processed area, and evaluate the operation processing index value of each processing step for the corresponding target operation area; The process continuity judgment management module is used to judge whether the process continuity exists in the same processing area for the target operation based on the characteristics of the same processing area; The contribution value calculation module is used to collect historical defect records of target chips that failed quality inspection after completing corresponding processing procedures; evaluate and calculate the contribution value presented by each processing procedure when processing the corresponding problem area of the defect; The detection prompt management module is used to sort out the distribution of contribution values of each processing worker in the corresponding historical defect records, and assist the management terminal in issuing early warning prompts on whether to carry out performance testing of the equipment performing processing at the corresponding processing station.
[0021] Among them, the operation processing index value calculation module includes an operation processing information sorting unit and a calculation evaluation unit; The operation processing information sorting unit is used to sort out the operation correspondence between each processing step and the corresponding area to be processed; The calculation and evaluation unit is used to evaluate the operation processing index value of each processing procedure for the corresponding target operation area.
[0022] Among them, the contribution value calculation module includes a historical chip defect information collection unit and a contribution value calculation unit; A historical chip defect information collection unit, used to collect historical defect records of target chips that failed quality inspection after completing corresponding processing procedures; The contribution value calculation unit is used to evaluate and calculate the contribution value presented by each processing step when the defective processing corresponding problem area is processed.
[0023] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.
Claims
1. A chip detection method based on image analysis, characterized in that: The method comprises: Step S1: sequentially setting a camera assembly on a corresponding processing station for each processing step in the target chip processing process; pre-acquiring all the to-be-processed areas of the target chip; sorting out the operation correspondence between each processing step and the corresponding to-be-processed area, and evaluating the operation processing index value of each processing step for the corresponding target operation area; Step S2: for the processing procedures of the target operation in the same area to be processed, determining whether there is a feature of process continuity based on the same area to be processed; Step S3: Collect historical defect records of target chips that fail quality inspection after completing corresponding processing steps; evaluate and calculate the contribution value of each processing step in the corresponding problem area of defect processing; Step S4: sort out the contribution value distribution of each processor in the corresponding historical defect record, and assist the management terminal in issuing an early warning prompt on whether to carry out performance testing of the equipment performing processing at the corresponding processing station.
2. The chip detection method based on image analysis according to claim 1, characterized in that: The step S1 comprises: Step S1-1: capturing images of the target chip before and after each processing step, respectively, to obtain a first characteristic image and a second characteristic image corresponding to each processing step; wherein the image captured for the target chip before processing is the first characteristic image, and the image captured for the target chip after processing is the second characteristic image; Step S1-2: Based on the first characteristic image and the second characteristic image captured for each processing procedure, the characteristic information presented by each area to be processed before and after the execution of each processing procedure is identified; for each processing procedure, the area to be processed whose corresponding characteristic information is not completely the same is captured, and the area to be processed is determined to be the target operation area of each processing procedure, and each processing procedure is a characteristic procedure of the area to be processed; Step S1-3: Obtain the operation processing index value β=1-α of each processing procedure for each corresponding target operation area, wherein α is the similarity between the feature information presented by the corresponding target operation area before and after the execution of each processing procedure, α∈[0,1].
3. The chip detection method based on image analysis according to claim 2, characterized in that: The step S2 comprises: Step S2-1: sorting the characteristic processes corresponding to each area to be processed according to the execution order in the entire process of target chip processing, and obtaining the characteristic process sequence corresponding to each area to be processed; extracting each characteristic process of each area to be processed, and calculating the operation processing index value of each area to be processed; Step S2-2: If adjacently arranged processing steps a and processing steps b are captured from the characteristic process sequence of a certain area to be processed, wherein processing step a is ranked before processing step b in the corresponding characteristic process sequence, and β(a)<β(b) is satisfied, wherein β(a) is the operating processing index value of processing step a for the certain area to be processed, and β(b) is the operating processing index value of processing step b for the certain area to be processed; it is determined whether there is process continuity between processing step a and processing step b based on the certain area to be processed.
4. The chip detection method based on image analysis according to claim 3, characterized in that: The step S3 comprises: Step S3-1: Identify all processing steps that have been completed on the corresponding target chip before each historical record is generated, and obtain a set Y of problematic processing steps corresponding to each historical record; Step S3-2: extracting the area to be processed where each defect is located in each historical record, and setting the area to be processed as a problem area in the corresponding historical record; extracting the characteristic process sequence X corresponding to each problem area in each historical record; Step S3-3: extracting a defective processing procedure set Q=Y∩X for each corresponding problem area in each historical record; wherein the order of the processing procedures in the defective processing procedure set Q follows the execution sequence in the entire process of target chip processing; Step S3-3: Set a unit defect score value θ, and calculate the unit contribution value η of each processing procedure in the defective processing procedure set Q of each problem area when defectively processing the problem area, where η=θ / card(Q), where card(Q) represents the total number of processing procedures included in the defective processing procedure set Q, and evaluate to obtain the contribution degree value δ=η×P of each processing procedure when defectively processing the problem area; where P represents the ranking value of each processing procedure in the defective processing procedure set Q; where, if there are two processing procedures in the defective processing procedure set Q that satisfy the process continuity based on the existence of each problem area, for the two processing procedures in turn, the ranking value used in calculating the corresponding contribution degree value is adjusted to be the average ranking value of the two processing procedures.
5. The chip detection method based on image analysis according to claim 4, characterized in that: The step S4 comprises: Step S4-1: The initial defect score value of each processing step is preset to 0, and the contribution value of each processing step in the corresponding historical defect record when the defect is processed in the corresponding problem area is accumulated to obtain the total contribution value of each processing step; Step S4-2: When it is detected that the total contribution value of a certain processing procedure is greater than the degree threshold, the feedback prompt management terminal performs performance detection on the equipment performing processing at the corresponding processing station of the certain processing procedure.
6. A chip detection system, used to execute the target chip detection method based on image analysis according to any one of claims 1 to 5, characterized in that: The system includes: a processing index value calculation module, a process continuity judgment management module, a contribution value calculation module, and a detection prompt management module; The operation processing index value calculation module is used to sequentially set a camera component on the corresponding processing station for each processing step in the target chip processing process; pre-acquire all the to-be-processed areas of the target chip; sort out the operation correspondence between each processing step and the corresponding to-be-processed area, and evaluate the operation processing index value of each processing step for the corresponding target operation area; The process continuity judgment management module is used to judge whether the process continuity exists in the same processing area based on the characteristics of the target operation in the same processing area; The contribution value calculation module is used to collect historical defect records of target chips that fail to pass quality inspection after completing corresponding processing procedures; evaluate and calculate the contribution value presented by each processing procedure when processing the corresponding problem area of the defect; The detection prompt management module is used to sort out the contribution value distribution of each processing worker in the corresponding historical defect record, and assist the management terminal in providing early warning prompts on whether to carry out performance testing of the equipment performing processing at the corresponding processing station.
7. A chip detection system according to claim 6, characterized in that: The operation processing index value calculation module includes an operation processing information sorting unit and a calculation evaluation unit; The operation processing information sorting unit is used to sort out the operation correspondence between each processing step and the corresponding area to be processed; The calculation and evaluation unit is used to evaluate the operation processing index value of each processing step for the corresponding target operation area.
8. A chip detection system according to claim 6, characterized in that: The contribution value calculation module includes a historical chip defect information collection unit and a contribution value calculation unit; The historical chip defect information collection unit is used to collect historical defect records of target chips that fail to pass quality inspection after completing corresponding processing procedures; The contribution value calculation unit is used to evaluate and calculate the contribution value presented by each processing step when the defective processing corresponds to the problem area.